User talk:Rich Farmbrough/Archive/2009 February

Previous · Next

Nobel Laureates

The reason Laureates is capitalized is because that is how they do it at the website of the Nobel Foundation. [1] I really wish you would have discussed it first. -- Scorpion⁰⁴²² 16:12, 31 January 2009 (UTC)

Regardless of the issue above, shouldn't the name of the prize be capitalized? I think that say "laureate in Physics" is still correct, as the "Physics" part is referring to the prize name, which is capitalized. — sephiroth bcr ^(converse) 06:35, 1 February 2009 (UTC)

Wikipedia Signpost, January 31, 2009

The Wikipedia Signpost Weekly Delivery

Volume 5, Issue 5	31 January 2009	About the Signpost

Large portion of articles are orphans	News and notes: Ogg support, Wikipedia Loves Art, Jimbo honored
Wikipedia in the news: Flagged Revisions, Internet Explorer add-on	Dispatches: In the news
WikiProject Report: Motto of the Day	Features and admins
Bugs, Repairs, and Internal Operational News	The Report on Lengthy Litigation

You are receiving this message because you have signed up for the Signpost spamlist. If you wish to stop receiving these messages, simply remove your name from the list.--ragesoss (talk) 20:49, 1 February 2009 (UTC)

Delievered by SoxBot II (talk) at 21:58, 1 February 2009 (UTC)

Photo in Nair

Dear Sir, The Photo album in Nair section is good but all the poeple are already expired except Shashi Tharoor and Mohanlal. Don't we have any more great people living in the Nair Community at present? And what was AK Gopalan's contribution to Nair community? Please add photos of some living legends like Dr.G. Madhavan Nair (ISRO chairman), Lt. Gen Satish Nambiar (UN under secretary), Vijay Nambiar( Foreig secretay GOI) and atleast One living woman like Parvathy Omanakkuttan. —Preceding unsigned comment added by 59.92.244.218 (talk) 06:39, 2 February 2009 (UTC)

SmackBot's edit of Vector space

Hi Rich,

SmackBot removed some “+” signs that it shouldn't have from the vector space article (the edit). It also added a “[” to “the polynomial ring F[x]]” where the correct fix would have been to remove the trailing “]” (but, of course, a bot couldn't be expected to know that).

Alksentrs (talk) 11:12, 2 February 2009 (UTC)

Smackbot and inline templates

Smackbot is (I believe) changing some template names in a way that it shouldn't. You can see an example here. In particular, it's changing {{or}} to {{original research}} and {{syn}} to {{synthesis}}. Those templates aren't equivalent; in each case, the former is inline and the latter isn't. Is this a known bug, a feature, or ? Thanks. Dori (Talk • Contribs) 00:32, 3 February 2009 (UTC)

Smackbot removed the uncat tag I posted

Is this because there was already a stub on the article? --Closedmouth (talk) 12:21, 3 February 2009 (UTC)

No it may have been been a bug with WP:AWB that has now been fixed. I will investigate. Rich Farmbrough, 14:10 3 February 2009 (UTC).

Thank you. (PS. Sorry for swearing at your bot, I really do love him.) --Closedmouth (talk) 15:01, 3 February 2009 (UTC)

Template "Confusing"

The "confusing" template is now putting code out onto the page on some of the pages on which it appears. I can't read wikicode well enough to tell whether your recent edit to the template is the reason for this or if it was like that before. Examples of what I mean: Admiral Kuznetsov class aircraft carrier, Solipsism#Neuroscience, Teletubbies#Overview. Not all pages seem to have the problem and I can't find any pattern which predicts whether it will appear properly or not. Since you recently edited the code near the area where the problem seems to be, I thought that you might know something about. Perhaps your edit fixed it on some pages but not on others. Would you mind looking it over to see if you can find out what's wrong? Soap ^Talk/_{Contributions} 22:45, 3 February 2009 (UTC)

I went ahead and reverted your edit. The problem articles are back to normal now. However I understand what you were trying to do and agree that it would be a good idea. I think that the source code on this page might help. I tried to figure it out myself but it is really beyond my skill level to understand what is going on there and how that kind of code could be applied to the Confusing template. Soap ^Talk/_{Contributions} 22:57, 3 February 2009 (UTC)

That's fine, I'll check it out later. It's what revert is for. Rich Farmbrough, 23:46 3 February 2009 (UTC).

