Wikipedia talk:WikiProject Articles for creation/October - November 2012 Backlog Elimination Drive

Referencing MOS

The project page recommends looking at areas in the manual of style. While, granted, you'll need to refer to this to get an article above stub or start class, or to stop a {{wikify}} template being slapped on it, I would not generally fail an article if it has MOS issues, but otherwise can pass WP:GNG. With that in mind, is referring to the MOS necessary or desired? --Ritchie333 ^{(talk) (cont)} 13:37, 18 October 2012 (UTC)

I included those articles so that if reviewers find problems with those sections they can notify the nominator, kind of like, "I'm passing the article but here are some little things you may want to improve on". The submission doesn't necessarily have to be failed.--Dom497 (talk) 18:33, 18 October 2012 (UTC)

Barnstar

The teamwork barnstar is the best looking one! Perhaps we need more variety of appearances to motivate people like me. Graeme Bartlett (talk) 09:28, 22 October 2012 (UTC)

I have made this - not sure how much use it is!! (Example for brownie :) )

{{User:Mdann52/sandbox|File:Brownie transparent.png|5}}

Mdann52 (talk) 16:21, 22 October 2012 (UTC)

Fastest way to record for score?

Does someone have a faster way to do the score totals? It took me literally 65 minutes this morning to enter in 160-some AFC reviews I did last night. What I'm doing is opening up my Contribs, selecting for just "WikiTalk", then hitting each diff, copying the URL, pasting into the edit window, copying the name, pasting into edit window, and then I save every 30 entries or so. Sure, that's 25 seconds or so per entry, but that really adds up and ironically cuts into my Reviewing time.

Any faster way to knock out the scoring? MatthewVanitas (talk) 15:12, 22 October 2012 (UTC)

• I have a notepad window open in my second monitor. When I've reviewed something I get the diff, write the article name, and then bulk copy and paste the whole lot at the end. It probably takes the same amount of time overall, but it's less noticeable because its integrated with the reviewing time. --Ritchie333 ^{(talk) (cont)} 16:26, 22 October 2012 (UTC)
• If everyone uses AFCH (or even a standardized edit summary), it would be trivial to write a bot/script to find them. Legoktm (talk) 19:22, 22 October 2012 (UTC)

I do use AFCH (the little menu thing by Favourites and Move where I pick "Review", yes?), so if there is some easy way that'd be great, so long as it's not challenging for someone to program up. MatthewVanitas (talk) 17:38, 23 October 2012 (UTC)

The other option is to not review so many articles. :-p   -- :- ) Don 17:54, 23 October 2012 (UTC)

I know this section is a tad stale, but I think programing and automatic scoring system into AFCH wouldn't be that hard. At the moment I simply manually copy-paste each diff into my file after each review. --Nathan2055^{talk - contribs} 18:51, 8 November 2012 (UTC)

For me at least, this has been a big impediment. I have nearly 1000 reviews (still have about 200 I just don't feel like tallying right now), and I've blown easily 7 hours or more making this silly list for score, so that's several hundred reviews I could have done if I didn't have to manually make a list. Not trying to be unpleasant here, but I honestly don't think I would compete in a Drive for AFC in the future, or any similar score-based drive with such high scores, unless the scoring is somehow automated. Manually counting a dozen Good Article reviews is no hassle, spending an hour to tally 130 reviews is a real pain. MatthewVanitas (talk) 00:01, 18 November 2012 (UTC)

IMO, I feel if we can get our hands on the tool used to add up RFA totals, that may be better for next time we do this. Mdann52 (talk) 08:50, 18 November 2012 (UTC).

Collapsible tables?

Since this page is going to get long very quickly it might be a good idea to use collapsible tables for the totals lists such as those used by WikiProject Wikify for their drives. France3470 (talk) 15:20, 22 October 2012 (UTC)

• I agree, plus the reviews really should be sorted in alphabetical order of reviewer. --Ritchie333 ^{(talk) (cont)} 16:27, 22 October 2012 (UTC)
  • I've had a go at this, but it still looks a bit fugly. If anyone else can do a better job, please do! --Ritchie333 ^{(talk) (cont)} 17:12, 22 October 2012 (UTC)

Might be best to just use the Template:GOCE Article list, with two columns. I gave it a go on mine and it works quite well. The counter is handy too. France3470 (talk) 20:50, 22 October 2012 (UTC)

• The collapsible tables break the notes at the bottom, if you are trying to click through all the copyvios for example. Legoktm (talk) 21:24, 22 October 2012 (UTC)

Just tried it now and it seems to work for me. Perhaps I fixed it? Ideally it would be nice to have the text smaller, but I can't seem to make anything work. Any thoughts? France3470 (talk) 22:01, 22 October 2012 (UTC)

Not working for me. I'm talking about if all the tables are collapsed (like how they are by default), and I hit the a or b next to the first copyvio note, it doesn't do anything. Is it just me? Legoktm (talk) 22:23, 22 October 2012 (UTC)

