This is an archive of past discussions about Help:Notifications. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.
Latest comment: 11 years ago8 comments8 people in discussion
I don't like the default zero messages "notification". I would prefer that the notification button didn't show when there is no messages. In addition, when I do get messages I can't read them properly, but this may be because I use the Cologne Blue skin and not the default. Regards, Iselilja (talk) 05:37, 6 May 2013 (UTC)
Hi Iselilja, due to the limited design and development bandwidth we are able to optimize visual design for Vector and perhaps Monobook, a lot of the visual design work for new components will be optimized for these skins and I would encourage use of these for best user experience; Its also not easy to create a clean crisp design for so many skins with the front end limitations of Mediawiki Vibhabamba (talk) 18:20, 6 May 2013 (UTC)
The only reason there is a zero button is in case you want to re-visit any of you recent notifications that you've already seen. Otherwise there would be no way to get to them other than manually going to Special:Notifications. Kaldari (talk) 03:59, 7 May 2013 (UTC)
"Best user experience"? I thought that the reason different skins were chosen was to improve the experience, for that user. Apteva (talk) 13:31, 12 May 2013 (UTC)
Apparently not. They've already got rid of several of the much loved (by some) old skins, and the plans are to get rid of anything that isn't Vector in due course. You have been warned. An optimist on the run! 12:06, 13 May 2013 (UTC)
Given that most power editors use Monobook, I very much doubt it's going anywhere for a very long time. I would consider Vector and Monobook fully supported, as Vibha says. --MZMcBride (talk) 05:28, 19 May 2013 (UTC)
Orange Bar of Death (OBOM) / Red Badge of Doom (RBOM)
Next steps
Resolving message indicator issue
Upcoming features (e.g. HTML Email)
We invite you to try out the prototype gadgets we created for key options. Just go to Special:Preferences#mw-prefsection-gadgets and scroll down to the "Testing and development" section. Be sure to uninstall any of the prototype user scripts first (if you had any), then clear your caches after you install a gadget. You will also need to send yourself a message from a separate account. As long as you don't click on the message indicator, you can switch between different gadgets and still see the new indicator.
Here are the options that we have has posted so far:
Talk page alert prototype D1 - blue tooltip, fades out
Talk page alert prototype D2 - blue tooltip, dismissible
Talk page alert prototype E1 - orange bar, floating
Talk page alert prototype E2 - orange bar, docked to top
Talk page alert prototype E3 - orange bar, docked to bottom
Let's all evaluate their respective merits against these goals:
Prominence (users should be able to notice they have new messages)
Persistence (users should be reminded if they do not check messages)
Clarity (disimbiguate from other notifications)
Consistency (with our UI design goals and best practices)
We will aim to summarize our discussions later today, and propose some steps towards resolving this issue together, based on your feedback.
We will post another update soon. Thanks for your constructive comments for helping improve notifications. We know there are still issues, and are working around the clock to address them as quickly as possible. To be continued ... Fabrice Florin (WMF) (talk) —Preceding undated comment added 19:57, 8 May 2013 (UTC)
IRC chat update
I'm happy to report that we had a very productive conversation in today's office hours chat on IRC about Notifications and the new message indicator. Many thanks to all community members who joined this discussion -- especially Edokter and Ignatz, who made some invaluable contributions to move this project forward (see IRC chat log).
Together, we reviewed proposed designs for message indicators, and discussed how each option serves our goals (prominence, persistency, clarity, consistency and compatibility). We then brainstormed a new version combining the best of options E and F -- using the orange background from E and the navbar integration from F, as shown below (see this full screenshot).
In collaboration with community members, our developer Ryan Kaldari and designer Vibha Bamba made live changes to our prototype gadget for Option F2. After testing that revised gadget, we reached consensus that this would be a reasonable solution to integrate in the Echo extension -- which we aim to do next Tuesday, May 14. We think this solution meets all of our goals:
it is prominent: the orange background makes it stand out on its own, and the animation further draws your eye to it;
it is persistent: it shows up each time you load a page, and stays on until you click on it (or go to your talk page);
it is clear: it differentiates messages from other notifications, and is combined with the 'Talk' link so you know what it's about;
it is consistent: it integrates with the main nav bar and 'talk' link, and conforms with user interface guidelines and best practices.
We encourage you to try out this new solution for yourself: go to Preferences > Gadgets and scroll down to the "Testing and development" section, then select 'Talk page alert prototype F2'. Be sure to clear your cache after you save your preferences (and uninstall any of the earlier user scripts from your '<username>/vector.js' file). You may also need to post a message on your talk page from a separate account to test it.
Please let us know what you think. We hope that this temporary solution will work for most of you -- so we can get it deployed and resume work on the next features for Notifications. Rest assured that we will continue to improve the user interface based on community feedback, but we would like to move on to other important tasks, such as supporting HTML emails. Thanks again to Edokter, Ignatz and other community members for all their creative suggestions -- and for helping reach a compromise that meets our shared objectives. Onward! Fabrice Florin (WMF) (talk) 00:34, 9 May 2013 (UTC)
What happened to your goal of compatibility? It's a bit rich to say this meets all of your goals when you've just changed your goals... Talk page notifications need to work on all browsers regardless of their configuration, to avoid giving people excuses for ignoring talk page warnings. As I understand it, what you're planning will only work if the browser supports JavaScript. As a compromise, could you at least restore the old orange bar for non-JavaScript users and users of screen readers, so everyone will get a talk page notification of some sort? – PartTimeGnome(talk | contribs)01:08, 9 May 2013 (UTC)
The current prototype uses javascript, but once it is integrated into Echo, there is a good chance the notification will be HTML-only (using javascript only for the animation). The chosen layout allows for this, as it sits inside the document flow and not in some popup. — Edokter (talk) — 01:20, 9 May 2013 (UTC)
I think E1>D1>D2 overall, especially given the concern for (new) editor awareness. E1 really provides that obvious/obnoxious feedback. Any of the Es would be okay, but E1>E3>E2. F2 is, I agree, an elegant solution. The animation makes it abundantly obvious but it's not so annoying as to irritate too many people. I'd sooner see a fade-away on it similar to D1 but it's a very solid framework. ~ Amory(u • t • c)01:39, 9 May 2013 (UTC)
I too cannot work with animation on any web page. If I cannot immediately turn it off, I must immediately scroll the page to hide it , which is a nuisance, or I cannot read the text or even stay comfortably on the page. This is a question of accessibility. Animation, blinking, and similar visual devices should never be used in the interface to a site. The color is OK-- but restricting it to the navbar makes it too small. It will be noticed better if it were separate, just like the old bar. Some people seem to like animation, I notice, but this is so utterly wrong for a minority that it really shouldn't be considered as the basic way. Whatever we decide, it would be good to have a gadget to personalize it--there is no worry about the inexperienced users, for they never think of adjusting their preferences. (I have the turn off animation button set in my gadgets, but it does not seem to turn off all animations that people use. ) and the comment about js is very much to the point--we don't want a "good chance" we want to avoid js to the degree possible at least for basic features. DGG ( talk ) 06:07, 9 May 2013 (UTC)
If you have the 'no animation' gadget enabled, this notification should not animate. (If it does, I need to fix the gadget.) — Edokter (talk) — 09:07, 9 May 2013 (UTC)
What does "animation" mean? If it's a blinking orange bar, that's extremely problematic for a subset of users. I won't bother to explain why here. Truthkeeper (talk) 10:38, 9 May 2013 (UTC)
"Animation" means that the navbar starts out with just "Username, notificationsbox, sandbox..." and then the orange "talk" thingie scrolls out right-to-left between the notificationsbox and the sandbox links. No blinking or flashing or anything. You can try the gadget if you want (let me know if you need a test on your talk page). Ignatzmice•talk11:45, 9 May 2013 (UTC)
Thanks, I'll probably ping you later - off to work right now. Blinking would be terrible, movement is an less of an issue but still suboptimal (I always click out of user pages with animation) so would like to see how quickly it moves. Truthkeeper (talk) 11:53, 9 May 2013 (UTC)
(edit conflict) I think this is a really nice solution, thanks to everybody contributing to it! Regarding the animation: that's pointless for me, since the interface is semi-animated anyway, on the same time scale, due to the loading of the clock gadget and Twinkle. At the moment the 'no animation' gadget doesn't seem to turn the animation off, but I guess that's fixable (and it's not a big deal anyway).
If people think there's still too many new editors who miss their messages, we can think more about a second line of defense similar to the Prompt me when entering a blank edit summary feature. After all, it's not really important that people notice their messages as fast as possible, as long as they notice them before (successfully) submitting another edit. — HHHIPPO10:42, 9 May 2013 (UTC)
Um ... animation is very much a big deal for some and if the developers aren't aware of this, they should be. Truthkeeper (talk) 10:45, 9 May 2013 (UTC)
Sure, it should be fixed, what I meant by 'no big deal' is 'no reason for a big drama calling for an immediate rollback'. To answer your previous question: it's not blinking, it's just appearing in an animated way, similar to what Libre Office calls 'peek in from left'. — HHHIPPO11:14, 9 May 2013 (UTC)
Hi folks, thanks for your constructive feedback about this proposed feature! I'm glad to hear that this new version of F2 seems to be an acceptable solution for you overall, with a few more tweaks. On the issue of animation, I share some of your concerns, largely for compatibility reasons: on slow machines and on tablets like the iPad, the animation is too slow to be practical. So I am going to propose to the team that we try a version of the gadget without animation, so we can all look at it together and determine our next steps. If that new 'static' version seems to work well, we could provide it for slow processors and tablets -- or possibly for all platforms. But let's take it one step at a time. Sound good? Thanks again to everyone who collaborated productively to finding this solution: we couldn't have made this much progress without the active participation of community members like you! Fabrice Florin (WMF) (talk) 16:25, 9 May 2013 (UTC)
Well, it's better than F, which hides in the scenery of the page like a stick insect in a bush. This is an orange stick insect in a bush. That's my problem with it - it's still in the bush. For newbies, we need something that obviously is not bush. (Not a political comment - note small 'b'...) The colour intensity will be similar to the blue of Modern, but newbies will mostly be on the wishy washy Vector skin, as, by the time they discover skins, they'll either have discovered communication or be blocked anyway. A newbie oriented indicator needs to be where it's not expected. Away from a row of little buttons or words. Peridon (talk) 17:13, 9 May 2013 (UTC)
I hope that whatever choice is made by those making it (seemingly not us...), it will capable of being turned off. I don't want animated orange boxes and definitely not flashing ones. The red pimple is fine when you get used to it. It's not the established editor that needs blatant warning. As to the idea I saw somewhere in one of these threads of having red and green pimples for different types of notification - remember R-G colour blindness. Peridon (talk) 17:20, 9 May 2013 (UTC)
Not focusing on the animation issues, this fits most of my suggestions, so I am pretty happy. My only real issues are that it perhaps not large enough to be prominent for new users and that, as mentioned above, that those that want to can turn it off.--SabreBD (talk)18:02, 9 May 2013 (UTC)
Hi DGG, Hhhippo, Ignatz, PartTimeGnome, Peridon, Sabrebd|, Truthkeeper and other contributors to this discussion: I am happy to report that Ryan Kaldari has modified the F2 gadget to turn off the animation, as requested by many users on our talk page -- and to address the compatibility issues with slow platforms and tablets like iPad. What do you think? From our perspective, it appears that the orange bar continues to make this indicator very prominent, even without the animation. So if it works for most of you, we'd like to go ahead with this implementation for now (we can always tweak it later). Special thanks to Edokter, who was kind enough to replace his temporary RBOD gadget with this new F2 gadget, which we we created in collaboration with him and other IRC participants. And thanks to the rest of you for helping us make such rapid progress on this feature! Fabrice Florin (WMF) (talk) 21:50, 9 May 2013 (UTC)
Orange, good; non-animated, good; still on the small side (and I am still mutterring that people who think it should not be "obtrusive" have forgotten the purpose of a warning sign); but I can live with this one. (My concern all along has not been about noticing warnings myself, but about whether newbies I send messages to will notice them). Thanks. JohnCD (talk) 22:26, 9 May 2013 (UTC)
I pretty much agree with JohnCD. I am happy not to have the animation (which was clunky on my laptop), but it was never really about what I see, but how clearly a warning is communicated to new users.--SabreBD (talk) 22:31, 9 May 2013 (UTC)
Thanks, JohnCD and Sabrebd|. Glad this new version works for you. If anyone doesn't like it and would like to turn it off, simply go to Preferences > Gadgets and scroll down to the "Appearances" section, then uncheck 'Alert me when I receive messages on my talk page. (new)'. Next week, we plan to make it part of the Echo extension and offer a preference for it as well, as described in this feature requirement. I'm happy that by putting our heads together in a constructive way, we were able to find a reasonable solution to this issue that seems to address your concerns. To be continued .... Fabrice Florin (WMF) (talk) 22:47, 9 May 2013 (UTC)
I like the new little orange bar; it seems less ominous than the OBOD. I just have two concerns, one for newbies and one for myself:
It does need to be more prominent not only for IPs but also for new accounts (and autoconfirmed wouldn't be a high enough threshold, since many users go for hundreds of edits before ever being welcomed, warned, or otherwise engaged with).
I would very much like the notification to go away after I've opened my talk page. Having to click on the Echo thingy as well is an extra step—not a huge deal but I suspect over time it would become an annoyance, especially on days when there's a lot of traffic on my talk page. Rivertorch (talk) 22:59, 9 May 2013 (UTC)
Hi Rivertorch, I am glad that you like our new little 'Orange Bar of Love' (OBOL) -- which more and more folks I talk to seem to be growing fond of as well. Regarding your first point, we are now providing welcome and getting started to new users right away, and plan to add more notifications beyond that, so they will quickly learn to use the red badge and message indicator, which they are already accustomed to on all major top sites except Wikipedia; so bear with us for a little while longer, as we unfold our plan for engaging users more effectively from the start. We totally agree with your second point about removing the orange bar of love after you've viewed your talk page -- this is a high priority bug we hope to fix very soon. Now that we have reached a resolution on the message indicator issue, we're getting back to work on all the other issues on our plate, and this one is topmost on our minds. Thanks again, and stay tuned for more ... Fabrice Florin (WMF) (talk)
I too think its an improvement over what you did earlier, though not as good as the original--my own idea of what was needed was to reduce the size a little. The main feature I find lacking is that it does not seem to go directly the difference page that shows all the new messages--sometimes people will send e messages to things a little up on the page, & I have to go into the history to find them. DGG ( talk ) 03:41, 10 May 2013 (UTC)
(@ Fabrice Florin) Glad to hear a fix is in the works. Thank you. I admit to being a little worried when you say "we are now providing welcome and getting started to new users right away". I'm not sure what I may have missed in that regard, but I hope we haven't embarked on a scheme to indiscriminately welcome all new users. One of my pet peeves around here is the use of standard welcome templates for users whose only contributions have been vandalism. I have taken multiple editors (most of them relative newbies themselves) to task for doing that. Rivertorch (talk) 06:21, 10 May 2013 (UTC)
Still not convinced. It's a massive step ahead of the original Echo solution, but I still see primary consensus for bringing back the OBOD, and still feel it's the best solution. As many said, it didn't have to come back in orange, but it's still by far the best option, and seems to simply be being ditched because it's "old". If it ain't broke, don't fix it - and that wasn't broken. Now, this new solution is fine for me (although not my preferred solution), but for IPs and inexperienced editors, whom don't know what they're looking for, it still isn't enough. Lukeno94(tell Luke off here)07:13, 10 May 2013 (UTC)
IPs do get the OBoD. For new editors, we should gather some actual experience on how well the new system works, and then see what to tweak or add. — HHHIPPO08:22, 10 May 2013 (UTC)
I find it very nice, enough to catch your attention, but in nice style as well. I think it's better than the original orange bar (which I voted for in the RFC above). Mohamed CJ(talk)20:02, 10 May 2013 (UTC)
more generally
3 questions (I'm addressing it to Fabrice, because he's taking the lead at the moment, but I really am asking this in common to everyone involved in programming the notifications system or the interface in general):
1. Why did you guys not think of doing this sort of inquiry before you introduced the change? Was it because you thought the merits of your first solution so obvious you knew it would be accepted, or that you had tested it yourself as well and as broadly as the community could have done, or you though the details of presentation-- as distinct from the underlying programming -- didn't really matter? Or perhaps that the time saved in getting something to work without further discussion was so important that the extent we liked the actual interface could be ignored?
2 For the other features you are working on, are you going to consult us first with an adequate pilot before introducing it, or are you going to again assume you can do all that is needed just as well yourself, and see what is the feedback later? Are you going to consult broadly, or do it on IRC? Are you going to let the community decide when it has consensus, or decide it yourself?
3 what is the relative importance to your satisfaction with your work: your own pleasure in it, the approval of the officials at the WMF, the approval of the part of the community likely to complain, or the benefit to the makers and users of the encyclopedia?
These questions are really more important than this particular feature; I think much of the outrage wasn't over the deficiencies of the feature itself, but the lack of prior consultation--of indeed of any consultation, had we not forced it on you (at least, that's how it appeared to some of us). I, and I think others, measure the degree to which you have understood by your willingness to address these points. I, and I think others, feel a little distressed you had not addressed them voluntarily much earlier. I'm not one to go by reluctantly extracted confessions of remorse, but I don't know if that is saving face, or the intent to ignore the community as much as we let you. DGG ( talk ) 04:36, 10 May 2013 (UTC)
Seconding/thirding these questions! Changes after changes in notification system! The OBOD (Orange Bar of Dignity/Dedication/Demonstration) was really very good! --Tito Dutta (contact) 07:26, 11 May 2013 (UTC)
Hi DGG., Peridon, Pete and Tito Dutta. Thanks for asking these questions, which I think can help us establish a better mutual understanding -- and pave the way for more effective software releases in the future. I've answered them briefly below, and I look forward to discussing them in more depth with you by phone tomorrow.
Q1. Why did you guys not think of doing this sort of inquiry before you introduced the change?
A: As we do for all of our software tools, we reported regularly about the development of the Notifications tool and its upcoming release on the English Wikipedia, over a wide range of communications channels (e.g.: here on en-wiki, mediawiki.org, blog, on-wiki newsletters, email updates and IRC). We are grateful for all the guidance we've received from community members throughout the past six months, and appreciate their invaluable contributions to the development of this tool. We have also repeatedly invited users to test the product in advance on MediaWiki.org. Since no significant issues were reported during this extensive testing period, we did not expect the removal of the OBOD to cause the strong reaction we have witnessed in recent weeks. At the same time, we had heard numerous concerns from many community members that the so-called 'Orange Bar of Doom' was not the most effective way to inform people of new messages, and our own observations confirmed that this legacy feature was poorly designed and violated widely accepted web conventions. When we reviewed best practices on other top sites (see research slides), it became clear that new design patterns such as the use of a red badge for notifications have now become the norm and match user expectations more closely than OBOD. So it seemed reasonable to deploy this new implementation first, then address any community concerns after everyone had a chance to try out the new tool. There was no malice on our part, it was largely a practical decision.
Q2. For the other features you are working on, are you going to consult us first with an adequate pilot before introducing it?
A: We always strive to get community feedback about features we develop, as we have in the past -- and our pilot for Notifications has been running on MediaWiki.org since December 2012 (though it's worth noting that few of the English community members who have posted on this page took the time to test it on that site and give us advance feedback). In most cases, it is not practical for the WMF to create special pilots for every feature we develop on every site we support -- nor is realistic to get prior community approval before we deploy these features. Instead, we aim for a more agile development approach, where we deploy new beta features quickly to production sites like the English Wikipedia, where users can test them more easily -- we then adjust these features through rapid design iterations based on user feedback, as we did last week with the new message indicator feature. And while we appreciate the consensus principle, it is primarily intended for editorial and social processes -- not for the MediaWiki software platform which powers our sites. In fact, community members explicitly added this exception, which states that 'the community of MediaWiki software developers, including both paid Wikimedia Foundation staff and other volunteers, … [may] operate however they deem necessary or appropriate, such as adding, removing, or changing software features, or accepting or rejecting images, even if their actions are not endorsed by editors here.' This is not a new observation: MediaWiki development has always been a distinct process, and there's always been some amount of (usually constructive) tension between the technical and editorial community, as appears to be the case now.
Q3. What is the relative importance to your satisfaction with your work: your own pleasure in it, the approval of the officials at the WMF, the approval of the part of the community likely to complain, or the benefit to the makers and users of the encyclopedia?
A: Personally, I consider all of the above factors as part of my work, but the most important by far is the overall benefit to all users of MediaWiki projects. In the past year, we have paid particular attention to the needs of new users, to invite productive participation that might offset the active editor decline that is jeopardizing the growth of our movement. In order to better engage these new users, we are now actively modernizing the MediaWiki software platform and user interface, a process that is likely to last several years. To that end, WMF teams are continuously developing and releasing new features and tools, as well as retiring old features which no longer serve their purpose effectively. We realize that this process can seem disruptive for many current users, and we regret this inconvenience. But in today's world, we all need to become more comfortable with change, particularly in the digital space, where user interfaces are constantly being improved at a rapid pace. Rest assured that we will continue to inform the community and invite feedback about upcoming changes -- and we will keep looking for better ways to involve community representatives at key milestones during software development. But the Wikimedia Foundation still needs to reserve the right to make independent decisions about its software releases, because it has to serve many stakeholders beyond the English Wikipedia, and cannot effectively develop different local solutions for every feature it releases on every single site which it supports. With that in mind, we would be very grateful for your cooperation when we make changes which you didn't get a chance to discuss or do not agree with. Thanks for your patience and understanding. Fabrice Florin (WMF) (talk) 20:15, 14 May 2013 (UTC)
Hi, Fabrice Florin (WMF), thanks for answering these questions! I have started feeling that some editors like User:Double sharp, me are being misunderstood for our negative feedback. But, I want to tell you, though our feedback have been negative, actually we are trying our best to provide constructive opinion on the tool. It is unfortunate that our feedback have been negative. But, believe me, we also care for Wikipedia very much like others here, so, our aim has never been to criticize you/developers without any reason. Next time, you add (some other) new exciting features and you'll definitely find "us" "wholeheartedly" congratulating you and celebrating the feature/tool at first. On Q1, do we have an editors' team of "Trusted Testers", please see Suggestion 3 here. If you roll out a critical new feature like this to some trusted editors before making it public, they can give initial reactions.I have been a trusted tester of Google and Gmail for a long time now. I wish I could tell you my experiences there, but, I can't due to the NDA (Non disclosure agreement). Anyway, TT's feedback might be very helpful. --Tito Dutta (contact) 20:39, 14 May 2013 (UTC)
That's a good suggestion :). I know that Erik (Eloquence) is very keen on the idea of deploying things in a beta on enwiki, in future, rather than testing elsewhere; hopefully this will allow us to discover problems 'natively' before they become a problem for many users. Okeyes (WMF) (talk) 20:49, 14 May 2013 (UTC)
Fabrice, I'm not sure you understand the complexity and motivations of the WP editing community. The volunteers are self-selected for eagerness to work in a chaotic setting with unknown colleagues, none of whom can be compelled to work in a predictable way on a predictable project, and who can & do walk away any time they feel unwanted or uninterested. The volunteers have come here very specifically so they can work in that fashion, and will always resent people here trying to tell them how they must do anything, regardless of whether they are reasonable or unreasonable. This poses certain limitations, but its inherent in the existence of WP. Even for those things that we know the staff can do better, & I agree that designing interface is one of them, we want to feel not just that they are acting in our interests, but that they are acting as our agents and subject to our approval--at least in the things that primarily concern us as editors. (I think we know that for aspects primarily concern the readers, we editors in general have no particular expertise other than guesswork.) We are not likely to change: if we were prepared to accept things imposed on us other than by our own choice, we would not be here: we know better than anyone the great inefficiencies of our methods, but are here deliberately to take part in a very large scale effort to see how good an influential a resource we can make using them--and by now we have the reinforcement of having been successful beyond any expectations. I & many of us have worked -- and I hope successfully -- in much more conventional settings. We intend this to be different.
Nor do I think you understand the way to communicate here. Almost all of us do our work on this Wikipedia, not on meta or the mailing lists. Many of us, as you will have seen, resent anything not totally open and widely broadcast, as a potential cabal. There are too many places for announcements for any one person to track, but we have evolved some relatively good methods , such as site notices. It's hard to judge what is worth such broad notice, and I think you did not anticipate this, not realizing the already very weak nature of inter-editor communication.
Relatively few of us are particularly interested in technical or organizational issues. We are, rather, here for either creating content, or dealing with the surrounding functions (screening & evaluating submitted material, guiding new editors, copyediting and reformatting, constructing infoboxes, categories & other structural features, resolving disputes, protecting against disruption). Within the general anarchy, we've evolved patterns of effective work, and we know our patterns are very unstable, and are concerned that the best-intentioned and best-planned alterations will result in destruction. Stability in a tenuous structure like WP is a very delicate balance.
We know there must be improvements in many things; if we did not eagerly seek innovation we would hardly even have thought of coming here. But it is never possible for even the most experienced here to predict what will happen on WP, and what changes will be accepted. the only way of accomplishing things is by trial--and repeated patient trial. We will not accept trials we do not know about imposed from outside, and many trials we do know about will be judged unsuccessful. Some things go much too slowly, but it is impossible to force them. This example proves it: you and your staff have by now spent several times as much effort explaining and modifying after the fact as you would have needed to do had you consulted and trial;ed very widely before it. And the response we have received indicates to me -- and probably many of us -- that you still do not understand: you are looking for ways to do things efficiently, but we regard business-like methods as unfriendly. Nor do I think we will really accept the decisions of an elite group, as some have suggested. It is part of the basic approach here that we think of ourselves as equals, and even the decision-making groups that we ourselves elect formally (such as the admins and arb com) operate in an atmosphere of pervasive suspicion, and nobody is suitable for such jobs who cannot accept it.