It seems to be working now. Thank you. Soap ^Talk/_{Contributions} 16:53, 4 February 2009 (UTC)

SmackBot

 —Preceding unsigned comment added by Mike6491 (talk • contribs) 13:41, 4 February 2009 (UTC) 

Please review changes to E2m.org. Thank youMike6491 (talk) 13:44, 4 February 2009 (UTC) Michael Garjian

Progress report: Country outlines

Development has been slow but continuous:

Penubag has done a fantastic job on the images for the awards we'll be using for our project's collaborations and contests. We now have 3 awards: a medal, a trophy, and a race ribbon. They all look tight. The trophy needs a small adjustment, but other than that, all 3 award images are complete and ready to use to create awards with.

Spartaz has warned us of (threatened to take) G4 (speedy delete) action if we run a competition that resembles the previously deleted Awards Center page. So whatever we do, any contests we run must differ substantially from the methods used there.

One type of competition I've been exploring is edit racing. I'm in the process of working the bugs out of this concept - the first race didn't work as expected - you see, because we only had an award for first place, the opponent didn't think it worthwhile to continue once it was clear who the winner would be. And since editors are in different time zones and usually need to start the race at different times, we need to base winning on personal start times - he who completes his assigned edits in the least time (rather than first), wins. And last but not least is quality control. What good is winning if your edits are ripe with errors? So I'll be exploring possibilities such as using a referee (whoever is overseeing a particular race), having participants watching each other for errors to knock them back, etc. I'm not sure yet.

Rich Farmbrough has been applying his bot expertise to filling in blanks in the country outlines (the population and area entries). I'm amazed at the number of edits he pumps out each day on a myriad of projects - ours makes up but a small time slice of his activity, and yet he has saved us many hours of manual work. Perhaps we should look into how he gets so much done.  :)

Zlerman has chosen to work on one outline at a time, and is taking on Azerbaijan, Tajikistan, Turkmenistan, and Uzbekistan. He also has been keen on noticing and reporting design issues pertaining to the whole set of country outlines. Keep up the good work!

Highfields has been filling in the names of capitals, and is our first race winner. Check out the award on his user pages.

As you probably know, this project has expanded to include working on any and all sets of pages that are linked to from the country outlines. Once the set of country outlines go live (in article space), traffic will likely increase for all the links included on them. The quality and usefulness of those pages will reflect heavily on the country outlines (the outlines, which are essentially lists of links, are only as good as the links they present), and therefore we've branched out to solve the biggest problems with those as well. So far, we've taken on:

• The creation of disambiguation pages for country adjectivals ("German", "French", "Taiwanese", etc. About half done.)
• The clean-up of the CIA World Factbook statistics on the "demographics of" country pages. We've been renaming those sections to provide a key string that AWB can use for targetting (for skipping and filtering). Once that's done, we'll be able to break the clean-up down into simple AWB search/replace tasks, because we'll be able to target just those pages that include the CIA stuff.
• Renaming the "Cuisine of" articles to their adjectival forms ("Chinese cuisine", "Italian cuisine", etc.)

Blackadam2 and Thehelpfulone have been helping out with the "demographics of" pages mentioned above.

And we have a couple speed addicts (addicted to wiki-velocity, not drugs)...

Both Robert Skyhawk and Thehelpfulone prefer (and excel at) simple AWB search/replaces. Robert hasn't actually joined our team yet, but he has been helping out quite a bit from the sidelines (via the WP:AWB/Tasks page. Unfortunately, there has recently been a non-AWB chore that has been holding things up on the AWB front - an edit to all the the headings which had to be reverted before too many new edits were made, because any new edits would make the reversion more difficult. The headings have been restored, so now the way is clear for AWB operations, and there are many search/replace tasks in the queue. AWB assignments have started again!

There's a similar bottleneck on the "Demographics of" pages (the "keying" mentioned above), but that's almost cleared too.  :)

With my internet access somewhat crippled, I've been finding it difficult to keep up with you guys. However, I expect to be accessing a Linky-capable workstation on a faster server (I'm on it right now, as you can probably tell from my contributions list for today), and so I should really pick up speed. Feels goooooood.  :)

Recruiting has been a bit slow (but steady), due in part to my crippled access, and because we've been waiting for the images for the awards to be completed. I expect the team to grow more rapidly as the bottlenecks are removed.

Well that's what's been happenin', and here's what's in the pipeline...

I'm about to begin work on a set of lists that corresponds to all the standard links on the country outlines, and these will be presented on the Topic outline of countries which will be organized exactly like the country outlines. Aside from being an extremely useful navigation aid, it will allow editors to easily see the state of country coverage on Wikipedia. I'll provide you with a link once I get up to speed on this.

In the meantime, keep up the good work!