You're talking about clicking the link in the footnote back into the table? I don't think I have ever seen that work in a collapsed table. I don't know how it could work without each link being scripted to initiate the expansion, but I'm no JS expert. -- :- ) Don 23:08, 22 October 2012 (UTC)

If you expand the table, it works. I guess when the table is collapsed, the anchor is invisible. -- :- ) Don 23:10, 22 October 2012 (UTC)

I wasn't thinking of it that way; I guess I tend to click on the number to take me to the footnote not the other way around. Is this particularly problematic? I would have thought this was minor in comparison to the foreseeable issues with trying to navigate a huge page. France3470 (talk) 23:36, 22 October 2012 (UTC)

Since there really is no good way to make collapsible tables and footnotes work, The "IRC cabal" (all 2 of us) has come up with a solution: a template! Simply use {{afc review|<reason>}} next to whatever you review. Pretty easy. Use it if you want, don't if you don't like it. Legoktm (talk) 23:52, 22 October 2012 (UTC)

Protocol Independent Links

Hi everyone, Please leave links in the following format:

[//en.wikipedia.org...] <-- USE THIS

[http://en.wikipedia.org] <-- NOT THIS

Doing so will have the software pick whether to use http or https based on what the user is currently using. I've gone ahead and done so for the current page, but would prefer not to keep doing it for the month. Thanks, Legoktm (talk) 19:20, 22 October 2012 (UTC)

For those who use a text document to make a list of reviews, I made a little Perl script to make this easier. It's probably not the most effective way to do it in Perl, but I'm still learning it, and it works, so hah! Wywin (talk) 15:04, 25 October 2012 (UTC)

What on earth is happening?

Can someone explain to me why we ended up having 3 different pass/fail icons on the totals section. It looks like such a mess!--Dom497 (talk) 00:42, 23 October 2012 (UTC)

Different templates, different personal opinions? I justified my creation of a new template in the section right above this one, but with the disclaimer that if people didn't want it, not to use it. Legoktm (talk) 01:15, 23 October 2012 (UTC)

@Dom, Good point. Legoktm pulled me unwilling into his Cabal and made me agree it was a good idea...   Just kidding. The icons are nice, but consistency is good also. Shall we look for a concensus? It's Dom's idea and work, so he should have the last word IMHO. The response appears to be excellent. Appears because we have not really been tracking how hard AfC has been working. -- :- ) Don 01:26, 23 October 2012 (UTC)

Proposal on failing posts

Proposal: If you give a fail vote, a reason should be provided. The reason can be provided in an afc comment, a reference, on the person's talk page... As long as some brief comment is provided which explains why a fail vote was given, which will allow the editor to learn for next time and to improve. These instructions will be included in our section on Reviews.

Thoughts? --Jethro B 00:50, 23 October 2012 (UTC)

• Support as nom --Jethro B 00:59, 23 October 2012 (UTC)
• Support Go Phightins! 00:52, 23 October 2012 (UTC)

Really? WP:AGF first off. WP:POLL secondly. If someone makes a mistake, leave them a note on their talk page. No big deal. Legoktm (talk) 01:17, 23 October 2012 (UTC)

Let me just clarify my position. I believe that this is a good idea so reviewers can improve, not because I support a conspiracy theory. I am not endorsing the conspiracy theory that Jethro is alleging. Go Phightins! 01:18, 23 October 2012 (UTC)

I'm not alleging any conspiracy, my main reason is in order to help other editors but also to ensure that this program isn't manipulated. It's a precautionary measure. WP:AGF doesn't have anything to do with this, if you actually looked at the diff and thought it failed, you should be able to just briefly explain why it failed so the editor knows for next time, and knows you looked at it. Not everyone is going to leave a note on someone's talk page, that hasn't been what's happening. --Jethro B 01:29, 23 October 2012 (UTC)

Sure, so just leave them a talk page notice. Thats where you normally leave editors messages so they can improve. Legoktm (talk) 01:24, 23 October 2012 (UTC)

That's acceptable, I just think there should be some sort of brief explanation for why, if nothing else, so the editor in question can improve as a reviewer. Go Phightins! 01:25, 23 October 2012 (UTC)

• (edit conflict) Neutral—I'd recommend that users elaborate on a contentious rationale in the form of comments, where the author and reviewer are equally able to access the information, rather than as footnotes in the project namespace. Mephistophelian (contact) 01:19, 23 October 2012 (UTC).
• Abstain - I think AfC is above that. Explanation is optional. -- :- ) Don 01:34, 23 October 2012 (UTC)

AFC is above telling users what they did wrong? Or above reckless conspiracies? Go Phightins! 01:36, 23 October 2012 (UTC)

Can you please stop accusing me of a "reckless conspiracy?" Have you considered that I'm actually trying to avoid a dillema and a bad-faith editor and improve the project, rather than promote reckless conspiracies? I find that term very offensive. --Jethro B 01:48, 23 October 2012 (UTC)

Sorry, I meant we should not be worrying about someone scamming the system, but I do thing the feedback is good, but not mandatory. -- :- ) Don 01:38, 23 October 2012 (UTC)