As practical suggestions, probably the English WP is too large a structure to test things on; many things are probably best tried on the smaller projects. Even after that sort of a test, we still need to see if it works here, for we are more varied & less consistent than any other of the projects. The use of a beta test is really necessary, and those things that can be done on a sample should be (like protected changes, or article feedback). Most successful changes to the interface have started as opt-in features, that then became opt-out, and eventually universal. Some things do have to be uniform--notifications that cannot be depended on to be uniform are useless, and such features pose particular problems. And the key problem here of dealing with newcomers we know to be our weakest point, and therefore is particularly sensitive. We'd be very grateful for improvements, if you can suggest them.
We are deliberately very tolerant of mistakes. It is extremely hard to do anything wrong enough to be rejected by the community. The editing method has been constructed very deliberately to allow anything to be reversed, even at the sacrifice of stable content; the rules of procedure are designed as a basic principle to permit any exception, even at the sacrifice of consistent decisions. But we have built in this flexibility because we expect mistakes to be corrected. Speaking for not just myself, I think the error that you made in assuming approval of the changes (and it's clear that it was not unreasonable for you to have anticipated approval) would have been insignificant, if you had reversed it pending further discussion. That this still hasn't happened strikes some of as deliberate defiance of the community, and the rathe flowery language you use in complementing us while saying you intend to continue the same path strike us as what we expect from conventional administrators and bureaucrats. Conventional administrators and bureaucrats are what many us us profoundly distrust, both on principle and from experience--when we use the term for our own positions, it's meant as irony. If that's the way you intend to work, we need to consider how to defend ourselves against you.
Those are wrong who say that the responsibility for the programming is that of the foundation. The programming for WP was done primarily by volunteers, and the volunteers regard the foundation is their agent for certain purposes, including legal formalities and the custody of the physical resources. Physical resources are inescapable, and legal formalities are necessary; it is not uncommon for them to usurp a volunteer project, but I rather expect we will resist it. There is no need for such a confrontation if you can learn to work with us, and not depend on formal status within the foundation.
Please understand that I say this not of you personally, but of an entire general attitude. Six moths ago I literally walked out on an attempt to co-opt me into a formal role in one of the WMF-affiliated programs. Six years before, along with most of their volunteers, I quietly left Citizendium because the formal organization became counter-productive. WP was formed in large part by anarchists. I do not share it as a general social or political philosophy, but I honor it as a necessary corrective within the larger society, and I regard WP as the best illustration of Kropotkin's optimism about human potential. The WMF has the power to ruin this, and now might is the time we must persuade it not to make use of it. DGG ( talk ) 06:52, 15 May 2013 (UTC)
Hi DGG, Tito Dutta and others: thanks for your thoughtful responses to my last post, which I appreciate very much. I am particularly grateful for DGG's thorough description of the motivations and expectations of the WP editing community: while I was already aware of many of these attributes, it was helpful for me to read your detailed and eloquent overview. I look forward to discussing these points in more depth with you later today, and we'll definitely address these observations during our team retrospective for this project at the end of the week. Some of the general questions you raise about the respective roles of the foundation and community in making software decisions seem beyond the scope of this particular project talk page and should probably be discussed in a different forum -- which might include some of our other WMF colleagues, such as Eloquence, Philippe and/or Okeyes (WMF), who are more knowledgeable on this topic than I am. However, I would like to point out that key decisions regarding the OBOD were made in consultation with many other WMF team members, not by a small group of newcomers like me. I hope I have represented the team's rationale accurately and look forward to participating in a broader discussion of how the community and the foundation can collaborate more effectively for new software development, if someone can set it up in a more appropriate forum in coming weeks. For now, we will continue to engage the community in discussions and testing of important new features as soon as practical, for this and other projects we are working on. To that end, TittoDutta's proposals to form an editors' team of testers seems very reasonable -- and we would be happy to support this initiative. Thanks again to everyone for your constructive recommendations. Fabrice Florin (WMF) (talk) 17:36, 15 May 2013 (UTC)
A belated response - I'm not at home and have had trouble getting online. As DGG says, a lot of us are too busy with things here to know what is going on at Meta and other places. I still feel that a lot of the opposition to the Orange Banner was from people who had a personal dislike (an off switch coupled with the red pimple would have dealt with that), and fail to see how it could prevent people from getting messages, while a lot of the support from it came from the people who aren't seen on Meta, but who do have a lot of difficult contact with the newer users. Could some thought be given somewhere to getting a way of communicating with the people who work mainly in certain areas when developments that will affect them come up? A register of interests, maybe - mine would include the deletion arena, pages for translation and SPI for a start. Most of us do occasionally work in more areas - but the main ones as considered so by the user in question would do. Peridon (talk) 20:03, 16 May 2013 (UTC)
This is definitely something I'm interested in working on. With stuff like Page Curation it was easy - who are patrollers? People with patrols! More general projects are going to be a problem. But it's something we should work on, and something we will work on. Okeyes (WMF) (talk) 20:16, 16 May 2013 (UTC)
I have had the talk with Fabrice that he mentioned, though we mainly exchanged our understanding about the general working and decision patterns of the volunteers in the community and the programmers at the Foundation. They are always going to be different, and there are always going to be problems, but I hope the next round will go better. DGG ( talk ) 17:47, 17 May 2013 (UTC)
Revert notification
Latest comment: 11 years ago3 comments3 people in discussion
First of all thanks for implementing this new notifications system. I really like it. One of my edits has been recently reverted (see here) but I received no notification about it. Why was this the case? I've already received revert notifications before and they seemed to work... Thanks. --SnowBlizzard18:25, 12 May 2013 (UTC)
I just got a notification that my edit had been reverted, when in fact it was another user's edit which was reverted and I was simply the last person to have edited the page before the revert [1]. Sounds like a related issue. Jafeluv (talk) 09:38, 19 May 2013 (UTC)
Linking notifications
Latest comment: 11 years ago5 comments3 people in discussion
Love the tool, especially the functionality which indicates when pages you have created get new links (I think that is a great way to show the usefulness of materials throughout the community). However, I have noticed an inconsistency in the tool: I got a notification which said "William Blake Archive was linked from And did those feet in ancient time" when the link on And did those feet in ancient time is not recent by any means. Also, as someone who would like to see the newly linked article so that I can explore why it is linked and how to build the web more, it would be really nice to link directly to the newly linked article, Sadads (talk) 13:31, 14 May 2013 (UTC)
Odd! Let me know if it reoccurs? And, yep, we're talking through it - our designer has objections to many links, but it seems reasonable to me. Okeyes (WMF) (talk) 20:57, 14 May 2013 (UTC)
That's just strange :/. I'll stick it in bugzilla and see if our devs can work out a reasonable way around it; thanks! Okeyes (WMF) (talk) 18:48, 16 May 2013 (UTC)
I think this may be related.
Can someone explain what in fact triggers the "Page links" notification.
The documentation states
"Page links: when a new link is made to a page you created"
but does not say anything about indirect links or the interaction between links, redirects, and renames.
But this edit to Hitler Cabinet generated the notification
"Federal Ministry of Economics and Technology was linked from Hitler Cabinet: See all links to this page. Yesterday at 01:50:02"
The diffs in the email notices and on the old orange bar were the most useful part of the old notification system. I'm actually quite shocked that they weren't integrated into this system before its release. I'm seeing all kinds of things about colours and bars (and actually, that new yellow bar isn't obvious at all and I managed not to even notice it for 20 minutes), but it's still only going to my talk page, not even to the right section, and obviously not to the diff. I'm good with the idea of adding functionality, but I'm getting quite concerned that 10 days in, we're still not back to anything close to the old functionality. I'm not trying to trash this new system, but I'm really disappointed that we've moved so far backwards here. Risker (talk) 02:25, 15 May 2013 (UTC)
And perhaps to prove the point...someone left me a talk page message on another wiki that has this enabled. And I scrambled around here on this project trying to figure out where in heaven's name the darn notification was on my talk page, until I realised that it was on another project. A link is not good enough. Give us diffs please. While you're at it, why is this new feature live without having been checked out on suppression? We were told after the last big new feature went live without suppression ability that nothing would be enabled even on opt-in basis that did not have the suppression facility sorted. Risker (talk) 03:03, 15 May 2013 (UTC)
Apparently, the temptation to reinvent the wheel constantly is just too strong. :) --Nemo09:16, 15 May 2013 (UTC)
I'm confused; Risker, I was under the impression that you were pretty heavily involved in the conversation about suppression yourself? Okeyes (WMF) (talk) 21:06, 15 May 2013 (UTC)
Not for this feature, at least not in its current iteration; I may have had some questions put to me some time ago in a hypothetical manner. Indeed, I didn't know about the concerns until it was pointed out here, and I certainly didn't have any opportunity to test before go-live. Risker (talk) 01:06, 16 May 2013 (UTC)
Hi Risker, thanks for bringing up this important issue, which is very high on our to-do list. We discussed your concerns with the product team yesterday and propose this short-term solution, if it works for you. We would like to add a '(show changes)' link for all talk page messages in the upcoming HTML email notifications, as well as in the archive page. So users will have the option to go straight to the diff page by clicking on this link if they want. This is a tough challenge to solve, because we're serving both new users and experienced users like you with this feature. To a new user, the diff page is largely incomprehensible and useless as currently designed. To an experienced user like you, it is incredibly useful. At the same time, it's not a good idea to offer completely different functions for different user groups, so we need a solution that serves them both effectively. For now, we would like to first add this 'show changes' link on the archive and HTML emails -- and hold off on adding it to the flyout, because putting more than one link per notification in that limited space tends to overwhelm users with too many decisions. However, we will test this specifically with users in coming weeks, and are prepared to revisit this temporary decision after we have more feedback. Would this proposal work for you, as a first step? (you can track this T50183 on Bugzilla) Fabrice Florin (WMF) (talk) 18:04, 15 May 2013 (UTC)
I'd note that this is in no way the final feature; I, for one, very much want diff links in the on-wiki notifications. Okeyes (WMF) (talk) 21:06, 15 May 2013 (UTC)
Andy, I appreciate that the process here has been highly frustrating for everyone involved; as someone who looks at things from both a Product and community mindset, I've been getting both sides of the issue. But I don't think we do the process or ourselves any favours by perpetuating posts beyond the point at which they contribute to the discussion, and I'd argue this does that. Okeyes (WMF) (talk) 18:42, 16 May 2013 (UTC)
We need the diff link on the Orange notification bar so the you can get a view of the change via the pop-ups without actually opening the link. Keith D (talk) 11:13, 16 May 2013 (UTC)
Agreed. I'm having that conversation now; it'll hopefully branch out into wider on-wiki discussion so we don't make decisions in a vacuum, or behind closed doors. Okeyes (WMF) (talk) 18:42, 16 May 2013 (UTC)
Seconded. I use alpine (email client) and so cannot easily view HTML emails. HTML-only emails also frequently end up in my spam folder (almost always correctly), where they may sit for several weeks. Thryduulf (talk) 04:01, 16 May 2013 (UTC)
Agreed, HTML email for a simple notification is major overkill. If we are trying to plan for a wide variety of mobile devices, and keep things lean for people with ancient computers and on slow Internet connections around the world, a plain text option is important. (I understand that a long diff link might be unattractive to some, but it's not hard to get used to clicking something -- it's not like people need to read the diff link.) -Pete (talk) 05:54, 16 May 2013 (UTC)
FFS. "There will not be any email format preferences unless we hear a strong user demand for this" - I wish to express strong demand for a user preference option for plain text. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits09:57, 16 May 2013 (UTC)
In case my comments immediately above are in any way ambiguous, I too wish to express my view that plain text emails are absolutely essential for accessibility and other reasons.
Please bear in mind that HTML emails can't be as accessible as text emails, primarily because most email clients were originally designed to deal only with text, and the HTML functionality has been added to their features, rather than being at the core of the design. This results in the email client providing relatively poor support for screen readers, which can misread the text. We recommend that, wherever possible, recipients are offered a choice of email format, to avoid this problem.
Thanks for your feedback about HTML Email for notifications. Please note that plain text emails will continue to be sent along with HTML emails, so if your email client cannot read HTML emails, you will see the plain text version instead. Also note that most email notifications are turned off by default for current users, so this may be a moot point for users who do not chose to enable them. That said, HTML email notifications are now widely used on other top sites nowadays, as shown in these screenshots of best practices -- and I am not aware of any major destination that still uses plain text emails as a primary communication medium. The main reason HTML emails are so widely used is because they present information more effectively than plain text emails, by using clear visual cues to inform users of new activity -- and invite them to take action more easily. For these reasons, we are not planning to provide a preference to switch between HTML and plain text email for this first release, but will closely monitor community feedback on this topic in coming weeks to plan our next steps. Fabrice Florin (WMF) (talk) 17:50, 16 May 2013 (UTC)
Indeed, I was about to ask why Wikimedia is comparing itself to social media sites when it is explicitly not social media? Thryduulf (talk) 18:09, 16 May 2013 (UTC)
LinkedIn is social media now? Trello, for example, is a project management tool. Onorem: if your standard is "only include changes that are based on what another online encyclopedia does" we'll either never change a thing or stop most of us from editing.
Thyrduulf; I totally agree. But we are going to give people a choice; it's not about making design decisions that only benefit a certain subset of users, it's about having options that allow for all users to be covered :). Okeyes (WMF) (talk) 18:16, 16 May 2013 (UTC)
I would note to everyone that Fabrice said, above (I appreciate it may be lost in the discussion - I'm very much looking forward to Flow :/) that HTML emails will display as plaintext to those with email clients configured to avoid HTML. Okeyes (WMF) (talk) 18:39, 16 May 2013 (UTC)
You again answer a different issue to the one raised. Nobody has claimed that the emails will not be multipart; but that some people - including me - want to receive text only. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits22:07, 16 May 2013 (UTC)
Then I'm confused. The argument made was for accessibility; my point is that the sending of multipart emails allows for this to be factored into acconut at the user's end. What's the argument for plaintext-only? If it's data transfer costs, that's something we should discuss, but it's a totally different argument from the RNIB-citing conversation. Okeyes (WMF) (talk) 11:47, 17 May 2013 (UTC)
Tim Starling pretty roundly debunked the myth that HTML e-mail is bad for accessibility in a mailing list post. I can't seem to find it right now, though. :-/ --MZMcBride (talk) 19:27, 16 May 2013 (UTC)
For what it's worth, as a screen reader user, I find that HTML emails are no longer a problem (I used to hate them with a passion). I only have experience with JAWS and NVDA, but other modern screen readers work similarly; they present *any* HTML textarea as a virtual buffer and allow users to navigate it easily. I think the above RNIB link is a bit out of date in this respect. Graham8707:33, 17 May 2013 (UTC)
Having said that, we really should make sure the HTML is formatted rationally for screenreaders; any objection to me giving you a poke once we've got something up so you can test it? Okeyes (WMF) (talk) 11:47, 17 May 2013 (UTC)
I'd be interested to know the justification for refusing to allow users on per-megabyte dataplans the option to turn off redundant and wasteful multipart emails in favour of light plain-text. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits22:07, 16 May 2013 (UTC)
I think your case would be bolstered by evidence that the increased message size would have a non-negligible impact. The bandwidth issue was discussed in Tim's e-mail. --MZMcBride (talk) 05:15, 17 May 2013 (UTC)
We can't know all the possibilities; but for example: a mother in the global south, with an expensive mobile data plan and no easy access to healthcare, and who doesn't edit Wikipedia, signs up and watchlists, with email notifications, the stub about the disease her child has, in order that she will receive an email when the article is changed or expanded. A student who want to be notified - with an edit summary - that their talk page has changed, on their expensive mobile data plan, in order to travel to the next town where they have landline access to edit. An employee who can only receive email notifications at their workplace, which blocks HTML or multipart emails. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits12:51, 17 May 2013 (UTC)
Your first suggestions could be shortened to "internet access is expensive for many people", which is a valid concern. However, for the latter, allowing people editing Wikipedia while on the clock to cope with dodgy firewall rules is hardly the Foundation's problem. — Scott•talk14:39, 17 May 2013 (UTC)
Ladies and gents, this seems like bikeshedding in the extreme. I dislike HTML email as much as the next old-school geek, but plain-text vs. HTML email seems like a pretty ridiculous thing to give an enormous shit about. The issue seems to be "the emails are a bit bigger". Have you loadedWP:ANI recently? Isn't there an encyclopedia to write? —Tom Morris (talk) 13:08, 17 May 2013 (UTC)
It seems strange that nobody seems to have mentioned the potential privacy concerns regarding HTML e-mail. Accessibility is one sword to wield, but surely privacy is another. HTML e-mail allows for much easier tracking via URL (with obfuscated or hidden URL parameters) and can track users by loading arbitrary server-side content (such as images). Interestingly, the Wikimedia Foundation privacy policy seems to go out of its way to exclude itself from governing any user tracking/data collection of this nature. Perhaps this sword is stronger to fight the evil beast that is HTML e-mail. ;-) --MZMcBride (talk) 04:12, 18 May 2013 (UTC)
Date format bug
Latest comment: 11 years ago8 comments4 people in discussion
Hi, in Finnish localization the date format is wrong. It says "TOUKOKUUN 12" ("MAY'S 12") while it should say "12. TOUKOKUUTA" ("12ND MAY"). Where can this be changed ? EDIT:Also, the date after the notification is not localized,it's allways English. --Olli (talk) 15:54, 17 May 2013 (UTC)
Hi. Would it be possible for you to upload a screenshot of what you're seeing? I think it would make resolving the issue much easier. --MZMcBride (talk) 16:58, 17 May 2013 (UTC)
Hi, thanks for you both. The fixing of the genitive case issue seems to be taking quite long time. I think it's not so difficult job to do... --Olli (talk) 05:22, 18 May 2013 (UTC)
You just see the translation, but these are algorithms and code that need to be fixed and rewritten. It's much more complex than it looks. —TheDJ (talk • contribs) 20:35, 18 May 2013 (UTC)
Latest comment: 11 years ago18 comments8 people in discussion
This message should have given me a mention notification, but it didn't. Is that because Cyberpower678's signature doesn't link directly to his user page, or some other reason? Or am I just losing the plot? Graham8703:32, 15 May 2013 (UTC)
@Graham87: After testing with my bot account ([4] — I did not get a mention notification), I think your hypothesis is correct: The signature of the user mentioning you must have a direct link to their user page and/or (?) their talk page. Interesting, isn't it? The Anonymouse (talk | contribs) 04:16, 15 May 2013 (UTC)
This would cause a problem for users who just link to their talk pages in their signature rather than their main user page (JzG, I'm "looking" at you). :-) It should probably be changed to accept either a user page or a talk page link. Graham8705:24, 15 May 2013 (UTC)
Ah, didn't notice the question mark in your message; my screen reader just paused as it would for a comma! This edit caused my main account to get a mention notification, so a link to a talk page is all that's needed. Sorry, Guy, for dragging you into this. Graham8706:30, 15 May 2013 (UTC)
I would certainly be unhappy with trying to require anyone to link to their user page, I have a redlinked user page for good [FSVO] reason. Guy (Help!) 10:25, 15 May 2013 (UTC)
I've now added the initialism to Wiktionary: wikt:FSVO. If you can give a concise definition for "for some values of" in this meaning please add it. Thryduulf (talk) 22:45, 15 May 2013 (UTC)
Thanks, both :). I'm not sure what the solution is here, to be honest :/. I get the problems of mucking with user signature requirements, but the alternative is to muck with a lot of onwiki templates and processes (can you imagine if a mere user talk link triggered it?) Okeyes (WMF) (talk) 18:47, 16 May 2013 (UTC)
I assume that a link to a non-existent user page would also work, because if it didn't, then this would fail for most new users. Also, if necessary, you could create a non-visible link like "" (yes, there's a link to User:JzG's user page between those quotation marks). WhatamIdoing (talk) 00:58, 17 May 2013 (UTC)
Yeah, it'll work fine; I think there's plausibly an argument to be made for not linking to userpages that don't exist (it could cause some confusion) but linking seems to be the least-onerous and problematic solution here. Okeyes (WMF) (talk) 11:49, 17 May 2013 (UTC)
As an aside, I recommend any admins or other long-standing users either to have a redlinked user page or an alternate account with a redlinked user page and a significant edit history. You would be amazed at the opportunities this provides for "teaching moments", and also at the valid points that come up re your own editing. Do use the email user feature for one-to-one feedback, though, as it is a lot less public and thus less likely to cause defensive reactions. Guy (Help!) 22:21, 19 May 2013 (UTC)
Certainly, but I'd make an argument it causes damage to the community as a whole - it breaks everyone down to identity-less pseudonyms. Okeyes (WMF) (talk) 18:17, 20 May 2013 (UTC)
Not at all; pseudonyms are not linked to your real-world identity - there's a big distinction. Someone can have multiple identities, all tracking different aspects of their achievements, actions and persons. So, for example, I'm Oliver Keyes, but I'm also Ironholds. At a time, these were completely distinct; Ironholds was my online handle, Oliver how I handled myself in the real world. My actions as Ironholds did not impact on my reputation as Oliver. But this is not the same as it not having an impact. Ironholds was my name for Wikipedia, the wider movement, code, gaming - something that impacted on any one of those things impacted on my standing everywhere. Indeed, if I chose to identify who I was in meatspace while I was Ironholds, it impacted everywhere. Pseudonyms can lack an identity behind them, but often don't. Okeyes (WMF) (talk) 16:08, 21 May 2013 (UTC)
I've seen what the notification is notifying me of. Get rid of the box.
Latest comment: 11 years ago13 comments5 people in discussion
That's just one of the problems forced upon us. One has to click thru ones talkpage history to see if there is another new edit(s) made before the announced one. The big orange bar lead us to the last change diff; So why is the new system doing away with it? Can anybody see an improvement in this? I don't. Oh, and yes, it doesn't away once you checked your talkpage. Very confusing and annoying. TMCk (talk) 01:48, 12 May 2013 (UTC)
Oh, boy. This damages one of the main purpose of the notification: making sure that new users see the messages left to them, so they can't claim that nobody warned them. --Enric Naval (talk) 10:36, 12 May 2013 (UTC)
I'm not sure; I can't give cast-iron guarantees of which patches are going to be done quicker than the others. I'd hope that it'd be included with the concrete rollout of the new talkpage notification, but I'll find out. In any case, I'll post a 'here are the things the new release will do' newsletter when said new release comes. Okeyes (WMF) (talk) 18:42, 14 May 2013 (UTC)
Looks like Kaldari has a meeting with our designer today to talk about this precise thing; I'll let you know how it goes. Okeyes (WMF) (talk) 19:56, 14 May 2013 (UTC)
From what I see, there's a new option that floats with the page (more intrusive than the old style) and goes away when viewing the talk page...but we're still stuck with a red # until we use the precious new version to get rid of it. --Onorem♠Dil17:29, 21 May 2013 (UTC)
Still nothing? Is there a timeframe that seems reasonable to expect a response? I thought a week was good, but oh well. --Onorem (talk) 19:14, 22 May 2013 (UTC)
Question 2 in the feedback survey
Latest comment: 11 years ago2 comments2 people in discussion
I tried taking the survey a few minutes ago, having received a number of watchlist emails, but no new-style notifications since the feature was deployed. However, even though I answered "no" to Q1 ("Did you receive notifications this week?"), the survey engine still wants me to answer Q2 ("are they useful?"), despite the "If you received new notifications:" conditional. If WMF people are still looking at that survey, is there a recommended answer if/when that question is not applicable? Likewise, it's unclear to me whether that conditional is also meant for Q2, or also for Q3 and Q4, and what to answer for Q4 if the latter. (FWIW, my answer to Q4 is: a few edits a week to 1 edit a day on average, because more often than not, there's a nit for me to pick at on the page I read, but I haven't kept track.) The Crab Who Played With The Sea (talk) 10:01, 22 May 2013 (UTC)
Hi The Crab Who Played With The Sea, thanks for bringing up this issue. We have now changed the survey so that Q2 is no longer 'required' -- and Q1 asks if you received notifications 'recently'. So you should now be able take the survey. It's also fine for you to say that you're 'not sure' in Q2. For those of you who are interested, about 62% of survey respondents find notifications useful (vs. 19% who do not), out of 440 responses so far. You can monitor the live results here. We are in the process of analyzing the comments field, to identify frequent concerns and feature requests -- and expect to publish a report in coming weeks. Thanks again for contacting us about this! Fabrice Florin (WMF) (talk) 18:27, 22 May 2013 (UTC)
Orange bar and notification count
Latest comment: 11 years ago15 comments9 people in discussion
When clicking on the new orange bar and reading the talk page this fails to clear down the count in the notifications area. When reading the talk page the notification count should be reduced by those notifications that relate to talk page messages. You should not have to enter the notifications area to get the counter reset. Keith D (talk) 11:29, 16 May 2013 (UTC)
No danger from me on that front...but you'll need to provide proof that after a few days, "We" have all got used to it and agree that it's an improvement. --Onorem♠Dil22:58, 16 May 2013 (UTC)
Is there any movement on sorting out the red box? While I understand the developers don't want to hear complaints, I'm finding it tedious having to click on the red box after checking my messages. Currently we have reduced functionality and visibility but increased maintenance of talkpage notifications. Such things are frustrating and demotivating, and I think it's important to feed that back. SilkTork✔Tea time16:00, 21 May 2013 (UTC)
Well, I could (surprise!) write a script to do that in the meantime, if you'd like. It was part of one of my old OBoD scripts, so it should be pretty easy to make. Writ Keeper⚇♔16:02, 21 May 2013 (UTC)
Hi folks, we know this is a high priority issue, and we're working diligently to solve it. If all goes well, we hope to release a fix by the end of next week. Meanwhile, you can track our progress on this Bugzilla ticket: T49912. Cheers. Fabrice Florin (WMF) (talk) 01:01, 24 May 2013 (UTC)
Deletion notification
Latest comment: 11 years ago2 comments2 people in discussion
Is there a way for article/template/category etc. creators (i.e. the first editor on the page) to be 'notified' if the relevant page is taken to CSD/PROD/XFD? I know that you are meant to notify them as part of then process, but as an Admin primarily working in the deletion areas of Wikipedia, I see a lot of good-faith editors failing to notify, through haste or ignorance etc. GiantSnowman10:32, 22 May 2013 (UTC)
Hello GiantSnowman, thanks for your question. Notifications are now sent to page creators when a new page they started is marked for deletion, as described here. However, this only works for new pages reviewed through the Page Curation tool at this time. So we encourage patrollers to use this tool, to make sure that creators are notified of new page reviews :) Fabrice Florin (WMF) (talk) 00:52, 24 May 2013 (UTC)
Mentioned but no notification
Latest comment: 11 years ago5 comments4 people in discussion
Mentions only work on talk pages and explicitly whitelisted pages (ANI, the Village Pumps, etc.). I would assume it hasn't been turned on for DYK pages. Devs, am I correct? Ignatzmice•talk03:06, 21 May 2013 (UTC)
There's no whitelist; the problem is that that's in the template namespace. It operates in the Wikipedia, Talk, User talk and Wikipedia talk namespaces. Okeyes (WMF) (talk) 15:22, 21 May 2013 (UTC)
When I click the notifications button, I don't see the notification - the number goes to zero, but I stay on the page I was on rather than going to the notifications page. However, if I click on it before whatever page I'm on is fully loaded, I do go to the notifications page. I can also get there by right clicking and asking to open it in a new tab or window. I'm using Windows, Firefox, Modern skin
I see something about a "flyout" on the page about the tool; that doesn't appear at all for me
The links to "mentions" break when there is a wikilink in the section heading of the relevant page. For example, the link in the notification about this diff should have brought me here, but instead brought me here
There is a white line in the heading on the notifications page. Hovering over it, there appears to be the survey link on one side and the preferences link on the other, but I can't actually see either of those things because of the colour of the text/icon relative to the background colour. Nikkimaria (talk) 21:46, 1 May 2013 (UTC)
I deployed some fixes locally for the modern skin. It's not complete, but at least it should make the flyout workable for you. 4 is also hidden away in that report btw. See the point about "H1 positioning". 3 might require separate reporting. —TheDJ (talk • contribs) 22:28, 1 May 2013 (UTC)
Yes, that page is also fully customized and only taking two skins in mind, causing trouble for the two unsupported skins. :( —TheDJ (talk • contribs) 22:59, 1 May 2013 (UTC)
Commenting here since you're here, but I don't have a major problem with the stuff on the header - it kind of makes sense and nothing changes up there except for a few scripts mostly limited to mainspace. Better it look different than not have the helpful links, jah? At any rate, every now and then the big ol' 0 has a big space next to it, mainly when editing. SO it looks like "AMORY 0 TALK" which is weird. ~ Amory(u • t • c)00:16, 2 May 2013 (UTC)
Another bug?