Cheers,

The Transhumanist    05:29, 5 February 2009 (UTC)

OR template error

Hello, 01:37, February 1, 2009, I added the template {{Or}} to the article cortisol. You may notice it under the section titled Effects, sub-section insulin, on the 6th sentence, or just about half way through. In this edit, you automated smack-bot created two mystakes.

1. It changed the template of {{Or}} to the template of {{Original research}}. It also added the date. Because the first template is inline and second one isn't you may notice this creates a formating issue.

2. It changed [[blood]] to [[blood|blood]] adding extra unecessary coding.

Best regards and I hope to hear from you soon. --CyclePat (talk) 23:45, 4 February 2009 (UTC)

1. Yes thanks, known bug already fixed.
2. No, it made the reverse change [[blood|blood]] to [[blood]].

Thanks for your note. Rich Farmbrough, 01:48 5 February 2009 (UTC).

Got your message. You are correct. Thank you. --CyclePat (talk) 14:58, 6 February 2009 (UTC)

Template:Articleissues‎

Hi. Do you know any idea what should add in the code that if someone tries to add Articleissues without any parameters to get an error message? -- Magioladitis (talk) 16:01, 5 February 2009 (UTC)

Answered on user's talk page. Rich Farmbrough, 17:50 6 February 2009 (UTC).

No content in Category:Articles containing Yi language text

Hello, this is a message from an automated bot. A tag has been placed on Category:Articles containing Yi language text, by another Wikipedia user, requesting that it be speedily deleted from Wikipedia. The tag claims that it should be speedily deleted because Category:Articles containing Yi language text has been empty for at least four days, and its only content has been links to parent categories. (CSD C1).

To contest the tagging and request that administrators wait before possibly deleting Category:Articles containing Yi language text, please affix the template {{hangon}} to the page, and put a note on its talk page. If the article has already been deleted, see the advice and instructions at WP:WMD. Feel free to contact the bot operator if you have any questions about this or any problems with this bot, bearing in mind that this bot is only informing you of the nomination for speedy deletion; it does not perform any nominations or deletions itself. To see the user who deleted the page, click here CSDWarnBot (talk) 04:20, 6 February 2009 (UTC)

SmackBot

Your bot usually seems to do a great job but it somehow messed up with dn templates on Submarine. I fixed the damage but thought you should know. Rees11 (talk) 14:00, 6 February 2009 (UTC)

SmackBot (mod_python)

In this edit smackbot changed a reference to "mod_python" to "mod python". The module name, and the article name, actually include the underscore, it's an error to remove it. TJRC (talk) 01:06, 6 February 2009 (UTC)

Thanks, yes this has come up before, I thought it was a resolved (WP:AWB) issue. There's some dozen page names it applies to. Rich Farmbrough, 17:52 6 February 2009 (UTC).

I'm probably misunderstanding your reply; but I don't think this edit was a WP:AWB issue. The edit was made by SmackBot. TJRC (talk) 18:33, 6 February 2009 (UTC)

AWB is the platform for SmackBot. Rich Farmbrough, 03:37 14 February 2009 (UTC).

Smackbot strange date fix?

The bot is removing dates from citation templates. [2] Was this intended? --JonIsaksen (talk) 01:37, 7 February 2009 (UTC)

Thanks for the note. This shouldn't happen I will look into it. Rich Farmbrough, 00:03 9 February 2009 (UTC).

American Society of Theatre Consultants

The page has been rewritten to remove what I believe you were thinking was advertisment type text. Please give another look and if you agree, please remove your notice at the top. Thanks so much. MaddyStephens (talk) 18:58, 7 February 2009 (UTC)

Cap Arcona

The nationalities (24/28) are visible on the memorial (photo : File:Neustad Holstein Cap Arcona.jpg). —Preceding unsigned comment added by 86.64.182.240 (talk) 16:44, 7 February 2009 (UTC)

SmackBot just dates the tags. Rich Farmbrough, 17:17 7 February 2009 (UTC).

Category:Articles containing Yi language text

You're not supposed to remove {{db-catempty}} notice from pages that you've created yourself. --Stepheng3 (talk) 18:46, 6 February 2009 (UTC)

• As the creator, you may feel some personal attachment to the page. The instructions say to apply {{hangon}} instead of reverting.

I'm curious: do you intend to contest the claim that the category has been empty for four days? Do you have plans to populate it in the near future? If not, I'll probably mark it for deletion again. - Stepheng3 (talk) 19:03, 6 February 2009 (UTC)

• If you'd followed the instructions, your present fears would not have been realized. Here's why: the admin would see your {{hangon}} and find your rationale on the Category talk page. The complete page history is available for inspection, so the fact that I blanked the page would not prejudice your case to the admin.