Why was my little red numeral a "1" this morning when I had two new messages (from two different users), both of which arrived while I was logged out? Rivertorch (talk) 16:21, 2 May 2013 (UTC)
I don't think that's a bug; it just collapses all the "you have a new message" things into a single notification and lets you know all at once. Writ Keeper⚇♔16:24, 2 May 2013 (UTC)
All right. Then that's a disadvantage of the new system. The OBOD told you how many new messages there were. Or rather from how many users. (SineBot counted as a user, which was sort of confusing. Maybe this isn't a big disadvantage.) Rivertorch (talk) 18:54, 2 May 2013 (UTC)
I don't know. It required clicking on and then counting the items, rather than the sum just being there on the page at a glance. It's not exactly onerous, but any decrease in functionality or user-friendliness isn't a good thing. Let me say that I thought I was going to warm to this feature because I imagine some real advantages in helping keep track of things, and I still think it has a lot of potential, but I'm beginning to think it was a bit of a blunder to eliminate the OBOD. A little redundancy wouldn't hurt when it comes to notifications about new messages in user talk, and I'm not clear why an opt-in choice for continued use of the OBOD is being denied. Is there a technical reason or what? (Apologies if that has been answered already. This page is an vast wall of text, and I have given up trying to follow every thread.) Rivertorch (talk) 17:22, 3 May 2013 (UTC)
Yeah, I get that problem. A count (or at least 'you have messages from [user/users/delete as applicable] is in the works. Sorry for the issues; as you can see from said wall of text, we've got quite a few bugs and we're working through them as we can :). Okeyes (WMF) (talk) 16:39, 24 May 2013 (UTC)
Only the righthand side of the message is shown
When I click my icon, I do not get an entire flyout screen. It is cut off so that I only see: ATIONS ? The screen is too far to the left. I've tried enlarging and shrinking my page size and tried eliminating my toolbars and still I can't see the left side. Funnily I have part of a post that says: BONO POSTED ON YOUR TALK PAGE. Please everyone run to see... ツ (Sorry it's now in an archived page). Is there some little knob I can twist to correct this? Fylbecatuloustalk02:39, 3 May 2013 (UTC)
Update: please refer to my edits above in the section entitled "Not getting a link": I now see that I have Bug 47932. Call the exterminator. Fylbecatuloustalk23:55, 3 May 2013 (UTC)
I'm using Monobook with Iceweasel on Debian at the current version of Squeeze. I'm getting numbers for talk page messages, but there is no red box surrounding them. Is this expected behavior for Monobook? Sophus Bie(talk)13:57, 3 May 2013 (UTC)
Oh, bother. I'm guessing, then, that it's because I have javascript disabled. (Please don't give the obvious solution, "enable javascript".) I didn't expect the flyout to work, but the lack of red highlighting is an issue. At least Special:Notifications itself seems to be working as intended. Sophus Bie(talk)14:13, 3 May 2013 (UTC)
Indeed! I wish they would enable using both. It may have been low-tech, but at least low-tech almost always works. Sophus Bie(talk)14:39, 3 May 2013 (UTC)
Not to badger your choice, but is there a particular reason you have JavaScript disabled? That closes off a lot of handy functionality, not just on Wikipedia, but pretty much everywhere on the Internet. EVula// talk // ☯ //17:41, 3 May 2013 (UTC)
For almost everything, I find the speed penalty outweighs the added functionality. With a sample size of 25, I've measured that javascript slows my browsing to about 1/10th of my otherwise expected speed. Sophus Bie(talk)00:04, 4 May 2013 (UTC)
Echo does work with Javascript disabled. You just don't get the flyout. If anyone knows how to create fancy flyouts without Javascript, I'd love to know :) Kaldari (talk) 18:29, 3 May 2013 (UTC)
@Kaldari: There's not going to be all the functionality in this that you need, I've spent no time trying to work out how/if it and php could ever give you the interaction you need - and maybe it can't... it may be utterly useless for this purpose, probably a zillion browser compatibility issues, but this (from a quick Google) indicates that kind of thing is not totally impossible: [5] (worked for me in Chrome 28.0.1496.0 dev-m with JS disabled - no other tests done) Begoontalk12:32, 4 May 2013 (UTC)
Clicking on the orange bar doesn't cancel the red box
I received a message on my talk page and clicked on the (pitifully small) orange bar to view it. After reading it and navigating to another few pages I then noticed the read box was still saying I had one notification. I clicked on the red box and was told again about the same talk page message. As the orange bar disappeared the system obviously knew I had read the message, so this would seem to be a bug. Thryduulf (talk) 20:44, 15 May 2013 (UTC)
Yes, I have the same problem. I find it tedious having to go to my talkpage and then click the red box to remove two notification. Just one is enough. The orange bar is fine. No need for the red box at all! SilkTork✔Tea time15:16, 21 May 2013 (UTC)
but the new system is very valuable for other things than talk pages. I very much appreciate the notices of when I'm mentioned in a discussion, for example. Dealing with both, however, is an utter nuisance. One way to deal with this is to let us turn off the notifications for talk page if we have the orange bar enabled. But even better would simply to incorporate both systems, and give a choice, with the default being the orange bar. DGG ( talk ) 20:18, 21 May 2013 (UTC)
Bug or feature?
Latest comment: 11 years ago5 comments3 people in discussion
I just got a "you got mentioned" notification (along with scores of other people) when someone accidentally transcluded one of the reference desks while meaning to link it. Was the software intended to notify us when links to our usernames get transcluded? I found it confusing, since my username wasn't on the code of any recent revisions of that page. Nyttend (talk) 00:21, 28 May 2013 (UTC)
@Nyttend: I use it as a feature for things like sending notifications to talk page stalkers automatically, and {{ping}}ing people, but it certainly makes it easier for anyone to spamalot. 930913(Congratulate) 00:44, 28 May 2013 (UTC)
Latest comment: 11 years ago4 comments3 people in discussion
got a notification today:
71.2.172.65 mentioned you on Wikipedia:Administrators' noticeboard/Edit warring.
Yesterday at 19:29
I don't see it anywhere, including the history of that user/time and around then. I wasn't involved in whatever it is they are talking about, anyway, so I figure this has got to be some sort of misfire. (i also got a previous notification of an edit to my talk page that was correct. not related to this item) Gzuckier (talk) 06:35, 29 May 2013 (UTC)
Latest comment: 11 years ago4 comments2 people in discussion
A Thanks notification.
Hi folks, here's an update on some of the new Notification features we released today on the English Wikipedia -- as well as our next steps:
1. Thanks notification
This experimental feature offers a new way to give positive feedback on Wikipedia. It lets editors send a private 'Thank you' notification to users who make useful edits -- by clicking a small 'thank' link on their history or diff page, as described above and on this overview page. The purpose of the Thanks notification is to give quick positive feedback to recognize productive contributions. We hope that it will make it easier to show appreciation for each other's work -- and it should be particularly helpful for encouraging new users during their first critical steps on Wikipedia. We have intentionally kept this notification as simple as possible, so we can all evaluate it and improve it together. Once you have had a chance to try it out, we welcome your feedback about this feature, and look forward to a healthy discussion on this talk page.
2. New features
Today, we also released a couple more features and bug fixes on the English Wikipedia and MediaWiki.org:
Talk page messages are now marked as read when you visit the talk page -- and the red badge is reset as well (Bug 47912)
Tooltips in preferences help explain how notifications work when you hover over question marks (Bug 47094)
These revisions were made based on community feedback and we would love to hear if they work for you. If you come across any bugs, please report them here, or post them directly on Bugzilla.
3. Next release
For our next releases, we're now focusing on these features, which were also requested by many community members:
Provide 'diff' links in the flyout for talk page messages and mentions (Bug 48183)
4. Next steps
Our follow-up goals for Notifications include:
Complete core features next month
Deploy Echo on international projects this summer
Develop cross-wiki and a few key features this fall
Support mobile + multimedia teams and developers
Our preliminary roadmap for 2013 is outlined in our E2 planning slides for editor engagement features.
Please let us know if these project updates are helpful to you -- and we will aim to post them every few weeks. In the meantime, many thanks to all the community and team members who have helped us create and improve this project! Fabrice Florin (WMF) (talk) 00:18, 31 May 2013 (UTC)
"Talk page messages are now marked as read when you visit the talk page -- and the red badge is reset as well" - Thank you. --Onorem (talk) 00:35, 31 May 2013 (UTC)
Latest comment: 11 years ago6 comments5 people in discussion
Notifications Flyout Single Targets Closeup
Hi folks, here's a quick update on what we've been working on for the Notifications project this week.
Here are some of the revisions we deployed on the English Wikipedia this week:
Talk page notifications now link directly to sections (when available)
Page link notifications now go to the article that linked to your page (unless page links are bundled)
Users can now be blacklisted by the community (if their notifications are deemed inappropriate)
Minor copy tweaks and link fixes in notification emails
Slightly resized the red badge and alerts
Most of these revisions were made based on community feedback and we would love to hear if they work for you. If you come across any bugs, please report them here, or post them on Bugzilla.
We're now focusing on these features, which were also requested by many community members:
Reset badge when you visit the talk page (Bug 47912)
Provide diff links for talk page messages (Bug 48183)
For the second feature, our design team recommends that we add secondary action links next to the timestamp on the flyout, as illustrated in this mockup (see thumbnail to the right). So for talk message notifications, the primary link would go to the talk page, but a secondary link to the diff page will also be available, for people who prefer that solution. We also plan to make the entire notification one large target, as suggested by several community members -- and to be consistent with best practices and user expectations on the web. When users hover over a notification, that target will underline in the flyout, as shown in the mockup. This seems to be the most practical solution, after careful review of all our options. The good news is that this solution lets us provide these secondary links on all touchpoints, including flyout, archive and most email notifications.
We aim to make progress on both features next week -- and will deploy them as soon as they're ready. This will make it easier for you to comment on their usability based on actual code, rather than on mockups. But we wanted to give you advance notice right away, so you know they're coming. :)
In other news, here are some of our first metrics for notifications on the English Wikipedia, as shown on our first metrics dashboard. About 295k notifications were triggered in the three weeks since we deployed on May 1st -- or a daily average of about 12k notifications. About 62% of these notifications were for existing users, and 38% for new users. About 45% of total notifications are for talk page messages, vs. 35% for system messages (welcome and getting started notifications); edit reverts represent about 9% of notifications, and the rest are split between page reviews, page links and user mentions.
Our ongoing survey results suggest that about 62% of respondents find notifications useful - versus 20% who do not find them useful. This is from about 500 respondents, who tend to be active editors (approx. 65% edit once a day). In coming weeks, we will provide a more detailed report on what we're hearing from both current and new users.
The little "Show changes" item in the image: will that take me to the diff that triggered this notification, or to all changes since the last time I visited that page? WhatamIdoing (talk) 20:46, 24 May 2013 (UTC)
Latest comment: 11 years ago8 comments3 people in discussion
I get a Red '1' at the top of my browser but cannot click on it to access Special:Notifications. This was working about a few hours ago and I've turned my computer on & off. A bug or just my browser having issues? GiantSnowman21:21, 30 May 2013 (UTC)
I got that mention just fine. I'm on Firefox 21 in Vista. I've had a similar issue before, but it's always cleared itself right up. You could try logging out, clearing cache and cookies and logging back in. If it's still there after that, and there is nothing in the dev console as Okeyes suggests, highlight the whole personal toolbar and viewing the source, there may be another element that you can't see that is stacked on top of it with a z-index or something (I've had that trouble before). Technical 13 (talk) 17:59, 31 May 2013 (UTC)
Browser = IE10, OS = Windows 7. I've logged out and purged my cache again. Can someone 'notify' me please to see if it works? GiantSnowman18:05, 31 May 2013 (UTC)
No, I don't usually but must have mis-clicked it; I've reverted back to normal and it's fixed the notifications issues. Thanks! GiantSnowman18:21, 31 May 2013 (UTC)
Watching categories
Latest comment: 11 years ago3 comments3 people in discussion
New notification: a quick way to say "thanks" for an edit
Latest comment: 11 years ago29 comments14 people in discussion
A Thanks notification.
Hi folks, we would like to let you know about our upcoming Thanks notification -- a new experimental feature which we plan to release on the English Wikipedia in coming days. This feature will enable editors to send a private 'Thank you' notification to users who make useful edits -- by clicking a small 'thank' link on their history or diff page.
The purpose of this notification is to give quick positive feedback to recognize productive contributions. We hope that it will make it easier to show appreciation for each other's work -- and it should be particularly helpful for encouraging new users during their first critical steps on Wikipedia. This small feature is now being tested on MediaWiki.org and we aim to release it on the English Wikipedia at the end of this week -- or the following week. We have intentionally kept it as simple as possible, so we can all evaluate it and improve it together, based on user feedback.
We welcome your feedback about this feature, and look forward to a healthy discussion on this talk page, once you have had a chance to try it out. If you would like to test it in advance, you can do so on MediaWiki.org right now, as outlined on this testing page. And any user who does not want to thank others or be thanked will be able to disable this notification in their preferences, as described here.
I've given thanks on mediawiki.org and can't wait to be on the receiving end to see what that is like. I look forward to it. :) Technical 13 (talk) 19:02, 28 May 2013 (UTC)
Perfect. Throw more bells and whistles at the thing before it's figured out. Wasn't there some other tool that already did this? --Onorem (talk) 19:11, 28 May 2013 (UTC)
If you've got specific issues with the tool we're not either resolving or have confirmed we're not resolving, as always, I invite you to bring them up. The other tool you're thinking of is Wikilove, which provides things like barnstars. As the thanks sub-page notes, barnstars are kind of the nuclear option; a lot of people don't give them for minor contributions on their own because it would devalue the award. Okeyes (WMF) (talk) 19:49, 28 May 2013 (UTC)
I saw this on Mediawiki the other day, and I think it's a great idea: click 'thank' with the same ease that you might click 'undo' and the person gets a quick message. It looks like it's logged but not a visible edit, so WP:Editcountitis won't drive thank-spam and no one really sees it except the person you're sending it to (unless someone checks the logs). This is a nice solution for encouraging everyday expressions of gratitude. WhatamIdoing (talk) 20:15, 28 May 2013 (UTC)
What if you added a "Did you click the right button?" option? I like the idea a lot, but given the link's location, I can imagine administrators accidentally thanking people when they're trying to click the (block) link, and the same is true for anyone trying to click the (undo) button. Nyttend (talk) 21:54, 28 May 2013 (UTC)
I can't wait until the "upload pictures of your supper via Instagram" feature gets rolled out. I assume, of course, that I can disable this latest Facebook feature clone? Resolute00:21, 29 May 2013 (UTC)
While I admit that Reso's tone was snarky, you must have caught on by now that many of us find this kind of cutesy-poo Facebookery repugnant and undignified and altogether inappropriate for an encyclopedic reference project. (I personally find the little cartoon heart especially appalling, more fitting for a pre-teen otaku forum than for us boring old grown-ups here.) It's harder to argue for the scholarly value of this project when bombarded with Wiki-love kittens and hearts. --Orange Mike | Talk13:27, 31 May 2013 (UTC)
Anybody who's seen my userpics or met me in person or looked at some of my userboxen or read my Facebook (yes, I have one) knows I'm not really a You kids get off my lawn! kind of guy. But something about officially institutionalizing things like smiley faces and little cartoon hearts here just goes down wrong with me. --Orange Mike | Talk18:47, 31 May 2013 (UTC)
To be fair I don't think actual pictures of hearts would achieve the same effect ;p. On scholarly value, there have been several studies showing a statistically significant relationship between positive feedback for editors and more edits...which has implications for the scholarly value of the wiki as a whole. Would you like me to send over the links? Okeyes (WMF) (talk) 18:52, 31 May 2013 (UTC)
Ya gotta have heart, Okeyes. I am just a cynic, and wonder what kind of editors the sparkles and shinies attract. (As a feminist, I will admit to a certain self-suspicion: since I am not now nor have I ever been a pre-teen girl, perhaps I am not the kind of person these things are supposed to attract and retain?) --Orange Mike | Talk19:25, 31 May 2013 (UTC)
As a fellow feminist Well, sort of. I consider myself a terrible feminist just so I keep an eye on if I'm doing okay the symbolism of a heart for positive, affectionate actions or statements is pretty universal and not limited to pre-teen girls ;p. Okeyes (WMF) (talk) 19:34, 31 May 2013 (UTC)
Ok, I just had to. It was too good of an idea not to.
I frequently kill conversations, by posting long heavily-edited synopsese, and I'm never sure if everyone is nodding quietly with me and simply not replying for fear of restarting the argument that I managed to quell, or, if they're politely ignoring my foot-in-mouth. Eg. (No-one's replied in 3 days. That's almost definitely a good thing. But knowing for sure would help me stop thinking about it!) The exact same goes for my perspective on other editor's contributions, particularly in talkpage threads - I usually don't want to start a "I agree. ~~~~" pile-on.
Along with that worry, I was contemplating the systems that slashdot/reddit/facebook/google+/etc have, for "upvoting", and pondering whether we were going to attempt something like that with Flow, and mulling over all the ways it could be implemented (anonymous #s versus listed-names, etc), and all the ways it could be misused/abused.
This feature seems like it might answer my contemplations perfectly. No longer will people hum and haw over whether or not to leave you (me) a talkpage message, or send me (you) a private email, or even reply-in-line, perhaps just giving a +1 template (which seems to be a recent trend, that has both positive and negative potential). All-in-all, this seems like a damn good idea. The only potential downside is the possibility of fewer editors giving in-thread encouragement/support; but hopefully this will end up just encouraging us to give private thanks with more frequency, and thereby feel bolder about giving public support when it seems like that would be helpful to a thread. –Quiddity (talk) 01:03, 29 May 2013 (UTC)
Looks like a nice way... wha- wait! Talk page edits? Edits from eons ago? I don't know about you but the only ideal circumstances I would plan on using this would be for a helpful article edit that appears on my watchlist. I'm also not too excited about getting "thanks" for edits that I can barely remember making half a year ago. Surely there will be some time limit before a certain edit becomes "un-thankable" by Echo means? FallingGravity (talk) 01:44, 29 May 2013 (UTC)
I can easily imagine that people in a tense dispute on a policy page or in dispute resolution would appreciate hearing a quick "thanks" for a well-written comment.
I don't expect many people to bother with saying thanks for an old edit, but if it continues to be helpful, then why not? (I believe that it gives you a link to the diff, if you've forgotten what it's about.) WhatamIdoing (talk) 03:59, 29 May 2013 (UTC)
Great initiative! I try to thank someone for their edit at least once a day, and I'm looking forward to trying this out. Lova Falktalk13:35, 29 May 2013 (UTC)
I've, not surprisingly, not been thanked for a post yet. I don't see thanks listed on the notification tab on preferences. I don't want thanks. Will we be able to decide to turn this off? --Onorem (talk) 15:14, 29 May 2013 (UTC)
Sorry if I wasn't obvious enough, though I don't expect many later. Will we be able to decide to turn this off? --Onorem (talk) 15:28, 29 May 2013 (UTC)
Sorry O, but I seem to get shit for not giving feedback prior to implementation, and I still get shit for giving it prior to imprementation. Please tell me at what exact point I'm supposed to give feedback. Otherwise, I'm just going to talk about things as I notice them. --Onorem (talk) 15:30, 29 May 2013 (UTC)
I'm confused, but: yes, you will be able to turn it off once it is deployed, hence the statement "any user who does not want to thank others or be thanked will be able to disable this notification in their preferences". The preferences links provided for in the FAQ don't exist yet because the software isn't deployed. Once it is deployed, they will, and you can enable/disable them. Okeyes (WMF) (talk) 16:00, 29 May 2013 (UTC)
Onorem, speaking as a fellow volunteer, feedback about proposed changes is likely to be more valuable if the feedback had something to do with the actual proposal. For example, given that Fabrice says that it's not been installed yet, your comment that you haven't been thanked and that you don't see it running probably sounds a lot like "I don't know what I'm talking about" to most of the editors reading it. Feedback on a wild guess about what the plans might be is not nearly as valuable to anyone as feedback based on the actual, published plans. WhatamIdoing (talk) 18:21, 29 May 2013 (UTC)
Thanks for the super helpful followup. Obviously, I had missed something...but without your input well after the fact, I'd have been lost. --Onorem (talk) 23:51, 29 May 2013 (UTC)
Orange bar, prominent talk-page notifications
Latest comment: 11 years ago475 comments224 people in discussion
Does the OBOD have to go?
Well, does it? Some users (myself included) have gotten used to the reassuring orange-y rectangle that has graced our screens for what feels like a small eternity. It'd be a shame to say goodbye just because Oliver and his clan decided to shake things up a bit. ;) All kidding aside, I wonder if it would be possible to incorporate some sort of "big notification for certain events" checkbox in the Echo preferences (so, theoretically I could check that box next to "talk page message" and then get a faux-OBOD whenever those events occurred...like the checkbox currently works for email). Is this at all possible, if only for us reluctantees? Theopolisme(talk)23:06, 30 April 2013 (UTC)
I think it's probably worth considering :). I did have some reservations about removing the OBOD - for every other type of activity Echo surfaces, it's an improvement. For talkpage messages (which are dang important) it's not. I imagine if there are substantial concerns we can talk through different ways of resolving them - this could be one of them! Okeyes (WMF) (talk) 23:09, 30 April 2013 (UTC)
The issue is simply that now, talk page messages "matter" in the UI the exact same as...getting a new article you created linked to from another one? Just something to consider, especially from the new editor standpoint. Theopolisme(talk)23:11, 30 April 2013 (UTC)
Theopolisme you bring up a critical point here. I would like to introduce myself since I'm relatively new. I designed the interface for Echo. The lack of scannability, filters & persistence for the Talk page messages is one we will need to address like Oliver mentioned. Some of the options we considered were 1. A separate badge for Talk messages, even though this will eventually be handled by another project around discussions called Flow. 2. An interim filter/ toggle within the flyout and the archive page that lets the user toggle between All notifications and Talk page Notifications only. Vibhabamba (talk) 23:29, 30 April 2013 (UTC)
might it have been a good idea to actually testy some of these things prior to release, or at least discuss them with the community? DGG ( talk ) 23:27, 1 May 2013 (UTC)
I'd like to request that the orange bar stay, or at least that we have the option to choose it for talk-page messages, rather than the new notifications. The bar is much more noticeable. I also think it's problematic to remove the bar from new users; the little numbers next to the name are easily missed, and people don't always add an email address to their preferences. SlimVirgin(talk)02:25, 1 May 2013 (UTC)
The truth is we need a way to differentiate unimportant notices (new Signpost issue), important notices (edit reverted), and vital notices (ANI mentions). The OBOD can't tell the first of these from the last – and Echo can't tell any of them apart. -- Ypnypn (talk) 03:29, 1 May 2013 (UTC)
Another Support for retention of the orange bar. To be honest, I was puzzled at first about references to the orange bar - I didn't recall the message having any background colour, and I had to check on another Wiki. To me, that shows how effective it was, that the message is more memorable than the colour. An optimist on the run! 09:14, 1 May 2013 (UTC)
No, I'm spartacus! err, wait, wrong thing... I also Support an option for retaining the OBOD. — -dainomite09:19, 1 May 2013 (UTC)
Very Strong support for retaining OBOD. It's hard enough to get new editors to look at their talk pages as it is. If all they get is a little number at the top (or wherever it goes on Vector), they aren't going to know what it's for. OBOD is in the face obvious. As for me, I missed four messages before I noticed it and worked out what it was for. I can see it as useful for the 'other page' notifications - which are useful and a step in the right direction. (I don't always complain...) I see it as a giant step backwards in terms of communicating with newbies (and experienced editors who are busy) for talk page messages. Peridon (talk) 11:39, 1 May 2013 (UTC)
Strong support for retaining OBOD (which, on my user page was always purple)
The miniscule teeny tiny red pipe notification is only as workable as an editor's vision. Visually challenged editors will not be able to see it due to size, and/or that combined with color. Some human beings cannot see variations of the red spectrum.
Too easy to get in the habit of ignoring it. How long before any editor develops a habit of ignoring the new red pipe and eventually forgets to even notice it's there?
Finally, an answer to why mine was purple while everyone else saw orange. However, since you pointed that out, I switched to monoskin just to see how the notification "blip" looks. No improvement. Different location on the page, but it's still just something that could be easily missed. Quite frankly, even if it was fantastic button on monobook, complete with flashing lights and accompanying music worthy of an approaching emperor, it still wouldn't be enticing enough to make me leave modern skin.— Maile (talk) 16:58, 1 May 2013 (UTC)
Strong support for retention I've already missed two messages today because the new interface doesn't catch my attention in the same way; if I hadn't seen them on my watchlist I still wouldn't know they were there. Admins are expected to respond promptly to questions; how are we meant to do that if we don't know they've been asked? Yunshui雲水13:02, 1 May 2013 (UTC)
To be honest, one of the things I like best about Modern is that the OBOD was a smooth, calm purple. Not jarring or angry, but relaxed. Compared to that, I find this big honking red button in a different font from the rest of my p-personal menu exceptionally attention-getting. ~ Amory(u • t • c)15:24, 1 May 2013 (UTC)
Strongest possible support for retaining it, and in order to make sure of reaching new users, it should be on by default. Iwouldn't have any objection to a preferences option for not listing it on one's own pages (though I would certainly leave mine on). As an admin working with problematic articles by new users, I frequently need to communicate in a unmistakable fashion with a new or relatively new user about a problem, and I know by experience they will not normally check their talk page routinely, the way most experienced users do. I want to know for certain that if they log in again, or edit again, they will infallibly see my notice. I was appalled yesterday to find this had stopped--I assumed whatever would be done would be in addition to existing features. Naïve of me, to be sure--I should have learned from the past that those who develop things here rarely think of the implications. This is not their fault--it is impossible for an individual to do this, but what is not impossible is to notify clearly, loudly, and unmistakably just what features are being removed before it happens. I hope it will not prove also impossible to get the developers to listen to what seems to be the unanimous feedback here. DGG ( talk ) 18:23, 1 May 2013 (UTC)
Strong support for retaining OBOD I do alot of repetitive tasks, including CSD, and the OBOD helps me to respond to talkpage posts right away, when I would otherwise go on with what I'm doing without checking anything for long periods. INeverCry19:07, 1 May 2013 (UTC)
Save the OBOD: I was going to start this thread, but someone else already has. While I like the "mentions" aspect of, well, mentions, the OBOD is much more visible. It's what we all know and love. Can't we keep it and enable it as we wish? The new notifications are too small, not as eye catching, and categorize too many things in one little notification feature. While I support notifications, I would love to use the OBOD for talk page notifications...perhaps more! MJ94 (talk) 20:42, 1 May 2013 (UTC)
I think OBOD needs to be turned back on, at least until ECHO can supports diff links. I also had not thought about the impact this is going to have on new users until I read this thread. New users need a MUCH more prominent notification than a silly little red square. SpinningSpark23:13, 1 May 2013 (UTC)
Support returning OBOD (at least as an option) I appreciate the color of the notifications square, but it just doesn't catch my middle-aged eye as well. Thanks for your efforts and all the best, Miniapolis02:29, 2 May 2013 (UTC)
Support A little red notification in the top right corner isn't the first thing that I look for when checking Wikipedia for a new notification. The OBOD was front and center and easier to see. Devin(talk)
Support for the return of the Orange Bar "Of Death" (love that). That is the one thing I don't like about Echo. Everything else is wonderful, but it's so much more fun to see the big orange bar than the little red square tell me I have a talk page message. Not only that, but the big, bright orange bar kept others, especially new users prone to make accidental mistakes and get notified about them, the ability to understand not only that they got a talk page message but that they made a mistake. Now they might miss the tiny red box. Also, a user who continued doing the wrong thing after being warned couldn't say they didn't see a big bright orange bar telling them they had a talk page message, but they can easily say such a thing about a tiny red box now. In summation: OBOD was great; but Echo is still awesome. Greengreengreenred03:09, 5 May 2013 (UTC)
Support returning OBOD with an option to let experienced users disable it by preference. The little red blot is not sufficiently obvious to new users that they have a potentially important message. This interface change should not have been sprung on us without discussion! —teb728tc20:50, 5 May 2013 (UTC)
Support the return of the OBOD The Orange Box of Doom/Death/Despair/Destiny/Dread/Dreariness must be put back. The first time I logged in after notifications were put into motion, I was working for almost 10 minutes before I first saw the notifications from the new system. Thekillerpenguin(talk)05:39, 8 May 2013 (UTC)
Are you nuts?
Seriously, are you nuts? I've only just noticed that Echo kills the Orange Bar for talkpage notices. Until now I was under the impression that Echo, being still very much under development, would supplement existing systems, including the Orange Bar. Did no-one think
yes, this is a major interface change which we should really ask people about.
hm, asking people is too much trouble. Let's just make sure we don't get rid of existing key features.
hm, let's at least give people LOTS OF CLEAR WARNING.
Can we have a show of hands of people who actually like the orange bar? I mean, sure, we're going to have that same, small group of people who hate and despise every interface change and who are probably still bitching about monobook, but seriously, the orange bar? The most horrible interface mod ever created? Please, just let it die. Viriditas (talk) 06:14, 1 May 2013 (UTC)
People always seem to complain there wasn't notice even when there was, if they personally were surprised by a change. To the credit of Oliver and others who built this feature, there were several days of watchlist notices, Village Pump technical posts, and multiple Signpost mentions (1, 2) of this as a major new feature under development. That list only covers about what was mentioned on-wiki here, and doesn't include the multiple IRC office hours devoted to the feature, monthly progress reports about Foundation engineering delivered via blog, wiki pages, and mailing lists, the public specifications on MediaWiki.org which have been available for months, or the public Editor Engagement mailing list which gives weekly updates about every software development effort on this team. Steven Walling (WMF) • talk06:23, 1 May 2013 (UTC)
That doesn't count because you didn't give us a chance to bite your head off for spamming by leaving talk page notices.</sarcasm> RyanVesey06:27, 1 May 2013 (UTC)
no Steven, you don't get to treat me like an ignorant moron because WMF failed to adequately document and advertise a major interface change. Echo/Notifications coming? Yes, lots of notice of that. And I've even used it a bit on mediawiki.org. A couple of days ago, when it was due to go live on 25 April, I read most of the documentation. And still until it was mentioned and confirmed on this page I was totally clueless that the Orange Bar was going to disappear. Is the Orange Bar perfect? No, but it's long-established, and you're replacing it with something which in important respects is materially worse (for now), because it makes talkpage messages much less prominent, and certainly very different. In short, your response merely proves that you really don't get the communication problem here. Rd232talk07:07, 1 May 2013 (UTC)
Yes, and that's on www.mediawiki.org, and I don't think everyone goes there on a regular basis, do they? I certainly don't usually. Double sharp (talk) 03:53, 11 May 2013 (UTC)
Shows hands Regardless of whether or not who likes or dislikes the orange bar (You certainly dislike it, and I certainly like it) we must think of the utility it has. Its fairly obvious many users are very comfortable with having the OBOD, which replaced by a small red box is not a good thing. Whats more important is that there are many new users and IPs who would pay attention to think a lot more if there was a flashing orange bar across the page. We've already got plenty of problems with users not seeing their talk page. If the OBOD goes, all they see is a red link listing dozens of reverts and one tiny warning to block, which would be missed very easily. TheOriginalSoni (talk) 08:02, 1 May 2013 (UTC)
Personally I am a big fan of the OBOD as it is a large bright notification that cannot be easily missed. and should not have been removed. Werieth (talk) 12:27, 1 May 2013 (UTC)
Holding hand up I don't like getting OBOD, but I find it useful (except for Signpost's arrival and bots signing things for people). I think it essential for new editors. Getting them to listen to what we're saying is bad enough - with this tiny thing they're not even going to know that we are telling them something. How can we claim to have warned them about something if they don't even know they've got a talk page? (Does happen...) If it's being kept for IPs, why can't it be applied to editors with less than an agreed number of weeks, or edits, with a helpful little bot posting on their talk page to tell them to start looking for little red things instead when they've been here that long? (With an orange banner to tell them of the bot message, of course...) Peridon (talk) 12:59, 1 May 2013 (UTC)
Another hand shown. Many new users do not understand about talk pages, (and some even manage to miss the orange bar!) They certainly will not notice the small red blob on the top line. The right solution would be to bring back the orange bar, ON by default, with an option for experienced users to turn it OFF if they choose (which I, for one, would not do.) JohnCD (talk) 13:05, 1 May 2013 (UTC)
Hand raise - there are several threads on this page requesting the same thing. Is it possible to amalgamate them? Perhaps an RfC could be raised. (Sorry, I'm rather busy at present or I'd do it myself.) An optimist on the run! 13:13, 1 May 2013 (UTC)
Comment I think 'bring back' OBOD is a very understandable, but somewhat oversimplified response to this development. I would say: Please make talk page notifications more 'noticeable'. This is per the common concept that direct user messages should be more distinct than environment messages. If this cannot be speedily done, consider bringing back the Orange bar for a while in addition to the notification indicator. —TheDJ (talk • contribs) 13:20, 1 May 2013 (UTC)
Agreed, the Orange Bar isn't irreplaceable (though for something so long established, replacement will always take a bit of getting used to). There's two clear issues: (a) direct user messages need to have due prominence (and not be piled in with much less important event messages) and (b) a transition to a new interface needs to be done as smoothly as possible. Both aspects have been badly mishandled here. Rd232talk14:28, 1 May 2013 (UTC)
Hand raise. I like it for myself, and as mentioned below, it's essential for IPs. Make it default with opt-out, since there's no point in forcing registered users to have it if they don't want it. Nyttend (talk) 14:24, 1 May 2013 (UTC)
Hand raise—separate from the IP issue, which is getting fixed (?) below, I personally like the orange bar, and it would be good to have something more noticeable. Ignatzmice•talk14:36, 1 May 2013 (UTC)
comment as Ignatz and Rd232, JohnCD and others working with new editors have said, it's the new users for whom this is essential. Sure , there are other ways of doing it, but the OBOD worked. If someone wants to propose something equally effective, let them propose it--and I mean propose it, not implement it site-wide and then see if there are objections. (this is not the first time a transition to a new interface element has been done wrong, and if we are going to have an rfc, it should be about how the users can recapture control of the public-facing parts of the project. It is possible to do better than this--some changes have gone well, but it's clear we need either some formal way or a major informal cultural change that will make it clear we expect to be consulted, and to have things adequate trialed. ) DGG ( talk ) 23:34, 1 May 2013 (UTC)
Sounds great in theory, but in practice it doesn't work. You can't design features by consensus, and those that do end up failing. The notifications system is a vast improvement over what we have and it is the way forward. For various reasons, the community is intransigently conservative in their approach, and they are neither open to new ideas nor welcoming to new features. If they allow the site to remain static and unchanging, it will eventually be passed like BlackBerry in the slow lane. An RfC isn't the solution, since we don't need consensus—we need design expertise and experience and leadership. There is a reason that all great art is at first attacked and despised and then, years later, acclaimed and celebrated. Unless you are open to new things (and for various physiological reasons, most people over the age of 30 are not) and are willing to give them a chance, you probably shouldn't have a say in any new technology that the site implements. I'm sorry if that sounds harsh or insensitive, but it's based on past experience. I mean, look what happens when you let Homer Simpson design a car. Do you tell your dentist how to do his job, or your doctor or your attorney? We expect professional designers to do their job, and to listen to our feedback, but I do not accept a design by consensus rationale as that will always fail. When it comes to feature design, we will always need people who are willing to take risks, experiment with new ideas, and to incorporate feedback from the community, and that's the best we can expect. Can you imagine if your medical procedures or your water treatment facility was determined by the consensus of the unskilled public? Viriditas (talk) 01:31, 2 May 2013 (UTC)
I dont tell the dentist to do his job, but neither does he do anything to make me unable to do my job properly. We dont dislike change- As a matter of fact, most of us love Echo, but hate only how OBOD was handled. What we do dislike is the way it was handled - OBOD was replaced without any proper warning or a possibilty of retention for those who want it, and its replacement fails miserably in fitting into its shows. Echo is disastrously inefficient in showing diffs, which are a must for most of us who would not want to scroll through pages of history. It also fails to be noticeable enough for anyone, and as Yunshui pointed out, comes in the way of us giving messages for us the priority it deserves. Whats worst about it is the way it was handled after these issues were pointed out - We still are yet to hear anything cohesive from the WMF except sarcasm and are nowhere near getting an "interim" solution while these issues are apparently beind "sorted out". TheOriginalSoni (talk) 04:15, 2 May 2013 (UTC)
Yes, this. Echo looks great, from what I've seen it do so far—but the orange bar (that wasn't really "doom"ful) worked much better for talk-page changes. Ignatzmice•talk04:22, 2 May 2013 (UTC)
Viriditas, leaving it to the professionals requires that they have some understanding of the function of what they are re-designing. For communicating with new users, replacement of the conspicuous orange bar by a small red blob is like finding that the doctor has amputated your foot because it had an ingrowing toenail and he didn't think it looked nice: "Oh, you used to use it for walking on? No, you can't have it back, feet are not the way forward, but we'll see whether we can think of some kind of prosthesis... " JohnCD (talk) 10:04, 2 May 2013 (UTC)
Veriditas, I do not go in the dental chair, say "Do whatever you please," and open my mouth. And if it is not right, I certainly let her know and expect her to work on fixing it until I am satisfied. What she can reasonably expect from me is to limit my wants to the grounds of what is possible. I certainly expect her not to take risks with unproven procedures, or experiment on me without my consent.
And on WP, I do not go around saying that people under thirty are most of them inherently reckless. Some are, and the ones who are, shouldn't be programming the site. Any over-thirty people who aren't willing to try new things aren't here on WP in the first place. Why do you think I joined 7 years ago? I was even willing to join Citizendium at the same time--not all new things work out. DGG ( talk ) 14:54, 2 May 2013 (UTC)
Hellooo-oo - much beautiful as the orange bar was, it worked negatively on my associations as an aversion banner. It got to the point where even if I saw that colour appearing anywhere else, an involuntary spasm would wrench my gut. No wonder it was called the orange bar of doom. In fact, it helps some to know it had that title. OBOD indeed. Now what to do about the spinning beach ball of death (BBOD)! Manytexts (talk) 07:10, 19 May 2013 (UTC)
Hey all; sorry for the problems with this - here's an update on what we're doing:
On the IP issue, this is being actively discussed and patched - hopefully it'll be fixed with some sense of urgency.
On the prominence of the OBOD, I've just started a thread on ways to make the talkpage note more prominent. The issue of reducing prominence for talkpage notices was something I brought up, and so we have done some mental prep for making a design change. What I'd suggest, however, is that the full orange bar is likely to be a substantial clash and change (it basically means taking talkpage messages out of Echo - which would totally work for this problem, but I'm convinced there's a better fix we can all think of). What I'd really appreciate is this: if anyone has any ideas for ways to increase prominence for talkpage messages other than the OBOD, please provide them here; I'll throw them at the Product team and hopefully we can find a solution that works for everyone. If none of the ideas any of us think of pan out, we can talk about the OBOD. Is that okay? Thank you, everyone, for your restraint thus far; I appreciate it must be very frustrating. Okeyes (WMF) (talk) 14:47, 1 May 2013 (UTC)
My idea is to give me a big orange bar that is extremely noticeable so that there is no question about whether or not someone who is actively making edits might have missed it...and so that I won't happen to miss a notification if I'm not specifically looking for it between each edit. Thanks. --Onorem♠Dil15:03, 1 May 2013 (UTC)
Why does it have to be "other than the OBOD"? Fine. Give me a Red Bar of Death...everyone's happy now. --Onorem♠Dil15:12, 1 May 2013 (UTC)
Because the suggestion "bring back the OBOD" is already on the table, which means you're not really adding to the conversation by providing it (ditto "Red Bar of Death", albeit for different reasons). Okeyes (WMF) (talk) 15:17, 1 May 2013 (UTC)
"If none of the ideas any of us think of pan out, we can talk about the OBOD. Is that okay?" - That is not on the table. That is on the backburner. I'm a big fan of adding options. I don't understand why you'd take them away. --Onorem♠Dil15:20, 1 May 2013 (UTC)
Also, I do think I'm adding to the conversation by adding my agreement that removing the bar was a bad idea. I want it back. Saying so adds to the conversation. If only one person says it, it's simply going to be ignored. It needs to be said more than once. --Onorem♠Dil15:22, 1 May 2013 (UTC)
As a user and a developer, I would say that anything that gives people the tendency to nickname something the OBoD is something to get rid of ASAP. Seriously, if you want an OBoD, code a nice little JS to turn your own notifications into a nice personal OBoD, but including an OBoD should not be a future design strategy. As I stated before, I'm not opposed to bringing it back temporarily in order to fix the short term visibility issue, but that thing needs to go in the long term. —TheDJ (talk • contribs) 15:37, 1 May 2013 (UTC)
Until today, I'd never seen anyone ever call it that. I shouldn't have to code for myself something that already existed and was removed for no reason. --Onorem♠Dil15:38, 1 May 2013 (UTC)
Sorry DJ, but that's a pathetic argument. Just because a few people give it a derogatory nickname doesn't mean the rest of us don't like it. Make it an optional extra - opt in, or opt out, I don't care either way. An optimist on the run! 15:41, 1 May 2013 (UTC)
It's an opinion. As far as I'm concerned, Wikipedia is already an option maze and we shouldn't add more to it, just to cater to a few users. We should fix the problems and don't get stuck in old ideas simply because they are what we are used to. How many websites do you know to throw an orange bar in your face when someone is talking to you, other than MediaWiki ? It's an outdated concept from 2003 when we simply couldn't do any better. Echo isn't better YET, but it should be better and as soon as Echo is better, we need to get rid of that Orange bar, without a 'i wan't to be in 2003'-option. —TheDJ (talk • contribs) 15:57, 1 May 2013 (UTC)
Out of interest, what websites are you comparing WP with? I can't off-hand think of any that have the same need for users to promptly read messages to them. Rd232talk16:01, 1 May 2013 (UTC)
I'd consider my talk page to be the equivalent of a private message system on a forum or the like. And in the forums I tend to visit, the notice that I have new messages is always obvious and in my face. Either a big red bar (Fark) or a pop-up message. I find that on sites that prefer subtle notifications (such as Twitter), I can visit many times over several days and never see that I have a direct message. Even on the Facebook notification format that obviously inspired this, I rarely notice the new messages on the website. I notice that I have new messages because it pops up on my phone. Resolute01:21, 2 May 2013 (UTC)
I would seriously suggest taking talkpage notifications out of Echo immediately. That fixes the problem with both IPs and "where's my Orange Bar" for both occasional users and regulars, and reintegration in a way that everyone is happy with can then be discussed and implemented at leisure, on a separate page (this talkpage covering all issues is not a good place for detailed discussion of alternative designs). Rd232talk15:14, 1 May 2013 (UTC)
What about coloring the entire "menu bar" area ([username] [Talk] ... [Log out]) when there's an active notice, rather than relying on a small box? So if I had a new message, the background of that entire area of my screen would go red, or orange, or whatever. I think for many of us, the top right of the screen isn't an area our eyes spend much time on when we're using WP, so we may need a more significant kick in the pants to notice things there than we would elsewhere. Another idea I heard on IRC (I think from user:Nick?): make the fly-out automatic for [logged-out users|newbies|non-confirmed accounts], rather than expecting them to notice and click to see what's going on. A big ol' box opening up at the top of your screen is much more noticeable than a little red speck. A fluffernutter is a sandwich! (talk) 15:20, 1 May 2013 (UTC)
Both reasonable :). I suspect our designers may not like prominently colouring such a large area, but I'll see what we can do. Okeyes (WMF) (talk) 15:31, 1 May 2013 (UTC)
Also reasonable, although I'd worry that if we create the option incoming users will play around with it - it's not just 'legacy' users we're hitting. Still worth bringing up, though :). Really what we need to do is increase the prominence of Echo, at least for talkpage messages, otherwise we've built something inferior to what it replaced. Okeyes (WMF) (talk) 15:33, 1 May 2013 (UTC)
It is my understanding that the OBOD is a function of the software, and there are thousands of wiki's affected. Echo should not have removed OBOD, that was a major error, and any removal of the OBOD should be a configuration issue, and not something that individual users can configure, but that when installing a wiki, the site developers implement or not per preferences for that wiki, and in this case for all of the Wikimedia wiki's. It should never be editor configurable. Apteva (talk) 15:07, 11 May 2013 (UTC)
Hey, all, I've created a cookie-based user script at User:Writ Keeper/Scripts/orangeBar.js to try and replicate some of the OBOD functionality. Obviously it's not as good as the real thing, but it's not totally awful. Let me know if there are any bugs y'all find. (Obviously, it requires cookies to be enabled in your browser.) Cheers! Writ Keeper⚇♔15:47, 1 May 2013 (UTC)
Comment One of the things I hate about the technical side of WP is when a developer says something like "Seriously, if you want an OBoD, code a nice little JS". Oh yeh? Some people can do that - and they're probably doing it. Don't get me wrong - I used to program in COBOL but I changed my line of work and became a Apple Mac user instead of a PC coder. I've never had to learn JS or the current 'in' things. I would think that the number of regular editors who can just turn out a little JS thing in a couple of minutes is rather small compared to those who can't. Peridon (talk) 16:19, 1 May 2013 (UTC)
Well the counterpoint to that is that it only takes one person to make the script; once they have, anyone of any level of technical ability can install and use it. (Of course, there's the question of how do you let people know that you've made the script; I know that spamming my orange bar script in different messages across multiple threads as I did made me feel uncomfortable.) But it's not a very good idea for the official developer to rely on JS widgets for important UI components like this; I know that my implementation of the orange bar is much less optimal than the normal one, and I hope to get rid of it when the orange bar is re-enabled. Writ Keeper⚇♔16:28, 1 May 2013 (UTC)
I think my personal preference would be to take talk page notifications out of Echo and maintain the orange message bar. An alternative would be to create a special flyout for talk page messages that automatically flies out. It couldn't be the current flyout which is far too intrusive. RyanVesey17:13, 1 May 2013 (UTC)
Question Can someone explain to me why talk p. messages cannot at least temporarily invoke the OBOD as well as the new system? I'm not a programmer, but it seems intuitively unreasonable that both could be triggered. (for talk p. messages--not the other notifications--I don't think anyone is asking for that). DGG ( talk ) 23:38, 1 May 2013 (UTC)
Why can't we make a permanent flyout in the new notifications system that spans the width of the entire article and is, albeit optionally, orange or another prominent color? I think that this would probably alleviate everyone's concerns as briefly summarized by DGG, especially if this span was not as easily dismissible as normal notifications as suggested by Ryan. — RandomDSdevel (talk) 01:09, 2 May 2013 (UTC)
Ok. Really? I just got a notification that Cluebat III archived my freakin' talkpage. It's a bot marking the archiving process as minor. That means Echo shouldn't notifying me.—cyberpowerChatOffline10:20, 5 May 2013 (UTC)
Can you explain why re-adding the OBOD (as an opt-out) would require taking talk page notifications out of Echo? Why can't it do both, for those that want it to? --Floquenbeam (talk) 15:04, 1 May 2013 (UTC)
I mean it in the sense that, if Echo becomes opt-in then (once we've settled the interaction issues around toggling between them) Echo has essentially stopped dealing with talkpage messages. It can do both - but it seems clear that if we make Echo opt-in for this feature, it will probably remain forever opt-in :/. Okeyes (WMF) (talk) 15:08, 1 May 2013 (UTC)
Well I have no problems at all with making Echo opt-out instead of opt-in; in fact, I don't even care if there is no opt out. Someone could probably write a javascript thingy if they really really hate the little square. I just want to make the OBOD opt-out. So everyone gets Echo, and everyone also gets the OBOD, unless they opt out. It sounds like you're saying if you opt in to the OBOD, you have to opt out of Echo. --Floquenbeam (talk) 15:14, 1 May 2013 (UTC)
Well, opt-out of Echo for talkpage messages, and then we have this conversation all over again in N months when we disable the opt. I'd note Rd's suggestion a few sections above to remove talkpage messages from Echo until we've got a proper solution; I'm going to propose that to the team, because I think it's totally worthwhile. In terms of what the solution is, any ideas for 'things that would increase prominence but are not OBOD-y' would be most appreciated :). Okeyes (WMF) (talk) 15:20, 1 May 2013 (UTC)
This appears to be a meatball:ForestFire; I'm not sure where to comment anymore, so I'll keep doing it here I guess. Rd's idea for now seems a good one, but I just want a clear answer to this: is it technically impossible to have talk page notifications included both in Echo and in something like the OBOD, or is it just what those in charge of this new interface want? --Floquenbeam (talk) 15:24, 1 May 2013 (UTC)
More the latter, I think - but I'd add the caveat that I am most certainly not a dev. I write in R and Python. MediaWiki makes my eyes hurt. Okeyes (WMF) (talk) 15:28, 1 May 2013 (UTC)
It's not. Give me an annoyingly obvious visual notification that there is a message. You can format it however you'd like, but I don't want a tiny notification that can easily be missed. Seeing as the orange bar has worked for years, that seems like a good option. I don't want the new option to be less noticeable in any way. The 'only' plus I see to this is that the people who still think the fake new messages bar is funny will have to finally come up with a new joke. Could you explain to me why the orange bar is not acceptable? --Onorem♠Dil15:29, 1 May 2013 (UTC)
Okay, that's more helpful; I agree that increasing prominence is paramount. The orange bar itself takes up almost too much room, and doesn't fit within the wider workflow for notifications, which are the issues. Having said that, I am in total agreement that the notifications workflow needs to do a lot better at making talkpage messages prominent - otherwise we've effectively replaced something with a more inferior alternative. Okeyes (WMF) (talk) 15:34, 1 May 2013 (UTC)
Is there any way to have the orange bar along with the new system? After all, I think the bar was more noticeable than the new method (although I can't say for sure as I have yet to get any new notifications under the present system). AutomaticStrikeout (T • C • Sign AAPT) 15:27, 1 May 2013 (UTC)
We're discussing ways to resolve it - see #Update above. It may or may not be possible to have both, but I'd suggest what we should focus on is "ways to make it more prominent that are not the orange bar", noting that we always have the orange bar as a backup if we can't think of any. Okeyes (WMF) (talk) 15:28, 1 May 2013 (UTC)
Comment: There are parallels here with the move to the Vector skin, aren't there? This is a bit like switching everyone from Monobook to Vector, instead of introducing Vector as an additional option and letting people stick with Monobook if they want. It ought to be possible for Echo to do an Orange Bar for talkpage notifications just as the current system does (but integrated into Echo to ensure it all links up with notifications going to zero, etc). If it's one of a number of design options given to users, then everyone can be accommodated, and in particular newer designs can be the default. Rd232talk15:30, 1 May 2013 (UTC)
I agree there's a parallel - what we seem to disagree on is whether the way the Vector rollout was handled is a Good Thing. Now, personally, I use monobook everywhere; I can't stand Vector. I find that it hides editor-centric features that are, to me, very important, in favour of providing a reader-centric view. But a bit of me feels like the switch meant we avoided having difficult conversations about Vector, and it suffered as a result. I don't want the same to happen to Echo, for example. Okeyes (WMF) (talk) 16:23, 1 May 2013 (UTC)
Hey, all, I've created a cookie-based user script at User:Writ Keeper/Scripts/orangeBar.js to try and replicate some of the OBOD functionality. Obviously it's not as good as the real thing, but it's not totally awful. Let me know if there are any bugs y'all find. (Obviously, it requires cookies to be enabled in your browser.) Cheers! Writ Keeper⚇♔15:39, 1 May 2013 (UTC)
Thank you; perhaps we should hire you as a developer in place of the current ones. The situation with Vector is a great example of what we should have done here — we changed the default without disabling the old version. Why can't we do the same here? Nyttend (talk) 17:22, 1 May 2013 (UTC)
Keeping the old skins makes every new feature for MediaWiki take twice as long to develop. There's a reason other websites don't keep old features - it builds up technical debt until it is impossible to deploy any new feature to the site, which is basically what's been happening to MediaWiki over the past several years. Kaldari (talk) 18:22, 1 May 2013 (UTC)
There's a difference between at some point abandoning support for some little-used old skins after a couple of years of something better being well-established and used by most, and suddenly switching it for everyone, even before it's had time to get bedded in. Seriously: if old skins are an issue, then consider abandoning support for them (begin by taking them out of preferences, so you can only access them some special way, like putting a line in your common.js). Doing it gradually and with plenty of notice and clearly better alternatives is OK, whereas doing it suddenly, without clear notice and not clearly better alternative is not. Rd232talk20:34, 1 May 2013 (UTC)
For reference, I wouldn't be so blasé about this. Removing skins with <1% usage has taken months to work out logistically. Monobook usage is so high that it doesn't even enter into that conversation. The way the Vector rollout was handled meant that it was never improved to the point where Monobook could be replaced for the vast majority of our power users. Supporting Monobook as a requirement is a technical debt that we'll never likely get rid of. -- tychay (tchay@wikimedia) (talk) 21:52, 1 May 2013 (UTC)
It seems the main issue in the "bring back OBOD" discussion is the prominence of echo notifications doesn't approach OBOD prominence, and maybe special prominence when it contains a talk page notification. Would be nice if that were addressed right in Echo to make it useful instead of turning Echo into the "notification stream of crap I don't care about" which would certainly happen if we made it optional. (BTW, statistically OBOD only has a 50% clickthrough for IP users, so while visually ugly, it isn't that prominent) -- tychay (tchay@wikimedia) (talk) 21:52, 1 May 2013 (UTC)
Very interesting - suggesting that messages, at least for IPs, need to be more prominent, not much less. And if WMF was willing to take talkpage notifications out of Echo temporarily whilst a new design is worked out in collaboration with the community, that would be useful data for discussion. As it is, it seems we must humbly await whatever it is that will be bestowed upon us. Rd232talk15:36, 2 May 2013 (UTC)
Is there a better explanation of how they collected their data? I like having the new messages bar. I almost never clicked on it directly to look at my new messages. I almost always go to the history from my watchlist when it pops up...and just because people don't care enough to click doesn't mean it isn't prominent. --Onorem♠Dil22:03, 1 May 2013 (UTC)
I don't care if the notification is in the form of an orange bar or a little red square designed to look like it was ripped off of Facebook.