I blank pages as an advance warning to interested parties that the category is subject to deletion, which gives you an extra four days to react. If I hadn't blanked it, it might have gotten deleted before you had a chance to object.

Personally, I'm not swayed by your arguments. You say that the category is part of a complex system, but it looks simple to me; I think I could recreate the category using one of its sister categories as a model. You say that there will be articles in the category "soon," but the time-horizon for C1 is 4 days. Do you mean you within that time? The category has existed for over a years and appears to have been empty since at least October. Would you have used the same argument back then?

-Stepheng3 (talk) 03:51, 9 February 2009 (UTC)

A note to the creator would have the same or better effect. In order to recreate the category you would ideally need to know the ISO 639 codes for the language, and link it into the parent category. However you wouldn't know the category was needed because if it doesn't exist the template the transcludes it will not do so, in order to avoid redlinks and breakage in certain other cases. Youwould only find the category was needed by scanning the database, assuming a db dump is ever again successful, and running a custom extraction program, and comparing that to a list of ISO 639 codes, and in turn IO 639 lang templates, and eventually categories. Rich Farmbrough, 03:39 14 February 2009 (UTC).

El Greco

Hi Rich, this edit [3] broke the infobox rendering on the El Greco article. I reverted it. cheers Tom B (talk) 23:51, 8 February 2009 (UTC)

Thanks fixed. Rich Farmbrough, 00:02 9 February 2009 (UTC).

Wikipedia Signpost, February 8, 2009

The Wikipedia Signpost Weekly Delivery

Volume 5, Issue 6	8 February 2009	About the Signpost

News and notes: Elections, licensing update, and more	Wikipedia in the news: Wikipedia's future, WikiDashboard, and "wiki-snobs"
Dispatches: April Fools 2009 mainpage	WikiProject Report: WikiProject Music
Features and admins	The Report on Lengthy Litigation

You are receiving this message because you have signed up for the Signpost spamlist. If you wish to stop receiving these messages, simply remove your name from the list.--ragesoss (talk) 15:35, 9 February 2009 (UTC)

Delivered by §hepBot (Disable) at 22:48, 9 February 2009 (UTC)

Category:Articles containing Choctaw language text

I notice that this category you created is unpopulated (empty). In other words, no Wikipedia pages belong to it. If it remains unpopulated for four days, it may be deleted, without discussion, in accordance with Wikipedia:Criteria for speedy deletion#C1. I'm notifying you in case you wish to (re-)populate it by adding [[Category:Articles containing Choctaw language text]] to articles/categories that belong in it.

I blanked the category page. This will not, in itself, cause the category to be deleted. It serves to document (in the page history) that the category was empty at the time of blanking and also to alert other watchers that the category is in jeopardy. You are welcome to revert the blanking if you wish. However, doing so will not prevent deletion if the category remains empty.

If you created the category in error, or it is no longer needed, you can speed up the deletion process by tagging it with {{db-author}}.

--Stepheng3 (talk) 21:34, 10 February 2009 (UTC)

Same thing with Category:Articles containing Northern Sami language text. --Stepheng3 (talk) 21:35, 10 February 2009 (UTC)

Smackbot error

Hi, your bot didnt make this edit [4] properly, oddly it moved the wikilinks needing disambiguation out of the correctly location and to the top of the article. Russeasby (talk) 01:19, 11 February 2009 (UTC)

HI, this has happened on another article as well. I have not been able to duplicate this error yet. Rich Farmbrough, 17:51 11 February 2009 (UTC).

Problem is fixed in latest build of WP:AWB. Rich Farmbrough, 08:15 14 February 2009 (UTC).

"dead link" vs "Dead link"

I noticed that Smackbot is changing "dead link" tags to "Dead link" ones. The Wikipedia:Dead_external_links even uses "dead link" for its example usage. Also most Wikipedia tags seem to start with a lowercase letter. Is there a reason why using "Dead link" is worth changing? Jason Quinn (talk) 16:44, 9 February 2009 (UTC)

Answered on user's talk page. Rich Farmbrough, 18:46 9 February 2009 (UTC).

But is there a compelling reason to capitalize these tags at all? Your answer to Jason Quinn (talk) said, tags "should only be changed when a template is being dated or other fixes are being made." However, SmartBot changes case for all tags that it tests when it makes a single fix to an article.

It's true that some template names are capitalized, but the tags in the template descriptions themselves are lower case (or not... arbitrarily?), e.g., {{fact}} in Template:Fact, but {{dead link}} in Template:dead link (note that "dead" is lower case in the template name).