What I DO care about is that once again, a tech-or-Interent-related company is changing its interface or something about their product, and for the worse.
The orange bar did two things:
It appeared right after a message was left on my talk page, and not hours later.
It displayed the actual diff of the message when you clicked on it.
The new red thingee does NEITHER.
And yet, I got no notice for this, I was not asked how I felt about this, and yet the "developers" responsible for this fucking idiocy have "no plans to bring back the orange bar or even include as a preference". Wonderful. Now Wikipedia too can join the ranks of Google, Facebook, Apple, and every other techie company that has elevated making arbitrary and harmful changes to their patrons' product, and apathy towards the patrons' needs, to a virtue. Bravo, Wikipedia. Encore. Nightscream (talk) 23:57, 17 May 2013 (UTC)
Do we want the orange bar?
After reading TheDJ's latest comment above, I almost get the feeling that those who want the orange bar are going to be treated like they don't count. He argues that creating an option to allow users to have the orange bar would be catering to a few users and his statement that "we need to get rid of that Orange bar, without a 'i wan't to be in 2003'-option" indicates that the wishes of orange bar proponents are being ignored. It almost sounds to me like the message is that they should be allowed to change things without our input because they know better and they don't want their website to look outdated. However, it seems to me that if the editors want an orange bar, they should get an orange bar. After all, they're the ones for whom the notifications system is being designed. So, the question becomes: do we want to have the option of retaining the orange bar? Because if we do, our wishes should be honored. AutomaticStrikeout (T • C • Sign AAPT) 16:10, 1 May 2013 (UTC) (to clarify: I am not saying that the orange bar has to be the only option. I merely wanted it to be one of the options. AutomaticStrikeout (T • C • Sign AAPT) 20:08, 1 May 2013 (UTC))
Discussion
I don't understand how TheDJ's comment has any implication as to what'll happen to proponents of the orange bar; TheDJ is not associated with the Echo development team any more than you are. Let me be clear as to our perspective on Echo, and on the orange bar:
We need a way of surfacing notifications to the community. At the moment, this mostly doesn't happen, so we've built echo to solve for that. One of the areas where notifications do happen, however, is in relation to talkpage messages; those are currently surfaced very prominently via the orange bar. Conceptually, it makes a lot of sense to build talkpage messages into Echo - having notifications split over lots of different avenues is what's happening now, and if we don't solve for that we haven't built very useful software - but I have repeatedly noted to the product team that at the moment, our solution is actually worse than the existing features when it comes to talkpage messages, even if it's better for everything else.
So the solution is: find a better way, a more prominent way, of surfacing talkpage messages. This could certainly be the orange bar: that's an option available to us. But, riddle me this - is anyone who wants the orange bar actually wanting it because its attribute of "being an orange bar" is paramount to how they work, or because its prominence - how apparent it makes talkpage messages - is paramount to how they'd work? I strongly suspect it's the latter, and comments both below and above ("yes, or something equally as visible", for example) seem to go in the same direction.
Finding a way to make this work is of paramount importance to us; it's worth noting that with most of the developers (and all of the designers) living on PST, they've only just woken up. Once they've got into the office and had a chance to talk/think about this, I'd hope they'd have at least mock-ups for a solution later today. Okeyes (WMF) (talk) 16:21, 1 May 2013 (UTC)
That's no problem. I'm used to it. As much as as Wikipedians are used to be blamed for any piece of bad content that is on Wikipedia. People like to think in monoliths. —TheDJ (talk • contribs) 17:59, 1 May 2013 (UTC)
Oliver, on your second bullet point -- from where I sit, it is absolutely about it being an orange bar. I think we are disconnecting here: you seem to be focused on whether or not the new system is an improvement. But my concern isn't about whether we are getting to a better place, but how we get there.
Removing the orange bar (as it has functioned for some time) without warning is a problem because there are tens of thousands of people in a learning process with Wikipedia, whose learning process is being disrupted by changing the feature without warning or explanation.
I don't think the orange bar has to live forever, but it does have to be preserved while a thoughtful transition is executed. -Pete (talk) 17:07, 1 May 2013 (UTC)
I agree that socialisation was not handled particularly well; the E2 should certainly have done a lot more work socialising the more prominent changes, and also addressing the rather obvious problem with switching from such a large notification to such a small one, even with informed (albeit probably begrudging) consent from the community. I don't think we're in disagreement on the second point; "it is familiar" is a different attribute from "it is orange" :). Okeyes (WMF) (talk) 17:11, 1 May 2013 (UTC)
I was more worried about engineering risks, like disruption of service because we were replacing talk page notificatiosn and people might have them dropped or some other degradation in Quality of Service. I still feel that had that happenned, it'd have been far worse than this ;-) - tychay (tchay@wikimedia) (talk) 22:00, 1 May 2013 (UTC)
I do hope that people would give Notifications some time. A lot of this reaction seems to me to be a knee-jerk revulsion. For me, I'm getting used to it and enjoying the lack of the highly intrusive big orange bar. I detested it with a vengeance. It's the visual equivalent to a loud noise and makes me want to close the browser each time my screen lights up, and so I bid it good riddance. The current popup about talk page messages (that just appeared in the last 12 hours) is a bit too prominent for my liking and I've disabled it; the little red beacon is quite noticeable enough after a short adjustment period and I now systematically look at it each time I load a new page. However, it seems that a sizeable number of people want ORANGE back, although demands seem to be from people wanting to get their own message across to others rather than other users necessarily wanting it for themselves. Our techies could surely configure it as a preference that you could opt in to. Even better, turn the display into a self-controlled 'editnotice' type message so those who want it could have a bright red screen with "READ YOUR FUCKING MESSAGES, YOU BLIND MORON" huge and bold lettering. -- Ohconfucius ping / poke03:18, 11 May 2013 (UTC)
Yes, or something equally as visible. 'Raising the prominence' doesn't describe enough how much they're willing to change from just having the tiny red box next to my user name. It should at least be an option. --Onorem♠Dil16:13, 1 May 2013 (UTC)
Pretty much everyone who isn't me, on the team, works and lives on Pacific Standard Time. It doesn't describe it enough because the people qualified to make this decision are all asleep :). Okeyes (WMF) (talk) 16:21, 1 May 2013 (UTC)
Now I'm confused ;p. Onorem objected to the fact that we had not described enough how willing we were to change the feature, as it stands now. My reply was that the Echo development team, minus myself and one of our developers, lives in San Francisco. It is currently 9:42am in San Francisco, and before now has been, well, less than that. The team has not yet had a chance to describe how willing we are to change the feature, because the team was not awake. I cannot make judgment calls on product direction on my own. Okeyes (WMF) (talk) 16:43, 1 May 2013 (UTC)
Shouldn't things be rolled out when people are awake to respond to them. I guess I'm not sure exactly what time this launched. My 'raising the prominence' was actually commenting on the statement made by Eloquence. Once again, I still don't understand why you have to take something away to introduce something new. If I opt-out of Echo, will I get my bar back? (I'm assuming that's tied to fixing the unacceptable lack of notification for IP users. Did you really not test this idea with anyone who wasn't logged in?) --Onorem♠Dil16:38, 1 May 2013 (UTC)
Certainly, which is why we launched it at about 1pm PST, giving several hours for problems to become obvious. I note that this issue was raised during that time period - we just (obviously) didn't have time to address it so late in the day. There is, I'd note, currently no opt-out for Echo. Okeyes (WMF) (talk) 16:43, 1 May 2013 (UTC)
Fair enough. I didn't notice it last night. (That's kind of a theme now. /tongueincheek) If I opt-out of every option for notification, I just get nothing? Do you know yet what the notification for IPs is going to look like? --Onorem♠Dil16:48, 1 May 2013 (UTC)
Yes, I'd like to have the orange bar back, "on" by default, with a clear path to turning it off if and when a user is happy with using the new notification system instead. This seems like a no-brainer. Removing a core feature like the orange notification bar shouldn't be done suddenly. -Pete (talk) 16:31, 1 May 2013 (UTC)
Lets unbreak the system and {{trout}} the dev who decided that killing the OBoD was a good idea. The visibility and impact of the OBoD is needed especially for new users and issues on their talk pages. Werieth (talk) 16:33, 1 May 2013 (UTC)
Duhh. We need to question Eloquence's willingness to participate in a consensus-built encyclopedia project, since he's basically said up above "We don't care what you want — you don't get it back". Nyttend (talk) 16:40, 1 May 2013 (UTC)
Definitely. And as a default for new accounts. A bot can tell them how to stop it after they've been here long enough to be trusted to check their notifications. Peridon (talk) 16:58, 1 May 2013 (UTC)
While I have no problems with this system (save one minor issue I'll discuss elsewhere, if I can't find whether someone else has raised it), I think we should allow people to keep the old system in preferences, much as we allow them to keep older skins. Daniel Case (talk) 17:23, 1 May 2013 (UTC)
Yes, or something as least as conspicuous, as the default for IPs and newbies. No objection to an option for experienced users to turn it off. JohnCD (talk) 17:27, 1 May 2013 (UTC)
Yes I consider talk page messages to be very important, and I would like a less subtle notification of them. -- tariqabjotu17:32, 1 May 2013 (UTC)
Yes A prominent notification of a new post on my Talk page is essential because Talk page notification is orders of magnitude more important than all the other stuff. The rest of the "notifications" are irrelevant irritations - my Watchlist is perfectly adequate "notification" for everything else. Roger (Dodger67) (talk) 18:10, 1 May 2013 (UTC)
Yes I much preferred it, find it easier to miss the tiny Facebook-style notifications badge. I wasn't aware of this change until it was rolled out; talk about a solution in search of a problem. --John (talk) 18:19, 1 May 2013 (UTC)
+1 It's hard to overstate my Yes. And please explicitly tell the rest of us the net time you plan a change nearly as big as this. TheOriginalSoni (talk) 18:29, 1 May 2013 (UTC)
+1 I want a bar at the top of my screen every time someone posts on my talk page. I want it big and orange and I want it to instill a feeling of Doom in my heart if I don't click on the link in it to find out why it was there. Thank you. Technical 13 (talk) 19:08, 1 May 2013 (UTC)
Yea, verily. And would that it would stay forever on my screen if I cannot be bothered to click on the link to my talk page, that I may be forever haunted by the messages that I might have missed, by its greatly irritating continuous presence! (That's its main virtue, BTW; that it is so obvious and so annoying that you cannot but check your talk page to drive it away, and hopefully reply. The red square, while annoying once you see it, is pretty much the epitome of unobviousness.) Double sharp (talk) 14:32, 11 May 2013 (UTC)
Yes, or something equally visible, as a turnoff-able option. I have non-standard colour vision, and a tiny little badge in a colour that most(?) people may see as a strong visual cue but I don't, isn't a functional replacement for a bright persistent bar that won't go away until I visit my talk page. cheers, Struway2 (talk) 19:25, 1 May 2013 (UTC)
+1Yes, I agree with TheOriginalSoni, that it's hard to overstate how important it is to give some notice to the editors next time a change nearly as big as this is planned. Please bring back the Orange bar somehow. - tucoxn\talk20:17, 1 May 2013 (UTC)
Definitely. I like it; lots of other people like it; give us warning next time. (As a side note, I don't know if the IP-fix has also restored the bar for non-confirmed users; if not, that has to happen ASAP as well.) Ignatzmice•talk20:53, 1 May 2013 (UTC)
Yes It's hard enough to get new users to read their messages - they certainly will not see this one. Hey ho - more blocking will be the order of the day. Ronhjones (Talk)21:02, 1 May 2013 (UTC)
+1 To me, I support thus wanting to have it badly. I choose orange because actually red and orange are sensitive to your brain; it can actually contribute caution (or urgent) to your brain. 𝕁𝕠𝕣𝕕𝕒𝕟𝕂𝕪𝕤𝕖𝕣22 (talk) 21:56, 1 May 2013 (UTC)
+1 The orange bar works. It works for me, it works for new users, it works for pretty much everybody. If you have something that works, then don't break it. Jheald (talk) 22:36, 1 May 2013 (UTC)
As much as I dislike the Facebookification of Wikipedia and want the Orange bar back, I do need the developers for other feature requests, so this is more of a feature request then a complaint. MBisanztalk22:37, 1 May 2013 (UTC)
If he thinks we're all going to post little photos to go with our posts, I'm afraid he's barking. Up the wrong tree, that is. Talk pages aren't putting new editors off. It's things like trying to find where the hell you are in an edit window so you can actually do something. Soimething like Facebook and Twitter is OK for Facebook and Twitter. Why can't we stick to the encyclopaedia way and leave social media dumbing down (of finished results - not the edit window trouble) to social media? Peridon (talk) 16:45, 2 May 2013 (UTC)
Yes and, considering the impact on communicating with new users, I think, unlkie MBisanz, that this is the highest priority thing for the developers to do (unless the wehile site goes down, or something of that magnitude). Considering the extraordinarily great preponderance of yeses, the time to do it is before tomorrow. Even if it means reverting the entire change, it should be done, and then the good other features added back selectively DGG ( talk ) 23:43, 1 May 2013 (UTC)
Yes. Reposting what I wrote elsewhere: With the orange bar, we knew it was a new user talk-page post, and a diff was offered to the last change, so we could get there with one click. Now, we see the little number has changed, but don't know why. One click tells us it's the user talk page. A second click takes us to the page. We then have to scroll down to find the new post or posts. The orange bar is particularly important for new users, who might otherwise not notice that people are speaking to them. SlimVirgin(talk)23:54, 1 May 2013 (UTC)
Absolutely. The Echo interface would have been a good supplement to the very obvious OBOD and giving the users the ability to use one, the other, or both should have been a no-brainer. Personally, I want the obvious notification. Resolute00:42, 2 May 2013 (UTC)
Yes, at least until diffs are available in Echo AND yes, until new users/IPs get a more prominent notification BY DEFAULT AND yes, until regular users can turn on the prominent notification as an option. SpinningSpark00:46, 2 May 2013 (UTC)
+1Yes - Now and forever. The OB should be an opt-out option, not removed by fiat. BRING IT BACK! (...please?) Manning (talk) 01:35, 2 May 2013 (UTC)
Sort of. I dislike the orange bar but would prefer to see a much more prominent talk-page notification system than the one rolled out with Echo. I'm sure I could learn to live with squinting at my monitor each time I log in, but newbies failing to notice they have new messages is a huge concern, and I think this needs to be fixed asap. (Other than that, so far I think Echo is a great idea.) Rivertorch (talk) 05:51, 2 May 2013 (UTC)
Make it default, and make it mandatory for IPs. Most of the time I drop a message on a user's talk page, I'm trying to get his attention. Lots of times, its the last step before I block him. That's why I'm leaving a message. I don't want to have to wonder whether a newbie is ignoring messages or just doesn't realize he's got one. —Kww(talk) 06:15, 2 May 2013 (UTC)
Linguistically, there is more emphasis on the individual +1s and thus would illustrate a more deeply held belief than +5. — Crisco 1492 (talk) 09:50, 2 May 2013 (UTC)
Emphatic yes. I don't care if it's orange, vermillion, puce or heliotrope; I don't mind whether it's a bar, a banner, a flyout, a popup or an animated .gif of bunny rabbits frolicking in the snow; I don't give two hoots if it says "You have new messages", "Oi! Talkpage! Now!" or just "Guflurgle": just give me something that emphatically lets me know I've got a damn message. Yunshui雲水07:18, 2 May 2013 (UTC)
Yes, make the orange bar default for new users and IPs. I prefer the orange bar but it is okay if an option is included to turn it off for autoconfrimed users. --Hdt83Talk08:05, 2 May 2013 (UTC)
Definitely it's nuts to just remove the big oragne bar and replace it with a subtle tiny red box in the corner of the screen. The Orange Bar should be in addition to the red box. No need to remove that, just restore the perfectly good functioning Orange Bar please! GedUK11:45, 2 May 2013 (UTC)
Yes I am utterly disgusted with the removal of the orange bar without, AFAICT, any well publicized discussion. This proclivity of developers to foist some fundamental change on the community, done as a fait accompli, is a terrible problem. We should not wake up to one of the best features of the interface removed and then have to claw it back. That is not the way it should ever work, and this is iteration number... I've lost count.--Fuhghettaboutit (talk) 11:58, 2 May 2013 (UTC)
Yes The orange bar should definitely be the default option. If someone doesn't like it, let them opt out of it, but the orange bar should be the default option. Sideways713 (talk) 14:52, 2 May 2013 (UTC)
Comment: I'm now more comfortable with the little red rectangle, but when it was implemented there was suddenly this unlabeled little red button and I'm there wondering "if I push this will I have to pay someone to install a little blue button to get me back down?" The little red rectangle isn't so bad once I figure out what it's doing and why it's different but the shock of the sudden change made me dislike it. I still like the big orange bar, but I could, eventually, get used to the little red rectangle. RJFJR (talk) 19:43, 6 May 2013 (UTC)
Strong yes as a default option with an opt-out option. It can be hard enough to get new editors to start communicating and not being sure they actually saw the message, doesn't help. Lova Falktalk15:50, 2 May 2013 (UTC)
Yes The notifications doesn't work properly. had 2 messages on my talk, but notifications showed only 1 as unread. RedtigerxyzTalk18:23, 2 May 2013 (UTC)
Yes (I only clicked the little grey 0 to find what the hell it was...) I don't get it very much, but I like the orange bar and find it really noticable when it does appear - even during an editing session. And I really don't want an email - I almost never use that mail account. This notifications thing seems to me an easy way to miss that someone has left me a message - it is not in an area of the page I normally look at and if the notification number stays the same colour as the 0 I really doubt I'll ever notice it for weeks. Astronaut (talk) 19:15, 2 May 2013 (UTC)
It should be the default option at least until the developers come up with some sort of improvement to Echo's talk page notification that has been tested and accepted by the community. I'm finding it virtually impossible to deal with anons and new users at the moment because I have no idea whether they have any reasonable chance of realising that they have messages. And I'm finding it awkward to spot the things myself. Like it or not, user talk pages are right at the hub of the Wikipedia community and anything that makes them less efficient to use is retrograde. Comestic/appearance come a distant second to practical/functionality. Furthermore, all this twaddle about "surfacing" and "socialising" etc is making my brain hurt. Any chance of people speaking in plain English instead of elitist management-speak? And is there any chance that the development team will stop imposing stuff from above with little or no consultation? I am among the apparent many who were aware of Echo but thought that the bar would be retained, and now that it has gone we are presented with a fait accompli in the FAQ, along the lines of "there are no plans to reinstate the orange bar". Are developers somehow exempt from consensus? If so, should they be? - Sitush (talk) 20:05, 2 May 2013 (UTC)
I agree, the "there are no plans" and "yes we hear you, give it a few days and all will be well" are seriously irking. Is there any reason we can't have the orange bar back NOW, until the devs/WMF/whoever get something similar in place?Ignatzmice•talk20:36, 2 May 2013 (UTC)
Yes - It should be optional for those users who want to retain the previous setup. In fact, I see absolutely no reason why this shouldn't be an option - if you don't want it, don't opt to use it. I would object to this being forced on all users, however. Lukeno94(tell Luke off here)20:16, 2 May 2013 (UTC)
Yes - while I like the new system, it is not prominent enough and makes it easy to miss important messages. At the very least, the OBOD should be an opt-in preference. Also, new users who make unwanted edits may not notice it and will therefore not receive their warnings. This could lead to an increase in blocks. Oddbodz (talk) 20:25, 2 May 2013 (UTC)
Yes The orange bar should be the default for IP editors and newbies, and available as an option for others. It is the difference between "Excuse me, sir, you're making a mess of the encyclopedia" and "YO! STOP WHAT YOU'RE DOING BECAUSE YOU'RE FUCKING UP!" — Malik ShabazzTalk/Stalk20:46, 2 May 2013 (UTC)
Yes - default to having the bar with opt-out being an option in preferences for those that don't want it. I would like to know if my talk-page was edited as soon as it happens. Additionally, the email I began to receive with the header You have a new talkpage message is not as useful as the previous Wikipedia page User talk:Valley2city has been changed by _____ and I would like that reverted as well. Valley2city‽20:53, 2 May 2013 (UTC)
Yes - My heart rate does increase in fear when I see it, but in some cases that is the desired effect. How else do you notify other users (esp. newbies) to stop their actions immediately? I would vote that OMOD should be an opt-out. Another option would be to include a checkbox "Notify the user" similar to "This is a minor edit" and "Watch this page" when editing in user talk page. YLSS (talk) 21:38, 2 May 2013 (UTC)
The notification that you have a new message needs to be very obvious and hard to ignore, especially for new editors. If someone leaves a warning for a new editor, they need to know that the new editor is going to notice it. A tiny icon at the top of the page changing from grey to dark red isn't adequate. We don't necessarily need an orange bar, but we do need to have something equally obvious. Furthermore I don't think a response of "we are considering options for addressing this" is adequate: this is a serious issue with widespread implications that should have been addressed before the feature was rolled out, not at some undefined point in the future. Hut 8.522:01, 2 May 2013 (UTC)
Yes, exactly. "Don't worry your pretty little heads about it, we're sure you'll love this next one!" is not okay. Howzabout we get the orange bar back NOW, and then whatever new plan can supplant it once they've got it. Gggrrrrrgrr. Ignatzmice•talk22:07, 2 May 2013 (UTC)
Yes. By default the orange notification that "you have new message" is very useful, especially for new editors. --Dede2008 (talk) 00:31, 3 May 2013 (UTC)
Most emphatically yes - this new "tiny obscure thingie seemingly designed to be missed up there in the corner" is no substitute; and it looks gimmicky, like it was ganked from Google+. --Orange Mike | Talk01:06, 3 May 2013 (UTC)
Yes. I want the return of the orange banner and I will continue to happily use my watchlist and contributions list to keep up with what I care to be informed about. Fylbecatuloustalk02:20, 3 May 2013 (UTC)
Do want. I went ten, maybe twenty pages before I noticed the weenie red box tucked away in the top corner of my page. The orange box made a statement. It let you know authoritatively that someone had conversed with you. You took notice of the orange box. You respected it. And now it's gone. At least make it the default with option to opt-out for non-IPs. Deadbeef02:58, 3 May 2013 (UTC)
Yes, by default. Personally I prefer using Echo because it looks more elegant and also has more features like notifying when someone has tagged you in a post, but it's way too easy to miss. Ideally, we would have an orange "You have new messages" bar which also advertises the fact that you can switch, e.g. "Don't like this annoying banner? Click here to change your preferences." -- King of♥♦♣ ♠ 03:10, 3 May 2013 (UTC)
Yes Need an option to turn this on, preferably on by default. I don;t get much activity on my talk page. I want something to grab my attention when a new message is there. Some of the "no" comments say its intrusive. I agree, but I want it instrusive so it is hard to miss. Let folks who don't want it opt out. Don't take it away from folks who find it useful. I also agree with the comments about IP messaging. RudolfRed (talk) 04:05, 3 May 2013 (UTC)
Yes, it should be an option that can be allowed to be turned on or off by the user in their preferences or in their Notifications preferences. — Cirt (talk) 04:50, 3 May 2013 (UTC)
It must remain the default. In response to Okeyes' riddle, it works as well as it does because a lot of people despise the look of it; I think an aesthetically pleasing solution, however prominent, would be less effective. I'm comfortable with enabling people to opt out, and would consider it myself. But for me the crux of the matter is that someone who has received a talkpage message and continues editing should be treated as though they have seen the big orange bar. —WFC— FL wishlist06:07, 3 May 2013 (UTC)
If I'm doing something wrong I'd like to know immediately so I can stop. The orange bar does just that. I always stop what I'm doing to check what it is. All the other notifications can wait. I have my watchpage which I can check myself as and when I need to. I have set up the new notification system to give me a daily summary (as a starting point - I may move that to weekly). I do not want to wait for a day or a week to be told that I have just accidently deleted someone's user page. The orange bar is important - and talkpage notices should not be confused in with other notifications. At the very least there should be an option to keep the orange bar, and if the official developers don't understand that, then someone who does understand will hopefully write up a script quickly, and let everyone on this page know. SilkTork✔Tea time06:41, 3 May 2013 (UTC)
Per Hut 8.5. The tiny notification is hardly noticeable (even when you know it's there – like several commenters above, I also didn't notice my first notifications until after a few minutes) and that is not sufficient in a context where you need to get a particular editor's attention to a message as quickly as possible (e.g. warnings). Chamal's sockSockMaster•C06:52, 3 May 2013 (UTC)
I trained a class of new users yesterday. As an experiment, I left each of them a talk page message while they were editing, but didn't tell them. Not one of them noticed, during the following four hours; in previous classes, people have noticed the orange bar immediately. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits10:01, 3 May 2013 (UTC)
Personally I'd turn off the orange bar in favour of a less-prominent notification, but it or something equally prominent must be the default for new editors; it was hard enough getting them to find their talk pages as it was. Espresso Addict (talk) 10:05, 3 May 2013 (UTC)
New users in particular are much less likely to notice the tiny red box, and then get blocked for failing to respond to warnings. That isn't fair. The orange bar is annoying because it needs to be. Andrew Lenahan - Starblind12:25, 3 May 2013 (UTC)
On occasion good faith new users have ended up having to be blocked because they blissfully violate content policies while not noticingthe big orange bar. Logically it's only going to get worse with a tiny red square. The argument that it's 2013 and we're using a 2003 interface doesn't address the important fact that we also have a "2003" contribution policy: the "modern" sites require registration and email verification before allowing folks to post, so it's relatively easy to contact them if the violate a site's Terms of Service (TOS). Since we don't require that, we need an "in your face" attention getting mechanism by default. Once users figure out that they have a talk page, then opting out of the orange bar is a cool feature. NE Ent12:30, 3 May 2013 (UTC)
Yes The current implementation is not nearly attention grabbing enough. I would prefer something as or more prominent than the orange bar was. Preferably enormous and obnoxiously colored. This is especially important for new users. After all, it took me half an hour to notice the new Notifications tool when I logged in earlier; there seems to be little hope that a new editor would even realize it exists. Sophus Bie(talk)12:43, 3 May 2013 (UTC)
Furthermore, due to my preference for browsing with javascript disabled, my notifications consist of a small number, with no red. I didn't expect the flyout to work, and I'm okay with that, but the lack of any highlighting whatsoever is an issue. Sophus Bie(talk)14:46, 3 May 2013 (UTC)
Yes; removal of it makes redundant the entire series of IP warnings and makes it very unlikely that they're ever going to find out about their warnings. We need something appallingly obnoxious; the orange bar is not good enough.--Launchballer12:49, 3 May 2013 (UTC)
The "orange bar is not good enough"? You are in fact voting to bring the orange bar back, at least as an option right? --GoneIn60 (talk) 13:12, 3 May 2013 (UTC)
I'm saying we need a bar. The current one isn't flashy enough - something like this would be much better.--Launchballer19:42, 5 May 2013 (UTC)
Strongest possible yes - I may be an old fuddyduddy, but the OBOD does its job and I see no valid reason to get rid of it as an option.--ukexpat (talk) 13:02, 3 May 2013 (UTC)
Yes – I share a lot of the sentiment above. New users and IP's benefit from having a blatantly obvious notification streaming across the top of the page when someone has posted on their talk page. The new Echo notification has its advantages, but the problem, evidenced by so many 'Yes' replies thus far, is that the new format appears to branch off in an entirely new direction when it comes to talk page notification. It's not an upgrade from the previous format, but instead is an entirely new approach. A technical person in nature has a tendency to attribute the negative feedback as "resistance to change" before taking the time to fully understand the position. I think that's all we're asking for at this point: time. Let's embrace the desire to improve Wikipedia and its tools, but let's not ignore ideas and principles established throughout the process. Jumping to a different tangent as if we're starting from ground zero can project the notion of reckless abandonment. --GoneIn60 (talk) 13:06, 3 May 2013 (UTC)
Yes. This shouldn't even be a discussion--this is an editor retention issue. A new user will NOT see the number change from one to zero. Assume this is a good faith new user who doesn't understand one of our hundreds of very important but also very non-intuitive rules, and a veteran editor is trying to tell them about this rule. Well, they're never see the message. And then the veteran editor(s) have no choice but to escalate to warnings, and eventually to a block. Now this new user is blocked after having received, from their position, absolutely no warning. Who wants to hang around a website like that? To be honest, I'd love it if non-confirmed users got an orange bar that filled the entire page. We should be increasing the prominence of this message, not decreasing it. Qwyrxian (talk) 14:29, 3 May 2013 (UTC)
I totally agree the Product team needs to come up with a reasonable solution, both for the sake of newcomers and experienced editors. But research showed last year (or the year before, iirc?) that actually, the big orange bar only has a 50 percent clickthrough rate. It's big, and obnoxious, and it occupies the space that (on every other site) is an advert inviting people to punch the monkey and win a prize. It's actually not that effective, because users screen it out. What we need is for them to quickly develop a replacement - something that is noticeably not just more prominent but also better at getting users to pay attention. The current system, quite reasonably, I'd argue, is not being interpreted as doing that (and we don't have the data to show differently). But the OBOD isn't very good at it either. Okeyes (WMF) (talk) 15:23, 3 May 2013 (UTC)
Do you really mean "clickthrough"? Does this mean "people who click on the hyperlink contained within the orange bar? How do you account for users, such as me, who go direct to their talk page when the bar appears rather than clicking on it? - Sitush (talk) 16:33, 3 May 2013 (UTC)
I've never ever, not even once I think done a click through. The bar is there to announce a message on the user page and that's where I go to find the message. So that data might be meaningless. fwiw. Truthkeeper (talk) 16:41, 3 May 2013 (UTC)
You say the orange bar is "obnoxious", as though that was a problem. It's meant to be obnoxious, for pity's sake, in order to get people's attention. If the design team are redesigning it to be less obnoxious, they are totally missing the point of its purpose, like changing the red traffic-light to a soothing pale lavender-pink because it's prettier and red might alarm drivers. I don't think a 50% click-through is surprising: I quite often go straight to the talk page, and not infrequently finish something else first, but I certainly take notice, exactly because it's obnoxious and persistent, and I greatly doubt that anyone "screens it out". For data on the current system, see #76 above, where Andy Mabbett reports that not one of a class of newbies noticed the new notification in four hours. JohnCD (talk) 16:48, 3 May 2013 (UTC)
If clickthrough is going to be the measure of effectiveness, let's have the clickthrough figures for RSOL (red square of life) to compare to OBOD. There is every chance that they are a damn sight worse. Frankly, I think 50% is pretty good in my opinion. I too frequently do not click through, especially when I am working through my watchlist and can see who the message is from and how urgent it is from the edit summary. By the way, is RSOL marking the messages as read after the talk page is visisted without clicking through? SpinningSpark06:03, 4 May 2013 (UTC)
Yes. New editors and IPs need to have this by default. They aren't going to notice the number. I have a hard time noticing the number. ~Adjwilley (talk)15:32, 3 May 2013 (UTC)
I should probably clarify here...I love the new notification system. Keep it just the way it is; just add the orange bar back as a default notification for talk page posts. (I don't oppose an option to turn it off.) ~Adjwilley (talk)20:09, 3 May 2013 (UTC)
I'd really prefer it if we had the option to revert back to the old orange bar for new talk page messages while using the new notification system for the other things (mentions, etc.). Also, per Adjwilley, we really ought to have the orange bar as the default for talk page messages, since new editors aren't likely to notice it (as well as anonymous editors w/ problematic edits requiring talk page interaction. The orange bar is much more noticeable and, to be quite honest, I like it better. Call me old and out of date if you must, but I miss the orange bar already. Tyrol5[Talk]16:13, 3 May 2013 (UTC)
Yes - the orange bar is hard to ignore, but not visible when working (writing) in the edit window, easy to remove when the message is retrieved (not now possible w/ the red thing staring in the face), and generally should be kept. As a side issue, having an orange bar notifiying talk page messages is benign, the current notifications in my view encourages edit warring. Truthkeeper (talk) 19:02, 3 May 2013 (UTC)
Yes - I have missed several talkpage notifications. The force of talkpage notifications has been negated. Everything else is accessible through my watchlist. This has been presented as a fait accompli and there was insufficient notification and opportunity for discussion. This is not simple conservatism.--SabreBD (talk)19:28, 3 May 2013 (UTC)
Yes - At least as an option as the number can be easily missed - I agree that the orange bar ought to be on by default for this reason. CT Cooper ·talk20:56, 3 May 2013 (UTC)
Yes Nobody asked me if I wanted this notification rubbish. Give me the orange bar back please, and please allow me to turn the notification thing off altogether. Parrotof Doom22:15, 3 May 2013 (UTC)
I've been known to overlook "tiny little number" notifications on internet forums for a year or two. The orange bar is a whole lot harder to miss. --Carnildo (talk) 22:49, 3 May 2013 (UTC)
+42 - I love the orange bar. I always noticed it. It literally took me two hours to notice that I had received a message on my talk page because of the little red box. Some people may not like it, but there should be an opt-out option in Preferences. öBrambleberryofRiverClan01:57, 4 May 2013 (UTC)
Yes. The tiny little notification number is very easy to miss if you're just having a quick look on WP - especially when you've got used to the big orange banner. Sure, the old banner was intrusive - but that's because it's supposed to be intrusive. Nothing worse than suddenly discovering you've had an urgent request for admin assistance sitting on your talk page for three days... Grutness...wha?05:56, 4 May 2013 (UTC)
Yes There's no reason I can see why we shouldn't have the OBOD and Echo both running simultaneously. I like Echo a lot, but I also think that the OBOD should be standard for all users to make new messages clear and readily apparent. As many people have noted above, the whole point of a message notification is to be as intrusive as possible, and while the new features Echo introduces are really cool, the loss of the "LOOK AT ME! YOU HAVE A NEW MESSAGE. READ IT YOU FOOL" effect is a bad thing. --Jayron3206:06, 4 May 2013 (UTC)
Definite, strong, eternity Yes The best is to extinguish Echo totally once and for all, and restore the OBOD. BTW, just placing
Yes this is needed for new people here, the symbol next to the name is too subtle to notice. The orange bar is obvious when it appears. We can have it opt out. Graeme Bartlett (talk) 10:52, 4 May 2013 (UTC)
Yes Besides the comments made about getting the attention of newcomers, I find the tiny red box doesn't get my attention sufficiently. I'm not absolutely wedded to the orange bar per se, but we know it works, so why change it? Mangoe (talk) 13:20, 4 May 2013 (UTC)
Yes. I need to know immediately if I screw up, and I won't notice a little red +1 thingy. the bar, or something equally conspicuous is needed. Personally, I'd favor having a click through pocture of dante's hell, with the meme You just screwed up, inspect the damage on your talk page, but that's just me. Tazerdadog (talk) 18:05, 4 May 2013 (UTC)
Yes. The orange bar is needed to call to people's attention that they have a message. If someone is a new user and editing inappropriately, and they get a message to warn them of their errors, that needs to be brought to their attention overtly, not subtly. Could users be given the option to turn off the orange bar in favor of a more subtle notification? I suppose. But the default needs to be a big conspicuous notice, like an orange bar. (As to the statistic that only 50% of users click through the orange bar -- how many of the non-clickers were editors who ever came back to Wikipedia at all? We have lots of "hit and run" editors who make one or two edits to the encyclopedia and never come back.) --Metropolitan90(talk)18:26, 4 May 2013 (UTC)
Yes. I read through the above and I can't find if anybody mentioned this but not only is the new notification system too small but it completely ignores the fact that some people are color blind and will not see the red in the dot. Now some people won't have seen the orange either but they will have noticed the huge bar across the top of their page that is not there unless you have a new message. MarnetteD | Talk21:57, 4 May 2013 (UTC)
Yes - First get the orange bar back, then look at possible options to make it more prominent through echo. After that make the orange bar a preference. Eloquence blunt refusal is not the right way to do this. Garion96(talk)22:52, 4 May 2013 (UTC)
Yes - The FAQ for notifications flatly states "We're not planning to bring the OBOD back nor to make it available as a preference". So it seems they don't give a damn whether we want it or not. A central part of developing software use to be listening and responding to the people the software is being written for. Personally I find it most unpleasant trying to remember to squint at this coy little "notification" hiding away somewhere at the top of my screen. A notification should notify, otherwise its not a notification. --Epipelagic (talk) 00:19, 5 May 2013 (UTC)
Yes. Looking at the "No" side, the main argument seems to be "It's new!"... But change as such is a bad thing (for example, it results in confusion) and should be offset with something good - yet little evidence of that is provided. Second argument is "The orange bar interrupts my work!". But "that's a feature, not a bug". When I get a message, I am supposed to stop whatever I am doing and deal with the message first. For all I know, it is going to tell me that I am causing trouble and should stop. Third argument strangely seems to be "It is not noticed by the users." - I wonder how that is supposed to work with the second argument... Finally, the argument "It was ugly!"... Well, perhaps in such case the Foundation could hire a good artist who would make it more beautiful instead? So, it doesn't look like there was a good reason for a change... Of course, it doesn't mean that the world is going to end if the change stays (as a Lithuanian proverb says, "Šuo ir kariamas pripranta." - "A dog also adapts to being hanged."), but it would seem to be better to avoid it... --Martynas Patasius (talk) 01:03, 5 May 2013 (UTC)
Yes. I'd love having the orange bar itself back (because I'm a stickler for sameness), but I'd settle for an as-prominent notification. - Purplewowies (talk) 01:59, 5 May 2013 (UTC)
Yes. It should be available as a preference for established users — I would turn it on, because I find it too easy to miss new messages with the new system. And it should be the default for IPs and new users, because it should not be easy for them to ignore warnings. —David Eppstein (talk) 03:30, 5 May 2013 (UTC)
Yes - the new system doesn't give enough emphasis to new messages, users should have the option to use a system that does. --W.D.Graham09:34, 5 May 2013 (UTC)
Yes, definitely, for all of the reasons given above. Notifications of new talk page messages are far less visible without the orange bar. Prioryman (talk) 10:40, 5 May 2013 (UTC)
Yes! - While the new notifications system works like a charm, the orange bar is still a good way to know someone has left a message on your talkpage. However, I think that it should be a preference: by default, it should be on, but it can be disabled. Also, should it be brought back, something needs to be changed so that edits by SineBot don't count as a new message. Narutolovehinata5tccsdnew15:06, 5 May 2013 (UTC)
Yes - I usually don't comment on these technical proposals as I generally don't care about the changes and/or can't provide a long, detailed reason for my opinion on a feature. But to me, keeping the orange bar (or at least the availability to opt-in) is just common sense for me. The only notifications I ever really get are additions to my user talk page as I am not an admin or a bot user. But the only way I would have noticed a change to my user talk page was the fact that I saw a change in my watchlist. Luckily, this time it was just the dopey disambiguation bot, but in the future, it could be a user asking an important question. I hope the admins reconsider this awful decision to "We're not planning to bring the OBOD back nor to make it available as a preference." —Mr. Matté (Talk/Contrib) 18:17, 5 May 2013 (UTC) – I don't think they are operating as admins here. I think they are paid for behaving this way. --Epipelagic (talk) 21:50, 5 May 2013 (UTC)
Absolutely. I don't mind getting new features that may or may not turn out to be helpful. I don't even mind this much if, as seems to be the case, they impact some of our processes (see #Sockpuppet investigations). With the exception of obnoxious howlers such as the "wikilove" stuff, I don't even mind getting useless new features. And I don't care about the precise colour of the bar or whether it appears at the top or on the side. But replacing it by something inconspicuous was a remarkably stupid idea. I want it back. Those who are currently in charge of design have proved their incompetence; I don't trust them to come up with something equally functional. Hans Adler19:01, 5 May 2013 (UTC)
Yes As stated above, swapping a UI component without notification (pun intended) is a bad idea. Reinstate it as the default and give users the option to decide which one they want to use. — MrDolomite • Talk03:45, 6 May 2013 (UTC)
Jawohl - The new notification system is far less noticeable to both newbies and established users alike. At least keep the new system for reversions and the like; but keep the OBOD for all new messages. hmssolent\You rang?ship's log03:51, 6 May 2013 (UTC)
Yes. I think Echo is just great. It's working just fine for me, and find it very useful. But I remember what is was like to be a new editor:
I had no idea that there were notability guidelines
I had no idea that there were heaps of other policies and guidelines
I was unaware that that articles had talk pages
I was unaware that that I had a talk page
For maybe a year, when I saw (top) (now (current)) for an page on my watchlist, I thought is meant that I had edited a page by clicking on its title instead of on a section in the page
The orange message bar was the way I found about these things and much more, through messages from other users. I wouldn't have known what in the world a little number next to my username meant, and quite possibly would have ignored it.
Very strongly support bringing back the orange message bar, making it the default for all users, with Echo as an opt-in alternative.