Template descriptions aren't internally consistent. Template:Dubious can't decide which way to go (my emphasis):

Add {{dubious}} after a specific statement or alleged fact that is subject to dispute [...] To refer to a specific talk page section, use {{Dubious|talk page section name}}

I suggest that SmartBot not make arbitrary changes. Instead, SmartBot should be dumb and leave the case of tags as they are. --Mtd2006 (talk) 15:54, 13 February 2009 (UTC)

"Template:dead link (note that "dead" is lower case in the template name). " If you have popups you will see that this is automatically redirected to Template:Dead link.

All pages on WP begin with a cpaital. I have chosen to use, as I must chose one or the other, the captialised version. It's not a big deal either way, lets move on from this one. Rich Farmbrough, 08:25 14 February 2009 (UTC).

MMA Hacks

check out Jason Guida and Mongkhon Wiwasuk and tell me WTF is going on. there are like 40 to 50 other pages with this problem.User:Sepulwiki 23:50, 11 February 2009 (UTC)

There is a show/hide on the infobox? Rich Farmbrough, 00:09 12 February 2009 (UTC).

you dont see where it says headlinnig text over and over and over and it also says youtube sucks next to the dates? please tell me you see that.Sepulwiki 00:16, 12 February 2009 (UTC)

AWB bugs

Hi Rich, I'm working through the backlog of outstanding AWB bugs and was wondering whether you could provide some more information about one you reported a few months ago: link simplification too greedy. I could do with some examples of incorrect edits and related correct ones to understand what the problem is here. Thanks in advance Rjwilmsi 20:08, 13 February 2009 (UTC)

Answered on AWB pages. Rich Farmbrough, 04:30 18 February 2009 (UTC).

SmackBot and articleissues

SmartBot is changing article=y to article=February 2009 in articleissues tags. Is that what it wants to do? See this diff. The date does no harm, so it's not an urgent problem. --Mtd2006 (talk) 21:24, 13 February 2009 (UTC)

Yes, it's stressing the point, it's autoamtable (I build the rules from the template most days), it avoids having an exception, and it can possibly be used in future for the difficult "expert" parameter. Rich Farmbrough, 15:40 16 February 2009 (UTC).

SmackBot bug report

This edit. The {{quotation}} template has a funny tendency to not line-break even in places where text normally would, so when working within a {{quotation}}, you need to manually put in a bunch of <br> tags. In this edit ten days ago, SmackBot removed those necessary <br> tags, leaving incorrect line-breaking in the quotation.

Just FYI. Thanks, and best wishes! --Steve (talk) 05:19, 16 February 2009 (UTC)

Thanks, not sure what to do about this one, I'll discuss witht e AWB people. ~~

Wikipedia Signpost — February 16, 2009

From the editor — A new leaf

Commons Picture of the Year
Picture of the Year 2008 begins voting

News and notes
Flagged Revisions, historical image discovery, and more

In the news
Political tiff, error repeated in press, predictions of doom

Dispatches
How busy was 2008?

WikiProject report
WikiProject Gaelic games

Discussion report
Discussion Reports And Miscellaneous Articulations

Features and admins
Approved this week

Technology report
Bugs, Repairs, and Internal Operational News

Arbitration report
The Report on Lengthy Litigation

You are receiving this message because you have signed up for the Signpost spamlist.
If you wish to stop receiving these messages, simply remove your name from the list.

Delivered by §hepBot (Disable) at 07:30, 16 February 2009 (UTC)

Smackbot needs adjustment.

Edits like this one, where no date tag was added, IMHO, should be avoided. Smackbot is now forced to edit every single article. Can you adjust this in order Smackbot is not case sensitive? -- Magioladitis (talk) 00:58, 15 February 2009 (UTC)

"Date" is not a valid parameter name for the template. Rich Farmbrough, 15:18 16 February 2009 (UTC).

Ok thanks. You are right. My mistake. -- Magioladitis (talk) 15:55, 16 February 2009 (UTC)

Minor SmackBot bug

The bot tagged a page as needing to be wikified when it was already tagged as such: http://en.wikipedia.org/w/index.php?title=IberoAmerican_Federation_of_Mutual_Funds&diff=271272186&oldid=271269318 --Sigma 7 (talk) 04:43, 17 February 2009 (UTC)

See below. Rich Farmbrough, 04:31 18 February 2009 (UTC).