Bring it back! It was one of the greatest and most beautiful thing on Wikipedia! If possible make it an 'opt in', with default set to yes.Vanischenu(alt)12:14, 6 May 2013 (UTC)
Yes, please! - Thus far, I've missed 3 talk page messages because I don't have my pretty blue bar in place (I have a css hack in place to make it blue). It really helps to have it there. ~ MatthewrbowkerMake a comment!18:07, 6 May 2013 (UTC)
Yes - This needs to remain the standard. With the orange bar, I notice talk page messages immediately. With the red dot, I notice them twenty minutes later. --Bongwarrior (talk) 01:38, 7 May 2013 (UTC)
Yes -- Of all of the user interface changes to make. I understand the desire to be hip or like FB or ... whatever the thinking was, but not everyone is thinking that way. Let it be (for now) the default option, that can be turned off, or selected, and then in a couple of months change the default to off UNLESS THE USER HAS ALREADY TURNED IT ON! Friends don't let friends make interface changes like Facebook does. htom (talk) 05:02, 7 May 2013 (UTC)
The orange bar is a relatively new feature and is appropriately very annoying. It is a good feature that should be restored ASAP. The new notifications are fine, but they are more like an automatic watchlist than a notification to check your talk page. Opting out of the orange bar should not be an option. In the interest of Wikilove, the color should be a soft green, though. Apteva (talk) 18:24, 8 May 2013 (UTC)
Apteva, the orange bar has been around for as long as I can remember — I first saw it in 2007! That was back when I first started editing as an anonymous editor. It was only briefly blue for those not using the monobook skin (which was Wikipedia's default view for a very long while). Kurtis(talk)18:25, 11 May 2013 (UTC)
Yes. Same as others above. Have the orange bar by default, with the option for the new notification system in Preferences. SlightSmile01:36, 9 May 2013 (UTC)
Extremely Strong Support. First of all, the orange bar is very noticeable. The small number is not (I didn't even realise my talk page had been updated yesterday for some time until I clicked directly a link to my talk page!!). Also IMHO it is a bad idea to have the orange bar for unregistered users and the notifications for registered ones (I understand it is that way from WP:TP; is it so?). I think pretty much all registered users were once unregistered users. What are they going to think when they register and never see the orange bar again? They've already seen it while as an IP and know its function: they have new messages. Now when they don't see it and must use a completely new system, it is very confusing. Double sharp (talk) 10:20, 9 May 2013 (UTC)
HELL YES Explain to me and everyone else in this section why every time we as a community come to you with excellent on ideas on what to change you pick the things that are fine the way they are and annihilate them. It seems to me that you've all got to some research into product marketing, and you can all start with this: New_coke#Official_launch. TomStar81 (Talk) 06:50, 11 May 2013 (UTC)
I was looking forward to Echo, and it is a very welcome development, however: a) it has not managed to fully replace the orange bar functionalities (no diffs, no detection of whether the user has visited their talk page yet, not prominent enough); b) it is not prominent enough, especially for new and unregistered users; c) simply put, the orange bar works, and echo should be welcome supplement to it, not an unwelcome replacement. I'm unsure why on earth it was thought appropriate not to fully consult the community on such an important matter, and why it was felt that deployment to a top 10 website of a still experimental feature was a good idea, as any user would have known that removing the prominent notifications and not providing diffs is not exactly very useful. SnowolfHow can I help?12:30, 11 May 2013 (UTC)
This makes the system work. Why are we wasting developer time on changing something that actually works when we could get them to do something useful like improve the software for resolving edit conflicts without losing edits and continually biting newbies? ϢereSpielChequers18:39, 11 May 2013 (UTC)
Bring back the Orange Bar: it works. It alerts new and inexperienced users to the fact that they've got a message - or the fact that they've got a talk page - and is still useful for the rest of us. By all means introduce an option for experienced editors to switch it off if they want, but leave it on by default. The only thing I've found irritating is when one editor had an orange bar lookalike on their talk page, so every time I went there I thought I had a message - fortunately he eventually took it away. It wasn't broke, so don't fix it. PamD11:55, 12 May 2013 (UTC)
Bring back the Orange Bar - In complete agreement with Pam. I also notice substantial support above. Jusdafax18:25, 13 May 2013 (UTC)
Should be the default. Fair enough that not everybody wants it, so give them the option to choose a different design - even better make it customisable. The new notifications are good, and not all need to be as prominent as that, but a notification of messages on your talk page need to be as prominent as possible. Facebook is not a model of good UI design. Thryduulf (talk) 20:31, 15 May 2013 (UTC)
YES The orange bar did two things: 1. It appeared right after a message was left on my talk page, and not hours later, and 2. It displayed the actual diff of the message when you clicked on it. The new red thingee does NEITHER. I honestly don't care if it's a long orange bar or small red square ripped off from--er, I mean, that bears a striking resemblance to Facebook notifications. I just want those two capabilities back. And whatever programmer or programmers who made the decision to remove those benefits, without asking the community how they felt about it, really needs to get a clue. Nightscream (talk) 00:03, 18 May 2013 (UTC)
Yes Didn't realise it had gone, but do please bring it back; lovely soothing colour, just like Birds' Butterscotch Instant Whip in the 1970s. Ericoides (talk) 19:00, 18 May 2013 (UTC)
Very strong Yes. It’s the only notification that I’m getting at all without JS, so it’s the only thing that should be there by default. Please give it back to me and please never ever have the idea to put it also onto other projects like it is now. Please read unobtrusive JavaScript before developing anything new at all. It’s just not possible with old browsers and old JS to get these new notifications at all. Then you’re getting none and the communication is out of order this way. No no no no no. I’m not doing anything more here on en-WP without notification bar anymore, except these polls about getting it back. If you ever will develop anything else that also will see everyone without JS and is there by default, then it’ll be ok. But not this way, this has nothing to do with Accessibility, and that should always be the most important thing with core features like notifications for communication. I can live with being not able to add, change or remove interwikis at JS based Wikidata and I can also live with not being able to use the new JS based translation interface, I can also live with some more things based on JS that aren’t there, but without a visible notification bar, I can’t. That will lead to misunderstandings and bad communication, if you don’t notice them anymore. It’s just no good at all this way. Please change that and put it on the top list for developing things, so that everyone will get notifications again and not only the ones which use JS. And please don’t ever develop new core features anymore that aren’t accessible to everyone here. Then it would be better to stop developing new things at all, because then things are getting worse with that and no better. --Geitost10:36, 19 May 2013 (UTC)
Yes. Some older browsers do not show the Orange segment "Talk:You" and so a Notifications of "(1)" could be a user-gossip wikilinking to username, rather than talk-page alert how your major article or template was up for deletion 6 days 23 hours ago! -Wikid77 (talk) 15:55, 21 May 2013 (UTC)
Frankly, Wikipedia has been going the wrong way on its interface lately. The best site to model is the Drudge Report: Fast, predictable, efficient, and works as expected with no scripting at all. Jclemens (talk) 03:47, 23 May 2013 (UTC)
Yes the orange bar when editing normally can be a bit exaggerated, but we need such noise for new editors to get their attention. Most significantly as an old hand the orange bar greatest asset when I infrequently use AWB as it very quickly highlights that someone has contacted me so I stop the process to see what the issue is and ensure its not because I've unintentionally cause some distress, the new notification system doesnt do that.. Gnangarra03:57, 23 May 2013 (UTC)
Yes - strongest possible support to keep the orange bar, sorry, late to the party again, but the people who need the orange bar will never, ever, ever in a month of Sundays read any proposals about changing it ever, and so will just get at best an unpleasant surprise, or at worst not respond to any message you send them ever again. More to the point, people got used to the orange bar, it was familiar, and we all figured out what it meant and what it did. Removing that introduced an element of surprise to the casual user - a big usability fail. Ritchie333(talk)(cont)14:24, 29 May 2013 (UTC)
Restore the orange bar. It's the only known way of getting the best kind of new users (meaning people who are genuinely new to Wikipedia and are trying to make positive contributions) to realize that someone is trying to communicate with them. --Orlady (talk) 16:26, 2 June 2013 (UTC)
No
Based on AS's explanation below that this a "yes" here is intended to be "yes keep the orange bar exactly as it was, forever", I'm going to say No here. Yes, I support a much-more-visible-than-currently notification option than Echo currently has, and yes, the devs need to squash the "IPs get no notification at all" bug post haste, but once that bug is fixed, Echo is a vast improvement over a bar that impedes page functionality. Any such notification should be implemented within Echo (whatever form that notification ends up taking), since it can be (presumably) done there with all the qualities of visibility and intrusiveness that OB lovers appreciate. Why preserve the legacy, more-awkward way? A fluffernutter is a sandwich! (talk) 16:34, 1 May 2013 (UTC)
Does it really impede functionality that much? Looking at your talk page, you got about 1.3 new messages bars daily. Did the orange bar really disrupt your ability to use Wikipedia as an editor or a reader? (1.3 assumes that at no point did you have multiple messages left between checking your talk page.) --Onorem♠Dil16:42, 1 May 2013 (UTC)
I can only speak for myself, but yes, I have often found myself opening a page, putting my mouse or eyes where I thought a link, infobox, etc would be, and instead finding myself clicking on some random block of orange that has appeared. No, it doesn't keep me from functioning and render me gibbering senselessly, but yes, it's a substandard system in my eyes to anything else that doesn't shove page content aside. Move the orange bar up to the menu bar, fine. Turn my page background orange, fine. But why mess with the placement of the content I care about, if there's a better way? That's why I asked below what you guys are so tied to - is it the orange? the rectangle? the pushing-down-page content? Many of these things are fine with me, but not to the point where I think we need to maintain a legacy system doing them when Echo can do them in an integrated manner. A fluffernutter is a sandwich! (talk) 16:49, 1 May 2013 (UTC)
I have never had that happen. I've had problems with fundraiser banners doing that, but I can't say I've ever had an issue with the orange banner disrupting my ability to navigate a page in nearly 7 years. I'm not against a better way. I'm against getting rid of the current way when it isn't being replaced by a better way. Don't break it until you have the fix ready. --Onorem♠Dil16:57, 1 May 2013 (UTC)
I've had problems with the "new" toolbar loading slowly and doing that, and finally just turned off the toolbar, which I never use anyway. But I also have never had the orange bar do that. I don't think the orange bar jumps things down the way the toolbar and fund-raising banner do…does it? It's not Javascript loaded, is it? Maybe this is browser dependent? -Pete (talk) 17:00, 1 May 2013 (UTC)
Lots of us orange-bar-lovers are interested in seeing the bar available as an opt-out thing. Can't speak for others, but I at least see no reason for you to be stuck with it if you don't want it. Why would you object to an optional bar? Nyttend (talk) 17:19, 1 May 2013 (UTC)
I can sympathize with Okeyes's concern expressed above that having it as an option requires the orange bar to still be officially supported by the WMF, creating another thing to maintain in addition to Echo proper. Obviously I don't think that's enough of a concern to override providing an option (if I did, I wouldn't have created a replacement orange bar script that I have to maintain now), but it's not an illegitimate point. Writ Keeper⚇♔17:24, 1 May 2013 (UTC)
(edit conflict) Well, because "optional" legacy features tend to...snowball, as far as user demands. If, two years down the road, the devs are ready to deply, I dunno, let's make something up, watchlists via Echo, or allow people to post to talk pages via Echo, or some tweak to the system that affects talk page notifications, the people who kept the legacy-version orange bar will pipe up: "No, you can't do that! It would break my orange bar!" or "Nooo this feature doesn't work with my orange bar, make it work!" and suddenly we have issues over deploying a new feature, because we have to accommodate legacy-users who refused the better option. This happened with skin deployments, if I recall correctly - oldie users are allowed tocontinue use Monobook as a legacy retained in the prefs, but the devs develop with an eye to Vector, because that's what everyone else uses. And when a new feature rolls out, users of legacy skins invariably raise holy hell and require extra dev (and script writer) time and resources to make things work with their preferred legacy format - because they're using something that's not developed for anymore, but they expect that something to continue working indefinitely. A fluffernutter is a sandwich! (talk) 17:28, 1 May 2013 (UTC)
They need to provide a better option before you can accuse legacy users of refusing a better option. Until then, all you are saying is they shouldn't listen to complaints now because at some point in the future, they might make a change that would lead to complaints. --Onorem♠Dil17:41, 1 May 2013 (UTC)
(darn it Oliver, stop edit conflicting me!) That's a valid point, Onorem. Everything I'm saying here is predicated upon my assumption that the WMF is and will be working on a better "you've got messages" format than the current red speck. I'm confident enough in the team to assume that they're already on that, so I'm comfortable saying that a better option will be available and the orange bar will be the worse option; I suppose YMMV as far as confidence that that's happening, and if you lack confidence that they'll fix this semi-bug, I can understand how you'd feel that the OB is a better option than what you expect to see in future Echo versions. A fluffernutter is a sandwich! (talk) 17:48, 1 May 2013 (UTC)
I do completely believe that. The godawful hours part a bit less. (It's only 9:42 AM? boohoo.) My comment was not a reflection of how I think you have handled this issue this morning. You're putting up with me and that's good enough for me. My comment was meant for fluffernutter's method of dismissing the complaints. --Onorem♠Dil17:51, 1 May 2013 (UTC)
9:42am...PST. I don't live on PST, I work eight hours off it. My work time today lasts until 2am my time, and (theoretically) starts at 5pm. In practise I've been working on this issue since 11am. I would consider a 15 hour workday 'godawful', yes. Okeyes (WMF) (talk) 17:54, 1 May 2013 (UTC)
No, we don't need it, be gone oh 2003. Btw, this seems like a terrible place to hold this vote. Most people end up here giving feedback, which is generally skewed to criticism. —TheDJ (talk • contribs) 18:01, 1 May 2013 (UTC)
Let someone make a gadget. A number of editors who are used to the old format may be disgruntled, but just as with vector or a new change on Facebook, people quickly move on and forget what the old thing looked like. This is a convenient innovation and while some things should be worked out it is a very good implementation. ~ Amory(u • t • c)18:21, 1 May 2013 (UTC)
I don't think a Facebook analogy applies -- the way Facebook approaches UI changes is different from Wikimedia's in too many ways to count, the primary one being a matter of resources. Also, I don't think many of us are concerned about our own notifications or ability to adjust (as experienced editors), but rather the experience of new and learning users. My concerns arise principally out of concern for my students, not my own editing experience. -Pete (talk) 18:27, 1 May 2013 (UTC)
If someone can make a gadget to turn it on, I'll be happy for myself, but as Pete notes, it wouldn't help with getting the newbies' attention. Much better to make a gadget to turn it off. Nyttend (talk) 20:26, 1 May 2013 (UTC)
Apart from shortcomings in the deployment, this is a big step forward. The OBOD made Wikipedia look as if it were designed by sixth-graders. older ≠ wiser18:38, 1 May 2013 (UTC)
OK - it can be redesigned to look hi-tech (it looks fine to me, and I can't see any need for redesign) - just so long as it's there for getting through to new users. Peridon (talk) 19:17, 1 May 2013 (UTC)
Redesigned: We could change the Orange bar to look redesigned by a $multi-billion corporation as showing message-bar: You have Talkoogle. -Wikid7716:17, 21 May 2013 (UTC)
Maybe "as well as", but not "instead of", Echo. Also, humourously, this has just been canvassed on IRC by TheOriginalSoni. — foxj19:54, 1 May 2013 (UTC)
I object. It was not canvassing for two reasons. First I was asking everyone at the IRC, and not aiming it at specific users [Hence it is not canvassing]. Secondly, the only reason I went to the IRC in the first place was "ask a question", not to ask everyone to join the fight. TheOriginalSoni (talk) 20:15, 1 May 2013 (UTC)
Not long after you posted your support here, you piped up on IRC: "Anyone here wants the orange bar back?" This is pretty clear-cut canvassing. — foxj20:49, 1 May 2013 (UTC)
Per TheDJ and Bkonrad. It's about time someone came up with some notification system brings that begins to look and feel like what one would expect on other websites. Some people need to step out of their walled garden and look at the rest of the interwebs. --MuZemike22:18, 1 May 2013 (UTC)
The day the orange bar disappeared was the day Wikipedia finally did something right. Most IPs don't communicate, aren't interested in communicating, and their talk pages go mostly unanswered. I understand and acknowledge that the community needs a way to communicate with them, and I'm sure we will eventually have one, but the orange bar isn't needed for IPs nor is it needed for registered users. The notification system is the way forward, incorporates new features that allow us to work more efficiently and communicate more effectively, and presents us with a more adaptable but non-intrusive, streamlined interface that we need. This unhealthy desire for the "good old days" when the orange bar would scream across the top of the page like a bloodthirsty eagle bearing down on its prey isn't the way forward. This unfounded nostalgia for the stress-inducing orange bar that always interrupted whatever you were doing with the burning glow of a thousand suns never helped anyone and should not be brought back. Kill it, bury it, spit on it. Goodbye and good riddance. Viriditas (talk) 01:41, 2 May 2013 (UTC)
That orange bar is amazingly helpful to the community - all of us started out as IPs before we registered (even if only as a reader). I still regularly edit as an IP, and I am still pleased at how often I get the orange bar from someone taking the time to welcome me. Sure many IPs may not respond from an IP address, but how many of them went on to create an account? Without that data we may be doing the project a huge disservice. Manning (talk) 02:56, 2 May 2013 (UTC)
(Speaking in a community capacity; I've had no real involvement in the project as a staffer.) I'm hugely glad that the ghastly orange box has gone. Especially because it doesn't work anyway - just like enwiki's hilariously-awful editnotices, which are similarly so 'visible' they trigger the 'this is an ad, please ignore' part of Web users' brains. I have faith that minor improvements to highlight the visibility of outstanding notifications will be sufficient. Regressing would be a huge mistake. James F.(talk)02:50, 2 May 2013 (UTC)
People resist change. I understand that. I believe this system is an improvement and should be default. I have no issue with an option to include the orange banner, but would prefer it was not the default. WormTT(talk) 12:30, 2 May 2013 (UTC)
Please don't revert back to the Orange bar for regular users. The new system is better and is least intrusive. Devs will figure out a way to prominently display the talk message notifications to IPs. — Ganeshk (talk)12:44, 2 May 2013 (UTC)
It's not just IPs that's the problem. It's the new editors who won't think to look at their talk page, or won't notice this tiny little notification. And to all in this section - those of us who want the bar back for new editors aren't saying 'make it compulsory for all'. We're saying to make it compulsory for new users and let everyone else be able to turn it off. Not everyone has much to do with new editors. Some of us do. Not only warning them, but helping them become real editors by getting them to stop doing something they didn't know was wrong, and explaining how to do it right. I think those wanting 'not to be bothered with plebs' are taking a very short sighted view. Peridon (talk) 14:34, 2 May 2013 (UTC)
There's also the non-trivial and (you would have thought) bindingly obvious point that making talkpage notifications equivalent to everything else in Echo gives users a strong incentive to turn off the other notifications, so that talkpage messages don't get overlooked. Which then reduces the whole thing to an expensive exercise in making things worse. This aspect of Echo is really a case of shooting yourself in both feet. Rd232talk15:40, 2 May 2013 (UTC)
Yes. In pre-Echo land, we have orange bar notifications, watch list notifications, watch lists themselves, fund-raising banners, etc. competing for attention in different ways. Users have de facto differentiation among different kinds of notifications. A unified system like Echo offers a great advantage if it pulls these things together, but a great disadvantage if it does so in a way that eliminates useful differentiation among them. The current two choices (web notification and email notification) might not be granular enough to get the job done. -Pete (talk) 16:36, 2 May 2013 (UTC)
No objections, but the current situation is a bit like taking a wheel off the car to save on tyre wear... Peridon (talk) 16:50, 2 May 2013 (UTC)
No. The more different options we have for users, the harder it is for new ones to acclimate. Helping new users becomes harder: "When you get a talk page message you should see that orange bar pop up." "What orange bar?" "Oh, right, most people don't have that anymore." It also inhibits future development; how much interface development has been hindered by the need to continue supporting old skins? PowersT17:44, 2 May 2013 (UTC)
No. If you need a huge orange bar to tell you have a message... every time you have a message, then perhaps the accusation of this becoming a social network is really already embedded in the culture. I know...maybe we need that to be a panic bar with a sound effect of an alarm and have it blink? Nah....how about just accepting this as an improvement (which it is) and moving forward.--Amadscientist (talk) 19:14, 2 May 2013 (UTC)
Eh? Amadscientist, you have seen my talk page. Yes, there are comments that might reasonably be construed to be of a social origin (notably, concerning health issues) but I am already struggling without more prominent notification of changes because the vast majority of changes to my page are directly related to article improvement etc. The talk pages of people such as Drmies and Malleus Fatuorum are replete with many jests, jibes and arguably "social" comments but I suspect that the ratio is pretty much the same, and the placement is more high profile. We humans are social animals and you cannot get away from that but to vaguely suggest that the bar is somehow co-related to the accusation of WP becoming a social network is, well, unappealing and misplaced. Since the bar's replacement is pretty much cosmetically similar to what is in fact a social network ... the argument looks weak. - Sitush (talk) 23:31, 2 May 2013 (UTC)
That is just so much bullshit really. You are a social networker yourself Sitush. You went to Malleus to tell him you mentioned him but couldn't be bothered to let me know you responded. Not in your social loop I guess. LOL!--Amadscientist (talk) 00:25, 3 May 2013 (UTC)
What? You are already involved in this thread, whereas Malleus and Drmies are not. Mine was a courtesy note and, taking my cue from ANI, we don't go round informing people who are already involved that we have mentioned their name/replied etc. Your comment is procedurally invalid. - Sitush (talk) 06:05, 3 May 2013 (UTC)
I have frequently had messages which require urgent attention from me as an admin. Anything which makes it harder to see that these messages have arrived is not just a problem for me, but could lead to a situation which requires attention compounding. It isn't a case of "orange bar suggests social networking", its a case of "orange bar indicates something requiring attention". Changing it to an itty-bitty number which may indicate you've got a message, or just that someone has changed something you edited somewhere is akin to replacing a fire siren with a small indicator light that comes on when it's warm weather. Grutness...wha?06:03, 4 May 2013 (UTC)
Why are you trying to force it on me? There's nothing close to serious support for requiring the orange bar; the idea is an optional thing that you can get rid of if you don't want it. Nyttend (talk) 05:05, 3 May 2013 (UTC)
No. (But I'm real conflicted in !voting 'no', since the shocking orange bar is a real assurance that I won't forget to pick up my new orange prison suit after reading the "You have been blocked indefinitely" message that awaits.) Ihardlythinkso (talk) 01:08, 3 May 2013 (UTC)
The orange bar was disruptive and visually unappealing. I'll be most pleased to have seen the last of it. Now if we could just get rid of hat-note accretion. Praemonitus (talk) 01:17, 5 May 2013 (UTC)
The orange bar shouldn't come back in the long run. It's ugly, and I'm sure we can do a better job of making a noticeable talk-page notification system. However, we need a working system to communicate with new editors, and the current implementation of Echo isn't providing that, so we should have the orange bar back until that gets fixed. I think that the deployment of Echo has been rushed; there should have been an opt-in trial on-wiki to get the more obvious bugs worked out before the orange bar was disabled. — Mr. Stradivarius♪ talk ♪02:06, 5 May 2013 (UTC)
I'm (obviously) biased. But here is my 2c anyway :-). From an engineering perspective, I dislike OBOD because it is notification code that doesn't live in the notification workflow. Going back to it (esp. permanently) entails the exact same ad-hoc long term maintenance and technical debt that is the why no unified notification system have been attempted in the last 5+ years. (The fact that Bug 47922 needed to be done, shows that there might be a benefit to addressing notifications to cover IP users and doing OBOD correctly.) However, that is long term, not short term; that is engineering, not UI. I'll assume the RFC is asking for a short-timeline UI fix. A lot of discussion has as an operating assumption that OBOD is super prominent and permanent. However, the click-through on OBOD is 50 percent at last test, so I'm uncertain whether or not its prominence, is actually that prominent. (There is a lot of anecdotal evidence for this: see Phoebe's comment, for example. My guess is OBOD looks too much like a banner which the average user has been trained to ignore, It is possible the tiny red number in notifications may actually be better adjusted for training time. OBOD looks prominent to us because we've been trained to look at it, but to new users, the red number is actually more in line with affordances with the rest of the web and thus is actually noticeable.) Even if the new notifications do perform poorly vs. OBOD, there should be a more UI friendly way to achieve equal or better response from urgent notifications like user talk notices. I'm on board with UI tweaks to make talkpage notices more prominent, possibly more permanent (non dismissable?/not easily dismissible), and then iterating from there as long as there is some real analytics to determine if we're actually headed in the right direction. When notifications has something other than user-talk notices cluttering up their notifications section, I think some more radical adjustment needs to be considered (splitting it into two streams a la other social networks)? -- tychay (tchay@wikimedia) (talk) 20:07, 6 May 2013 (UTC)
Hi Terry, thank you for sharing your views. I'd like to point out a couple things:
First, like many others in this section, you are arguing against a straw man. Very few if any Wikipedians are arguing that the original orange bar should be reinstated for anything more than a temporary period while an implementation and transition are agreed upon.
The 50% figure has been repeatedly cited as an indication that the orange bar is ineffective. I am curious how much that premise has been probed. As a counterexample, I routinely ignore the orange bar out of choice -- I see that the notification is from a familiar bot or from a user I'm not interested in talking with right then, and I ignore it for a while. If I'm in an active editing mode, 5 or 6 notifications might build up before I click through. None of this indicates a lack of prominence of the orange bar. I don't know how common my experience is, but it should be considered.
As someone who hasn't been around for long enough to get habituated to the status quo, I find the current Mediawiki interface extremely clunky and a major hurdle for new editors. The orange bar of doom is just one part of this—the first time I saw it I thought it was a strange bit of vandalism that had been added to the article I was reading. The new solution may not be perfect, but I'd prefer the developers forge ahead with some of the new suggestions at Wikipedia:Notifications/New_message_indicator rather than take time reimplementing an old system out of fear of the inevitable adjustment period. —Neil13:01, 7 May 2013 (UTC)
When working with new users, many of them didn't see the orange bar. This system is much more obtrusive, and that popup box and red colouring does a much better job showing the reader that there is something to pay attention to, Sadads (talk) 03:36, 9 May 2013 (UTC)
No - I already receive an email AND a UI notification when my talk page is changed. I don't need a the orange bar of death as well (or instead). WaggersTALK13:30, 10 May 2013 (UTC)
No, per my comment above. To answer the question embedded in the header, the answer has to be an emphatic "no". I strongly oppose bringing it back because this is progress, although I wouldn't object if and only if it was an opt-in feature. -- Ohconfucius ping / poke03:18, 11 May 2013 (UTC)
I've always thought that the orange bar was awful and unprofessional (it was something designed by programmers, for programmers IMO), so good riddance. The new system is far superior. Nick-D (talk) 05:21, 11 May 2013 (UTC)
I haven't read the details of the arguments. All I can say is "well done WMF engineers", on this occasion. The new system is less visually disruptive and provides valuable new alerts. Having said that, if people are really attached to the old orange bar, well ... it's no big deal to let them have it back (I guess a pref. tickbox is out of the question?). But please keep the red number against one's name at the top for the other notifications. Count this as a no. Tony(talk) 09:48, 11 May 2013 (UTC)
The orange bar was always obtrusive. It filled up a large part of the top of any article you went to, forcing you to immediately address it, even if you didn't want to. The new system is much better, still perfectly noticeable, but unobtrusive, and it even allows you to look back into the most recent changes to your talk page and it notifies you when other people mention you directly on Wikipedia and when people revert you. It is a much better system. SilverserenC02:39, 23 May 2013 (UTC)
Don't care
I'm not much of a fan for the Notifications, but the orange bar popping out of nowhere sometimes scares me due (probably) to its color. ZappaOMati20:24, 1 May 2013 (UTC)
I personally don't care, and based on editathon experience I don't think new users notice the orange bar either, so nothing lost there. -- phoebe / (talk to me) 21:52, 1 May 2013 (UTC)
snarfa (talk·contribs) just made the same point to me. But I wonder if edit-a-thon experiences may skew our perceptions? A possibility: when a newbie is looking to somebody sitting next to them for guidance, he or she may be less inclined to notice any notification. Meanwhile, many of us know from onlinee experience that some newbies respond to our messages on their talk pages, and it seems pretty likely that they found out because of the big orange bar. No? -Pete (talk) 22:01, 1 May 2013 (UTC)
Who knows. But I have myself seen more than one person who was staring at the screen when the orange bar popped up. I said "oh hey, you have a message!" They said, "I do?" I said, "yeah, see the notification?" They said, "nope." So that's my anecdotal evidence. Sure, of course, some folks do notice it. Some people will notice the little red thing too. I don't think we have much evidence one way or the other about relative noticeability, unless there's some user testing out there I'm unaware of. -- phoebe / (talk to me) 05:48, 2 May 2013 (UTC)
Phoebe, just wanted to let you know I saw this. I think it's a good discussion to have at some point, but at the moment am much more interested in the Engagement Team rolling back the elimination of a widely-used feature, to create space for a discussion of (a) what the best implementation of Echo will be, and (b) what is the smoothest and most effective path to get there. It appears to me that (b) has been tragically overlooked to this point, and I hope that changes very soon. I'll be happy to get back to a discussion of (a) once it becomes clear that (b) is on the radar. -Pete (talk) 20:34, 3 May 2013 (UTC)
I at least think that each user can choose to get notified of talk page edits by either the "notifications" template, or by the Orange "New Message" Bar. Whatever they choose. Bacon-Cheddar Man 5000 (talk) 13:11, 2 May 2013 (UTC)
I think it would help here if we can narrow down exactly what the "orange bar proponents" (so to speak) are asking for. An orange bar that loads at the top of article text, exactly the way it does now (er, well, did until now)? Something that loads at the top of article text, but can be any color? Something orange, that loads wherever? Anything highly visible, with no particular preference for whether it happens in the "menu bar" or the article text? Personally, I think there should be at least an option for a highly-visible, colored notification (larger than a small red box) that displays by default, but I'm not tied to orange, I quite dislike things loading on top of article text and shoving down the content I'm trying to read, and I don't think we need to preserve a legacy, clunky feature exactly as it was to please a small subset of users when we have a better-integrated way of doing the same thing. Unless you can spell out a bit more what you're asking for, AutomaticStrikeout, I'm not sure where that puts me in your yes/no choices. A fluffernutter is a sandwich! (talk) 16:20, 1 May 2013 (UTC)
I am asking if people should at least have the option of using the orange bar. I haven't seen any reason why it should be replaced and I think that at the very least, people out to be given a Preferences option restoring their orange bar. I'm not saying that it has to remain standardized, but I think it should be an option. AutomaticStrikeout (T • C • Sign AAPT) 16:27, 1 May 2013 (UTC)
Every time we create a preferences option, we create something else that needs to be maintained - even in cases like this where maintenance is probably trivial or non-existent, it sets a bad precedent. Okeyes (WMF) (talk) 16:28, 1 May 2013 (UTC)
Definitely. Removing longstanding basic features like this is a very bad thing, and WMF needs to start setting the precedent of restoring basic features that it shouldn't have removed. Nyttend (talk) 16:34, 1 May 2013 (UTC)
Comment Just to make one thing clear, I don't have a problem with most of this new system. A lot of it has the potential to be incredibly beneficial. On the whole, it's not a bad feature. The talk page notification issue is all I'm concerned about. AutomaticStrikeout (T • C • Sign AAPT) 16:34, 1 May 2013 (UTC)
It's not the color that's the problem – it's the prominence. WMF removed a very, very obvious form of notification, without any consensus whatsoever (or even warning). No-one really cares about the orange vs red; we care about having features not unilaterally disabled. -- Ypnypn (talk) 16:41, 1 May 2013 (UTC)
Big +1 to Ypnypn. But also, the problem is any sudden and unexplained change to the one notification system we've had for many years, which many thousands of people are used to using. I have no huge attachment to the orange bar, but if it's going to be removed, it needs to be done in a way that respects the end user. Sudden and unexplained removal (and by "explained" I mean explained in the interface, where those confused by it will encounter the explanation) is a huge problem. -Pete (talk) 16:47, 1 May 2013 (UTC)
The orange bar is intrusive for good reason. By all means provide an option for experienced users to turn it off (though I, personally, would not take that option), but for the reasons why the default needs to be something at least as conspicuous, consider:
The happy schoolchild who has got at Wikipedia in his break and started writing "Poop" everywhere. Not infrequently, that type of IP vandal stops after a {{uw-v2}} or {{uw-v3}} warning. Unless there is something eye-catching to draw his attention to a warning message, the first feedback he will be aware of is a block message.
The new good-faith user, unfamiliar with Wikipedia - for instance, a student who has been told to edit Wikipedia for a class project. Those newbies are often very hard to communicate with - I have the impression that many of them think of WP as a website to edit, not as a community they are joining and need to engage with. They are not familiar with the Wikipedia screen, may not know they have a talk page, and may well miss a small red blob. The orange bar says "HEY!!! We need to talk!" and that is exactly the message we need to give.
We knew that talk page notification visibility was going to be an issue, but we decided to go with the simplest implementation (to avoid complicating the UI/prefs if possible), and see what the community's response was. There were no objections raised during the beta period on MediaWiki, but of course English Wikipedia is a very different animal. Given the responses here, it seems clear that we're going to have to make talk page notifications more visible, which seems like a reasonable request. Give us a little bit of time and we'll come up with something. Speaking in my personal capacity as a long-time Wikipedian, I've always thought the orange bar of death was clunky and obnoxious. But that's just my personal opinion :) Kaldari (talk) 18:14, 1 May 2013 (UTC)
As Oliver concisely put it above, "orangeness" and "sameness" are different attributes. I don't think the orange bar needs to survive forever, but it's important to have it there now until there is a plan in place for a smooth transition. (I'll be starting a 6 week class on Wikipedia next week, and it's no fun for first-time users to have turbulence around basic features.) Is it a major technical challenge to simply reinstate the orange bar for the "talk page" notification item, and gradually phase it out in favor of the new feature, along with clear tooltips etc.? -Pete (talk) 18:42, 1 May 2013 (UTC)
Comment Kaldari, you KNEW it was going to be an issue and decided to ... talk about it after making the change? Talk about designing and documenting from the inside. If I ever get around to writing a book about the user experience and how to fsck it up, this is going to be a leading example. THERE ARE TIMES WHEN CLUNKY AND OBNOXIOUS IS A FEATURE! Sneaking changes in on users and then thinking ... give us a little bit of time and we'll come up with something. ROLLBACK is your friend. Especially when you've fallen on your face in it. Roll it back and, then think. (Do not worry, I won't be holding my breath for either event.) htom (talk) 00:41, 8 May 2013 (UTC)
"We knew ...", "We decided ...", "We're going to ...", "We'll come up with ..." ... certainly gives the impression of either being the spokesperson for a group mind, or a royal we. Doesn't seem to have been a bystander requesting that the brakes be applied. Group mind I think is more likely, all inside the project, all looking inside at the wonderful thing they've created. And we dare to not like it. Been there, seen that, too many times. Customers are such fickle beasts, wanting what they want until they want something else; force-feeding them may be profitable if what you want is their livers, but not if you want golden eggs. I'm a little sorry for pounding only on Kalari, pass it on to the rest of the "we", please. htom (talk) 03:45, 8 May 2013 (UTC)
Comment I wonder how many of the opposers of the bright bar actually spend time dealing with new users. Fluffermutter does, but the others I don't recognise from my normal area of deletion - where explanations, cautions and warnings take up a lot of time. They don't always work - but if we have don't have a good clear 'you have a message' thing of whatever sort, we won't get through to any of them. I've been here five years. It took a while this morning before I noticed the little red thing, and a while after that - having started to wonder why I had messages but no warning - that I found out what it was. New users are more concerned with what they're doing to worry about little red pimples. Peridon (talk) 19:13, 1 May 2013 (UTC)
Yeah, I had the rather ironic experience an hour or so ago of reloading my watchlist and realizing that someone had left me a talkpage message and - surprise, surprise! - the tiny red box had failed to get my notice. No question we need some version of "big honkin' notification here" message, absolutely for newbies and probably for us old-timers as well. I figure if we give them a little time, though, the devs will figure out a way to make Echo honk, rather than making me have to continue to rely on the "Oh god what am I in trouble for now" orange bar of doom - and at that point, the OBoD will be vestigial. A fluffernutter is a sandwich! (talk) 19:33, 1 May 2013 (UTC)
Peridon wondered whether opposers of the bright bar spent time dealing with new users. I don't think you are in that category. JohnCD (talk) 15:49, 2 May 2013 (UTC)
DGG, you are a regular at CSD. I learned a heck of a lot from you when I was patrolling Edits be New Accounts. And since. You weren't in my mind for that comment. I'm referring to those who don't have to deal with 'plebs' and 'newbies'. Peridon (talk) 16:56, 2 May 2013 (UTC)
Many of us calling for the return of the orange bar are doing so (at least in part) out of informed concern for new users. (I am teaching an intro class on Wikipedia.) It seems to me that in evaluating consensus, we should be disregarding the poorly thought out arguments -- and there are plenty on both sides here -- and focusing our attention on the best points. Peridon, your comment -- while well-intentioned, I'm sure -- points in the direction of ad hominem arguments and, ultimately, unnecessary drama. -Pete (talk) 00:57, 3 May 2013 (UTC)
I can imagine situations where I'd prefer the old way over the new way. So I think the default ought to be the old way (for the reasons John suggests above), but established editors should be able easily to opt for the new way if they want it. I think most editors have just been taken by surprise by something new here. Martinevans123 (talk) 18:06, 5 May 2013 (UTC)
(moved from No section) Should we include in our vandal notices say "to avoid receiving notification of these reports..." It is not the people who need the notification who want it back – they are never going to opt in if they are by default not getting the messages, the people who want the bar restored are those leaving the vandals messages, so we are sure they will see them. If there is to be an option, it should be a privilege that is earned, just like rollback, not a blanket option. Apteva (talk) 13:31, 12 May 2013 (UTC)
Interjection
It seems to me that most of the "no" votes [when this was posted, there were 6] are overlooking an important point: few if any of us asking for the orange bar back are seeking that as a permanent solution. I think just about everyone accepts that the general approach of Echo is an improvement -- it's just an improvement that needs to be implemented in a careful manner.
In other words "put the orange banner back now" is not the same as "put the orange banner back forever." Establishing a context in which we can compare the features side by side would be a strong step toward making a calm and consensus-driven decision. -Pete (talk) 01:19, 2 May 2013 (UTC)
On that point, I am one of those "few" (assuming in good faith that "few" is a fair representation). I love both the idea and implementation of Echo. I also completely understand the desire to not be seen as yet another way of grabbing people's attention, and that streamlining is in general a good thing.
The point of the new system appears to be to ensure that people can more efficiently navigate their way through the maze of Wikimedia, keeping on top of things that are important to them. It also accomodates editors with different methods of keeping track. Some editors would go through their notifications once per session, if that. Others would try to keep that little number at zero as often as possible. That difference in approach, that very flexibility that Echo tries to give different types of editor, is the root cause of the problem with removing the orange bar.
Talk page messages are unique. If I go to someone's talk page, more often than not my hope is that they will read it before their next edit. More often than not, a message that can wait can be posted elsewhere. When a warning/request to cease editing and start discussing is posted, there are circumstances in which failure to heed that call could lead to hot water up to and including a block, and this isn't exclusive to IPs or newly registered editors.
So having made the fairly mainstream case for doing something special for talk page notifications, why do I go a step further and say that the orange bar should remain in the long term? Because it is absolutely HIDEOUS, and something that most people want to get rid of as soon as possible. I would cite the clampdown on OBoD spoofs as evidence. Why were they so annoying? Because old and new editors alike saw them and clicked on them immediately, in large numbers, some believing that the link on the other side was worthy of their immediate attention, others wanting to get rid of the ugly bar as soon as possible and ending up at something even less appealing to the eye.
Ultimately, the question for developers and the community alike is where the balance lies. In life as well as on-wiki, I lean heavily towards getting the job done as effectively as possible, regardless of other considerations. That said, I accept that "regardless of other considerations" puts me at one end of the spectrum, and don't have a clue where the middle-ground is. Spend time and resources on a more elegant solution if you must, and if you go down that route I concede that you are likely to find a compromise that achieves consensus. That said, I'm certain that there is nothing you can implement to do the job more effectively than the OBoD. —WFC— FL wishlist07:00, 3 May 2013 (UTC)
To follow up with this: I've made some improvements to my orange bar of doom script; it doesn't use cookies any more, so most of the problems people were having should be fixed now. As a bonus, it uses the Echo API now, so visiting your talk page will automatically clear the Echo notification for your talk page as well as the OBoD. Big thanks to Legoktm, Kaldari, et al. for helping me figure it out! Writ Keeper⚇♔17:48, 3 May 2013 (UTC)
Another followup: I believe I just fixed a bug that prevented it from working in older browsers, notably IE8. As ever, please let me know if any other bugs turn up. Thanks! Writ Keeper⚇♔06:32, 4 May 2013 (UTC)
The proper method for rolling out such a train-wreck as ECHO would have been to:
Do alpha testing
Beta testing
Roll out to smaller wikis
Deploy to enwiki as an optional opt-in feature
Extensive testing and feedback over an extended period (several months)
Switch ECHO to default notification system for new users, while leaving it opt-in for those veteran users.
If at some point after this ECHO finally doesn't suck, discuss the options with the community about phasing out the old system.
Developer actions via edict with such a large impact should not be possible. The reaction that the developers are taking (you are screwed the OBoD will not come back regardless of consensus because we say so) is appalling. Given the large amount of issues ECHO caused the update should have immediately been rolled back and the devs should have gone back the the drawing boards to fix it. This IP editors get zero notice about talk page messages for 24 hours is really problematic. How many users ended up getting blocked/where deterred from editing because they where blocked for disruptions because they didnt get the talk page note saying what they where doing was inappropriate? Werieth (talk) 17:00, 1 May 2013 (UTC)
And you think we shouldn't deploy on enwiki until there are zero bugs? That's literally never going to happen; there are far too many possible things that can go wrong on individual wikis, from site-specific CSS to use-specific CSS to browser choices.
I'd argue strongly that Echo does not 'suck'. This specific feature, I agree, is worse than what it replaced. I warned the team that this was a problem with it. Right now, I'd suggest we all calm down and see what they'll come up with as a solution. Okeyes (WMF) (talk) 17:12, 1 May 2013 (UTC)
This is what is called incompetence, you do not break a top 10 website and then go home and hope to fix it the next day. I have been in the IT, software development process for many years. Before I put my name on a product and stamp it for release I make sure that it is more than 50% functional. When deploying a major upgrade I assume Murphy's law and have fallback plans. I dont go home until the system is functional. Yeah it might not be 100% bug free but it will not leave a major breakage while I go home and watch TV and relax. If crap has hit the fan its a matter of all hands on deck, or rolling back the updates until such time as a postmortem can be preformed and we can figure out what went to hell. Given the fact that this is one of the top 10 websites in the net I would hope that the paid techs who rolled this out would be competent enough to revert the changes if a patch to fix a major flaw wasnt implemented ASAP. Werieth (talk) 17:23, 1 May 2013 (UTC)
This is what's called 'the point at which I leave the conversation'. Please do let me know when you can be reasonable about the issues. Okeyes (WMF) (talk) 17:26, 1 May 2013 (UTC)
I see how it is, when a reasonable deployment process that will work is proposed in the aftermath of a train wreck that you orchestrated you decide to leave the conversation? I never said that it had to be bug free, in fact if you take a look at the plan that I suggested there is a period of several months where bugs can be identified and fixed before shoving it down the throat of your users. You dont just make a single mass switchover, take a look at the WMF move to ashburn they had everything but the master servers already tested prior to the migration. Even then they planned a fallback to revert to the tampa servers if there where problems. I really dont see the forsight and good planning that was shown there in this project. Werieth (talk) 17:32, 1 May 2013 (UTC)
No, actually, it's the bit where you start referring to people as incompetent. In what world does that aid the conversation or make participants more likely to listen to you? Okeyes (WMF) (talk) 17:37, 1 May 2013 (UTC)
Sorry I am unsure what else to call it. Any other term that comes to mind is far more derogatory. I was just attempting to describe the actions as I saw them. Why wasnt something like what I proposed implemented? Instead something half baked and full of issues was thrown out and left running for an extended period of time without fixing it. Werieth (talk) 17:43, 1 May 2013 (UTC)
"One night" is not an extended period of time. We have fixed the IP issue already, and will be iterating improvements to the general UI problems over the coming day. Okeyes (WMF) (talk) 17:52, 1 May 2013 (UTC)
That's context dependent. In this context, I agree with Werieth that one night is a very, very, very extended, long period of time. -Pete (talk) 18:06, 1 May 2013 (UTC)
(edit conflict) Werieth, Okeyes is telling you that he agrees with you, that he and the dev team are doing their best, and that they're going to be working on it, even if a bit later/slower than you wanted. Is there any value in continuing to hurl nasty words at them telling them how terrible they are, rather than just describing what you see as the faults in the software and/or development system? Nobody at the WMF or among the devs is trying to do anything badly or to offend you, so please give them the courtesy of speaking to them as if they're human beings. A fluffernutter is a sandwich! (talk) 17:41, 1 May 2013 (UTC)
Fluffernutter, given the fact that this isnt the first time recently that the devs have unilaterally implemented features that have gone against the communities wishes. (see watchlist-gate within the last year). Given the backlash that a minor change like that caused in comparison to this change it was very poorly thought out, and even more poorly implemented. In most paid positions this kind of screwup would cost them their job. I am not asking for that, rather I am trying to figure out why common sense and careful roll-out did not happen. Given the impact that minor issues can snowball on a site this big a proper reaction would have been to revert the roll-out, get feedback about the bugs that where discovered and develop a phased implementation that wouldn't cause such a large scale disruption. Standard practice when updating/migrating software in a large environment is to do it in small manageable waves where testing can be reformed, feedback can be assessed, and issues resolved before full migration. Measured implementation and testing practices are key to smooth deployments. Like I said before compare this process to the data center move, things where extensively tested prior to the move and there was a fallback plan if there where issues to mitigate extended down time. In this case there wasnt a fallback plan or if there was it was ignored, causing a major disruption to site operations. I am asking why logic and a practice of ensuring minimal downtime was ignored here. I am also appalled by the fact that the limited number of devs can say screw what the community wants, if we what to add a cool hip new feature, regardless of how broken it is we will and you can all go to hell since the old method will never come back. Personally I find that JavaScript intensive web pages are problematic especially when tools like resource loader are used which makes filtering out annoying javascripts impossible. I remember when BITS was first introduced it was geolocating and doing other things that easily added 3-10 seconds to the load time of a web page. I just opened up my browser settings and blocked that script. Since everything is being minified its no longer possible. I tend to prefer the monobook skin due to the fact that its not bloated with all the "cool" looks and bloated javascipt that vector uses. Remember KISS when developing. Werieth (talk) 00:20, 2 May 2013 (UTC)
The value would be in underlining that the problems caused could and should have been avoided, and that lessons should be learned about how this went live. Rd232talk20:22, 1 May 2013 (UTC)
The number of issues is probably a distraction. It's the significance of the issues. No transition from the orange bar planned out? That may be just one issue, but it is…well, I think "large" would be an understatement. (I think there is a lot of quality in Echo, but as has happened before, a poor rollout strategy will sink you regardless of how good the core feature is.) -Pete (talk) 17:15, 1 May 2013 (UTC)
I do not care about anything that Echo has added. I use my watchlist and contributions tab to keep track of those things. I care about what it subtracted. None of the new notifications are worth getting rid of the prominent talk page notification IMO. --Onorem♠Dil17:19, 1 May 2013 (UTC)
Indeed Pete. The IP issue alone makes it "really suck" unfortunately. Most of the policies for editors making test edits revolve around them being warned on their talk page - without that there is a huge risk of resorting to mass amounts of blocks to good faith editors instead of a simple talk page message. That said, I fully expect a fix to be deployed within the next day or so (or to see it taken down temporarily). Also, to be fair, I've seen plenty of buggy things deployed over the years on en-wiki. Like it or not it does tend to be a testing ground for things like this despite being a huge website. Ryan Norton17:35, 1 May 2013 (UTC)
Agreed; the fix is scheduled to go out in a couple of hours (crossing fingers!). I spent a chunk of thisd morning poking people at godawful hours to secure a deployment window. Okeyes (WMF) (talk) 17:37, 1 May 2013 (UTC)
Thank you; I've already blocked a bunch of two-time-vandalism IPs because there wasn't any way to tell them to stop. Nyttend (talk) 17:44, 1 May 2013 (UTC)
Great! Just tested it, it works as expected. Still concerned about the orange bar in other contexts, as discussed above -- but glad to know this issue has been resolved. -Pete (talk) 18:19, 1 May 2013 (UTC)
Good, thank you! Inability to communicate with IPs was a really serious problem, and it is excellent that you have been able to resolve it so promoptly. As you have said that he orange bar will not be retained, I presume this is a temporary fix: what is the longer-term proposal for IPs? JohnCD (talk) 21:42, 1 May 2013 (UTC)
RfC?
I note that User:TheOriginalSoni has now turned this discussion into an RfC. Ignoring the substantial problems with trying to formalise something after a lot of conversation has gone on - frankly, you need to start an entirely new discussion if you want the outcome to mean anything - and the long lasting time of an RfC, Eloquence (Erik) has made his position on the orange bar clear. The Echo team is not going to be turning it on again permanently, even as an opt-in, whatever happens. I'd strongly suggest that someone who isn't me simply remove the RfC tag, because it's not going to lead to a different outcome, I'm afraid :/.
I woke up this morning to a set of emails from the team, basically outlining "we think we have a solution, but we'd like to A/B test it against Echo to make sure it's actually an improvement". I'll have a more full update and some screenshots to share in a bit, but I wanted to try and be transparent about what the team is doing (or, at least, what they report to me they're doing). I should have more from them in a bit :). Okeyes (WMF) (talk) 13:13, 3 May 2013 (UTC)
It's an RfC now. 83 support the orange bar, 19 oppose, and a lot of the opposes are saying they want the orange bar back temporarily. I suggest the WMF team get the rods out of their behinds and reinstate the orange bar, even if only until they roll out whatever "fix" they have in store. Ignatzmice•talk13:56, 3 May 2013 (UTC)
This is not something I have any control over, although, trust me, I have made clear to them that we need a result sooner rather than later. Okeyes (WMF) (talk) 15:18, 3 May 2013 (UTC)
Okeyes, I'd strongly suggest that you let the RfC continue, so that the community's consensus on the matter becomes clear. Whether the WMF acts on that consensus is up to the WMF. Besides, currently the RfC seems to suggest that something (not necessarily the orange bar) better than the current notification is needed, and I gather from your comments on this page that something is in development/QA right now. I doubt stopping the RfC now is going to score any points for the new release, the dev team or the WMF. ChamalT•C14:19, 3 May 2013 (UTC)
I'm not going to stop the RfC; it's not within my power :). I'm trying to persuade the Product team to fix this quickly, because I don't want Echo to become exclusively/specifically about this issue, and there's a risk it will. Part of that, however, is trying to stop the situation from escalating, and also (as a general point) trying to avoid users wasting their time contributing to a discussion that will not, simply via a template, be any more or less useful or binding on the Product team than it was before. Okeyes (WMF) (talk) 15:18, 3 May 2013 (UTC)
"Eloquence (Erik) has made his position on the orange bar clear. The Echo team is not going to be turning it on again permanently, even as an opt-in, whatever happens." I really hoped that after the ACTRIAL row Erik and the WMF would not return to the attitude of "We've decided what's best for you, and anyone who doesn't like it can go jump in the lake, because whatever you say, and however many of you say it, we're not listening", but here we are again. JohnCD (talk) 17:19, 3 May 2013 (UTC)
I could live with "we won't turn it on again permanently", if they would just turn it on again temporarily whilst developing improvements to Echo that are at least as good as the Orange Bar. I think this refusal is disappointingly pigheaded and disrespectful of the community. It's the sort of behaviour you expect from a corporate like Facebook, not from a Foundation which is basically there to (a) run the tech and (b) keep the volunteers happy so that the tech serves a purpose! Rd232talk17:29, 3 May 2013 (UTC)
I absolutely agree! The WMF and their techies need to remember they work for the editors - not the other way around. We've had this same struggle before, about a year ago over some really dreadful watchlist changes. Roger (Dodger67) (talk) 07:06, 4 May 2013 (UTC)
Stopping the RFC is not going to prevent the discussion getting bogged down in this issue. What will stop it is temporarily turning the orange bar back on, then we can move on to other things and start congratulating the devs for a wonderful notification system while they think about how to properly incorporate the functionality of OBOD. I have made clear to them that we need a result sooner rather than later—that's not soon enough, there have been over 600 blocks in the last 24 hours and most of them will not have seen the warnings. This is an urgent matter that requires dealing with NOW. SpinningSpark06:32, 4 May 2013 (UTC)
I assume your question is serious? It is the orange bar across the top of the screen saying "You have a new message" which, until Tuesday, used to warn you of the arrival of a new message on your talk page. You will no longer get that; instead, a small blob just to the right of your username on the top line will turn from grey to red to signal a new talk page message or various other events. See Wikipedia:Notifications and Wikipedia:Notifications/FAQ for more detail. JohnCD (talk) 20:42, 3 May 2013 (UTC)
As mentioned below, an option that does not have a dependency on the devs is to turn my script into an official gadget; this has been suggested by others, particularly MBisanz below. I've just fixed the last of the bugs that's been reported to me; once I hear back from the affected user (User:Nyttend), I won't be unwilling to have this happen. Writ Keeper⚇♔06:41, 4 May 2013 (UTC)
That's useful for experienced users who know about gadgets but (unless the gadget was ON by default, and I doubt whether that would be allowed) it does nothing for the IPs and newbies who most need to be made aware of a talk page mesage. JohnCD (talk) 11:42, 4 May 2013 (UTC)
I wasn't here for the Watchlist War, but it seems an admin was able to edit the site-wide CSS to revert the WMF's changes. Supposing we could confirm User:Writ Keeper's orange bar is (mostly) bug-free, and also supposing we could make it very easy for users to turn it back off again, I would totally support something like that happening here. We wouldn't turn off Echo, only import that bit of script. Although come to think of it, while it makes sense for there to be a site-wide CSS page, it's not as entirely obvious that there's a site-wide JavaScript page.... Ignatzmice•talk11:53, 4 May 2013 (UTC)
The IP editor, new editor, argument for retaining it for those editors sounds like a sound reasoning IMHO. If there is a way to allow more experienced editors (say who have made more than 500 constructive edits) to opt out of its display, I think we should go with that solution.--RightCowLeftCoast (talk) 13:56, 4 May 2013 (UTC)
Yes, turn on the gadget now. Even if this is only very short term, we have a serious problem right now. But no registered user should be prevented from opting out. If they have enough experience/clue to work out how to turn it off, they are probably aware of the consequences of doing so. In any case, it is their responsibility to read their messages. What we are concerned about here are the IPs and new users who have no clue about the messaging system and were completely unaware they had any messages to read - until they attempt to edit and discover I have blocked them that is. I really don't want that to be the first communication they have from me. SpinningSpark14:33, 4 May 2013 (UTC)
Well, the script is here. I suggest that, if this is to actually happen, Writ Keeper should update it so that the bar says something like See the documentation page if you want to turn off this bar and the doc page should have instructions on going to Special:Preferences#mw-prefsection-gadgets and unchecking the correct box. Or the script could be moved to a centralized location, and then anyone with sufficient competence could add that. I don't know what the protocol is for gadgets. Ignatzmice•talk14:39, 4 May 2013 (UTC)
Question Who has the actual authority to decide to restore the bar as a temporary measure while this is being discussed, or to block its restoration? DGG ( talk ) 19:15, 5 May 2013 (UTC)
Any admin has the power to make Writ's script into an on-by-default gadget, though we should wait until he gives the go-ahead. Only the Editor Engagement Team (Eloquence, Okeyes, Fabiance, Jorm, there are others I don't know; maybe only a specific person on the team) can bring back the original orange bar. Ignatzmice•talk19:18, 5 May 2013 (UTC)
Tiny orange bar
For the record, I oppose the new tiny orange bar as a replacement for the Talk link at the top, and request that developers or whoever either restore the original orange bar or quit making changes until this RFC has closed. Apteva (talk) 14:57, 11 May 2013 (UTC)
I like the tiny orange bar. It's quite visible, but much less disturbing than the orange bar of death. --NaBUru38 (talk) 18:40, 13 May 2013 (UTC)
Strong oppose for being a replacement, it could only be without JS and if there by default. But by now, this new bar isn’t shown at all without JS, so it isn’t able to replace it at all. There is a must for accessibility and for unobtrusive JavaScript, neither the old nor this design is able to replace the old orange bar in any way, ’cause both aren’t shown at all. I’ve now placed a warning message onto my talk page, so that users are at least aware that I’m not seeing new notifications anymore (except accidently or by looking for it) since this big change, and I think there will be many users who aren’t getting notification bars of any kind of design anymore, so it has to be changed and it should never get live on other wiki versions until this big bug will be fixed. --Geitost12:28, 19 May 2013 (UTC)
If anyone's still asking, I like the mini-OBOD. 'Twould be nicer if reading it cleared the notification count but no big deal. Jim.henderson (talk) 21:48, 26 May 2013 (UTC)
I kind of like it, too. I partly like it precisely because I can ignore it for a bit without losing so much of my screen real estate. That means I can finish whatever I'm busy with before being dragged back to my talk page again. I suspect that this desirable benefit is going to be exactly why some people oppose it, though. WhatamIdoing (talk) 23:44, 26 May 2013 (UTC)
THANK YOU. This is what I've been saying. I don't care what the damn shape or color of the thing is. But the orange bar appeared right after a talk page message was left for me, and provided a link taking me to the diff. The little red box does neither. Nightscream (talk) 00:09, 18 May 2013 (UTC)
heart, Okeyes. I am just a cynic, and wonder what kind of editors the sparkles and shinies attract. (As a feminist, I will admit to a certain self-suspicion: since I am not now nor have I ever been a pre-teen girl, perhaps I am not the kind of person these things are supposed to attract and retain?) --Orange Mike | Talk 19:25, 31 May 2013 (UTC)
Done The Anonymouse (talk | contribs) 15:22, 1 May 2013 (UTC)
Already done The Anonymouse (talk | contribs) 15:22, 1 May 2013 (UTC)
![[3]](http://i.imgur.com/5s0yRVw.png){kind=link}