SmackBot

Please ask the bot why it removed this orphan tag. — RHaworth (Talk | contribs) 04:57, 17 February 2009 (UTC)

SmackBot removed an orphan tag from Stealth conservative as well. Шизомби (talk) 16:03, 17 February 2009 (UTC)

This seems to be a bug in the AWB tagging I will ask for it to be investigated and meanwhile disable it. Rich Farmbrough, 04:29 18 February 2009 (UTC).

Template:Expert

I think that your (very) recent change to this template has created an error. E.g. the article Hamza Marri shows an "ifexist" beneath this template which seems to be caused by this template. Fram (talk) 13:51, 18 February 2009 (UTC)

Thank you for the message

(moved from User:Mtd2006 User talk:Mtd2006)

SB is explicitly coded to understand other template arameters, even the non-parameter "reason". I will investigate the cause of this bug forthwith. Rich Farmbrough 04:23 18 February 2009 (UTC).

I admit to frustration with some of SmackBot's automatic edits. I make mistakes and it's nice that SmackBot fixes them for me. But we need to get SmackBot under control. I'd like to discuss the problem further. Mtd2006 (talk) 04:45, 18 February 2009 (UTC)

Actually as far as the SmackBot specific components are concerned, I have more or less finished transitioning between a very complex rule-base built up over several years, and one that is built automatically from time to time, and is hence more consistent and flexible. While this has given rise to a few new problems it has already proved far more maintainable than the old system, a good example is that on Monday someone created 60 new cleanup templates, I was able to add these in a matter of minutes on Tuesday. Moreover the system now keeps track of redirects, once a major headache to retrieve and code individually, and individual template parameter lists. The specific removal of a carriage return you mention is, perhaps, not ideal, but it would require another exception for the {{Article issues}} template. The plan is to treat all cleanup templates in as near the same way as possible. Rich Farmbrough, 08:14 18 February 2009 (UTC).

I understand the complexity of the problem -- better now that you've explained the method you're using. I have a concern about SB's return to operation. SB is complex. It needs testing. More than testing, however, SB must fail-safe. When it encounters an unexpected condition, it must do no harm -- as it is, SB fails the harmless test. SB needs to be more bullet-proof before it's restarted.

I placed a message on admin notices. That may be the wrong place. Please look at it. Move it or remove it -- okay by me. I've looked at SB's design spec (its tasks) the history of SB's bug list and I have some suggestions to make SB more bullet-proof. N.B. I'm Perl-savvy. Reply here or by e-mail. Mtd2006 (talk) 09:01, 18 February 2009 (UTC)

• I moved it to Bot Owners noticeboard.
• The bug you mentioned is fixed.
• "Doing no harm" is only possible by doing nothing. However SB is reasonably conservative, more so now than ever, except that it is becoming more liberal in it's interpretation of date mis-speelings.
• My biggest problem is the rump of articles left over when a SB run is completed, which need dating manually (about 5000 when the new templates were introduced above, reduced to 4-500 now). Again the aim is to reduce these, but continuous changes to the wiki have made this hard up 'til now. I generally aim to clear the backlog by month end.
• Unsuprisingly I have a list of improvements in mind already.
• I look forward to hearing your suggestions.

Rich Farmbrough, 09:20 18 February 2009 (UTC).

Okay, I'll start on that right now. It's 3:30 AM here. For full disclosure, I've been seeding test cases for SB to get an idea how the bot works. I've used degenerate cases so that SB finds unexpected conditions.

Doing nothing is necessary when an unexpected event occurs. That's fail-safe. If something unexpected happens, do nothing, and the bot has done no harm.

I've looked at the template space... the ruleset for SB. It's complex and inconsistent -- not ideal for automated processing. The "dummy" parameters must cause havoc. I imagine the exception cases are becoming a maintenance headache. What you've done with SB is quite an accomplishment.

Is SB coded in Perl? I'll use Perl to explain some fail-safe tests if that works for you. My followup in e-mail. Regards, Mtd2006 (talk) 10:38, 18 February 2009 (UTC)

It's Noon here and I've not gone not bed yet...

SB relies on AWB to make the actual changes to the pages, and the way it works is this:

I have a local list of templates, a perl program reads this list, invokes wget to recover the source of the page and redirects, analyses the html to extract the redirect lists and parameters, then builds the rule base from a bunch of XML fragment files plus some ancillary lists (plausible mis-spellings of January, for example). This rulebase is loaded into AWB and run against a 1 level recursive category listing of the dated cleanup categories category. Rich Farmbrough, 12:14 18 February 2009 (UTC).

My suggestions for changes to SmackBot are irrelevant. I didn't understand SmackBot's design or how AWB works. That's more clear now. The bot owners noticeboard is the best place to continue. Thank you for your explanation. Mtd2006 (talk) 22:58, 18 February 2009 (UTC)

SmackBot and templates

Is it necessary that your bot makes edits like this? It appears that the only thing changed was one space after the template name. It seems like a pointless edit. It worked fine with the space there.--Rockfang (talk) 16:50, 18 February 2009 (UTC)

I have raised the same concern at the WP:BON thread. Are you not able to instruct the bot not to make insignificant changes? –xeno (talk) 19:46, 18 February 2009 (UTC)

Actually it was a testing edit, I was hand testing some functionality, the bot won't genreally save white-space only changes. Rich Farmbrough, 19:48 18 February 2009 (UTC).

And caps changing? {{deadlink}} to {{Deadlink}} ? –xeno (talk) 19:50, 18 February 2009 (UTC)

That is more subtle, AWB has a feature to avoid saving case changing only, but sometimes the fix is to change the case of a parameter name. See the FAQ for why this should be rare, and a dozen reasons it can happen. Rich Farmbrough, 19:54 18 February 2009 (UTC).

Or a parameter value, for example "may 2008" => "May 2008". Rich Farmbrough, 19:55 18 February 2009 (UTC).

k, cool. thanks for your time. –xeno (talk) 19:57, 18 February 2009 (UTC)

Thanks for the info.--Rockfang (talk) 20:05, 18 February 2009 (UTC)

Hello, Rich Farmbrough. You have new messages at Debresser's talk page.
You can remove this notice at any time by removing the {{Talkback}} or {{Tb}} template.

Information for Country Outlines--How?

Hello there! User:The Transhumanist tells me that you have used a bot in the past to pull data from websites and put them in country outlines. Well, I'm currently running a bot for this project, and I'm wondering, do you think you could tell me how you did that? And if you're the open-source kind of guy, maybe give me the source codes? I understand that you are probably very busy, but if you could enable me to do this work that you once did, our project would greatly appreciate it. Hope to hear from you soon! Robert Skyhawk ^{So sue me!} (You'll lose) 21:44, 22 February 2009 (UTC)

Rich pulled data from one page on Wikipedia, to fill in blank entries on the country outlines on Wikipedia (for population and then for area). Rich, I too am very interested in how you did it. It's a mystery to me. I've been scratching my head ever since. The Transhumanist    21:56, 22 February 2009 (UTC)

SmackBot

I have noticed a Strange edit by the bot. Please check Alex Bakharev (talk) 23:49, 23 February 2009 (UTC)

Thanks, fixed.Rich Farmbrough, 05:22 24 February 2009 (UTC).

Hi. Possibly related to the above is this edit. Not sure what the bot was trying to do... GDallimore (Talk) 10:10, 24 February 2009 (UTC)

Please! Help with the CNN article

Hi, I am fairly new to Wikipedia. However, in looking at the CNN article, I noticed that in the beginning (fourth paragraph down) the following statement appears: "Among the general public, CNN has a more centrist perception.” The source for this proposition is FAIR, FAIR being a “progressive media criticism organization” whose “target invariably is bias on the right” [5]. Needless to say, FAIR is not a reliable source. On the other hand, the academic literature (i.e., peer-reviewed, scholarly journals) is full of articles addressing CNN's bias. They have concluded that the public perceives CNN to have a liberal bias:

-->Joan Shorenstein Center on the Press, Politics and Public Policy, Harvard University, & Project for Excellence in Journalism, “The Invisible Primary – Invisible No Longer: A First Look at Coverage of the 2008 Presidential Campaign,” p. 32 [6] (noting that “The CNN programming studied tended to cast a negative light on Republican candidates—by a margin of three-to-one. Four-in-ten stories (41%) were clearly negative while just 14% were positive and 46% were neutral. The network provided negative coverage of all three main candidates with McCain fairing the worst (63% negative) and Romney fairing a little better than the others only because a majority of his coverage was neutral.”)

-->Jeffrey N. Weatherly, et al, “Perceptions of Political Bias in the Headlines of Two Major News Organizations,” Harvard International Journal of Press/Politics (2007) (12), 91 at p. 97 (finding that “Overall, participants . . . rated the headlines from CNN as significantly more liberal than those from FOX News. . . . CNN tilted more to the left than did FOX News. The conclusion that CNN’s headlines were perceived as more liberal than FOX News’, rather than FOX News’ headlines’ being perceived as more conservative than CNN’s, is supported by the finding that overall, the headlines from FOX News were rated slightly on the liberal side of neutral. If FOX News’ headlines had been perceived as biased toward the conservative viewpoint, then the overall ratings should have exceeded 50 percent.”)

The administrator who semi-protected this article knew, or was reckless in not knowing, that (1) FAIR is a blatently unreliable source, and - equally important - (2) actual,peer-reviewed studies suggest precisely the opposite of the proposition that CNN is perceived as "centrist." Would you please help in removing the unreliably-sourced, and inconsistent-with-real-literature statement concerning CNN's perceived "centrist" position (it's semi-protected)? I would also like some help in getting the aforementioned articles detailing CNN's perceived liberal bias included in the CNN article. Thank you so much for your time. Jm131284 (talk) 04:33, 24 February 2009 (UTC)

Thanks for your note:

1. Protection and semi-protection do not imply approval of the version protected.
2. semi protected articles are still editable by registered users of more than a few days standing.
3. The best place to discuss your concerns is on the talk page of the article involved. I have little knowledge of the subject, the regular editors of the page will be better informed.

Rich Farmbrough, 05:42 24 February 2009 (UTC).

Wikipedia Signpost — February 23, 2009

This week, the Wikipedia Signpost published volume 5, issue 8, which includes these articles:

• Philosophers analyze Wikipedia as a knowledge source
• An automated article monitoring system for WikiProjects
• News and notes: Wikimania, usability, picture contest, milestones
• Wikipedia in the news: Lessons for Brits, patent citations
• Dispatches: Hundredth Featured sound approaches
• Wikiproject report: WikiProject Islam
• Discussion Reports And Miscellaneous Articulations
• Features and admins: Approved this week
• Technology report: Bugs, Repairs, and Internal Operational News
• Arbitration report: The Report on Lengthy Litigation

The kinks are still being worked out in a new design for these Signpost deliveries, and we apologize for the plain format for this week.

Delivered by §hepBot (Disable) at 21:40, 24 February 2009 (UTC)

OR>original research?

I noticed on the meme page that your bot swapped out the Template:Or to Template:Original research with this edit (while adding the date like usual). As you can see from the template pages they are not the same thing, so that might be a problem. Vicarious (talk) 22:30, 25 February 2009 (UTC)

Yes thanks. this was because OR and Or were different I have fixed them up since then. Rich Farmbrough, 06:36 26 February 2009 (UTC).

Featured article review

I have nominated British African-Caribbean community for a featured article review here. Please join the discussion on whether this article meets featured article criteria. Articles are typically reviewed for two weeks. If substantial concerns are not addressed during the review period, the article will be moved to the Featured Article Removal Candidates list for a further period, where editors may declare "Keep" or "Remove" the article's featured status. The instructions for the review process are here. Cordless Larry (talk) 07:52, 26 February 2009 (UTC)

Why did you delete my Pop Culture reference?

(SmackBot)

Hello, I added a Simpsons reference to the Pop culture reference part of the page - why did you delete it? What is more pop culture than the Simpsons?

Judy

What page would that be? Rich Farmbrough, 05:46 5 February 2009 (UTC).

Backlog Tag on Category:Biography articles without listas parameter

Really? I have been dropping hints all over the place in the hope that some folks might jump into the fray. I hope your tag does some good.

Just so we don't bump into each other you should know that I am working from the back. I have cleared all the articles after Z and have started on the Z articles. I believe that there is a bot that is going through the articles and copying the DEFAULTSORT value from the article into the listas parameter on the Talk page but I do not know the scheme.

Glad to have you aboard!

JimCubb (talk) 06:27, 27 February 2009 (UTC)

Smackbot having some trouble with ISO dates

See [7], [8], and [9]; Smackbot seems to have grabbed the wrong number for the year. I glanced at his other edits and didn't see that happening to anyone else's files, so I assume I just had my dates formatted in a way the bot wasn't programmed to handle? rʨanaɢ ^talk/_contribs 05:00, 28 February 2009 (UTC)

Yes, thanks. Sorted. Rich Farmbrough, 08:25 28 February 2009 (UTC).

There are still issues: Jan 2006 instead of Jan 2007 (not sure this is wrong, as it is the creation date; the other edits don’t seem to have any real source for the date), Oct 2006 instead of Oct 2005, Oct 2006 instead of Feb 2006. --AVRS (talk) 18:11, 28 February 2009 (UTC)

Yes the first one simply goes from the date on the page. The other two go from the date the page was added to the category: this is recovered from the meta-data using the API. There is a substantial "bulge" in October 2006, I assume that is when the category was added to the template. Rich Farmbrough, 18:21 28 February 2009 (UTC).