OK, I agree with that. I was a little confused. Thanks for the clarification. Go Phightins! 01:39, 23 October 2012 (UTC)

As a matter of fact, I have already created a conflict with feedback. Should we have a three-way, so to speak? -- :- ) Don 01:41, 23 October 2012 (UTC)

• Neutral - sometimes a footnote is a neat solution, sometimes an afc comment on the draft article is sufficient, sometimes an explanation on the reviewer's Talk page is best. As long as an explanation for the fail is made somewhere, that should be sufficient. Sionk (talk) 01:48, 23 October 2012 (UTC)
  • I don't mind this either, and seems to be the general sentiment, so I'm modifying my original post to just including a comment somewhere, whether a footnote, on the person's talk page, on the draft article... My proposal I made was just to ensure a reason was given somewhere, I didn't mean to be so specific. --Jethro B 02:00, 23 October 2012 (UTC)

All right, so I think consensus is that somewhere, reviewers of reviews are to leave a message somewhere to the original reviewer as to why. Go Phightins! 02:02, 23 October 2012 (UTC)

Proposal 2 on failing posts

Proposal - a Fail of a review requires a third opinion and then the majority decision will determine to whom the points are awarded. This should be changed ASAP because of the following rules:

1. Any user may review other user's reviews if they wish.
2. If you fail an article, please add a colon ( : ) in front of the number ( # ) symbol.

This automatically gives the second reviewer more credence than the first. -- :- ) Don 02:07, 23 October 2012 (UTC)

Good thought. I support it. Go Phightins! 02:10, 23 October 2012 (UTC)

 Done--Dom497 (talk) 19:05, 23 October 2012 (UTC)

Proposal: subpages

The page is getting very long and thus I might not be able to review (I will start at Thursday, maybe Friday) if the page is getting that long that my browser is getting instable. We should move the individual reports to subpages and only linking them here (so no transclusion!) with the stats on this page. mabdul 04:19, 23 October 2012 (UTC)

Oh yes please. I keep getting ec's everytime I try and edit anything on that page. Legoktm (talk) 04:22, 23 October 2012 (UTC)

 Partly done Up to Jethro B. I'm going offline now, so someone else needs to take over. Mdann52 (talk) 07:21, 23 October 2012 (UTC)

 Pretty much done Everyone has their own subpages now, some formatting on the main page might need to be fixed as well. Legoktm (talk) 08:05, 23 October 2012 (UTC)

Too successful?

Well, a few days in now, and we have cleared almost 2/5th of the backlog! Almost too much of a success then :) - Mdann52 (talk) 13:09, 23 October 2012 (UTC)

You cannot be too sucessful... only too fast. If the speed costs are bad reviews, then it is too fast. If not, then everything is OK.--Müdigkeit (talk) 13:13, 23 October 2012 (UTC)

Problems which need to be addressed

1. CSD'd articles - I found a user doing 2nd reviews on articles yesterday who was failing articles because they had been CSD'd. This is an additional problem which probably nothing can be done about at this point, if the 1st reviewer was wrong about the CSD. Since we are in speedy mode, this is more likely to happen.
2. User space articles in AfC - This problem has been discussed over and over in AfC with no concensus.
3. 2nd proposal above - This should be clarified ASAP. The longer we wait, the more the repair that will be necessary.

-- :- ) Don 15:21, 23 October 2012 (UTC)

We are not in speedy mode. Do not do this faster if it comes at a cost for accuracy. There is no need to do it very quick. One month is long, and checking all the facts that have to be checked takes its time.--Müdigkeit (talk) 17:35, 23 October 2012 (UTC)

By definition, the speedy mode can not be helped when it is a Drive/Contest, but I believe that is compensated for by most, if not all articles, being review twice, which is the exception in normal mode. Except, for the CSD and conversely approval of articles. -- :- ) Don 17:59, 23 October 2012 (UTC)

Who said that once the drive ends everything must be in full order. If there are still issues after the drive ends, no barnstars will be given out until they are resolved. In the end, there is no rush as we have as much time as we need.--Dom497 (talk) 18:28, 23 October 2012 (UTC)

3rd issue has been resolved.--Dom497 (talk) 19:06, 23 October 2012 (UTC)

There are currently four of Eclipsed's reviews relating to copyright violations that require a majority decision. Mephistophelian (contact) 22:34, 23 October 2012 (UTC).

If there are any similar articles involved in the competition, please check what links to the namespace to verify whether the administrator's rationale for deletion corresponds to the reviewer's decision. Mephistophelian (contact) 23:45, 23 October 2012 (UTC).

AFC stats? Growth rate?

We have several folks on board, and the number is certainly lower, but we get new AFCs every hour. Is there somewhere with statistics like "average number of new AFC articles per day"? Also would be interesting to see stats on how many articles are Declined for what rationales; I wish there were some auto-decline for purely blank articles, as that would knock out 5-10% of the AFC workload automatically. MatthewVanitas (talk) 01:35, 25 October 2012 (UTC)

We need to ask Earwig, but he already providing the stat how many submissions are sent to AfC, accessible at the end of WP:AFC/S. mabdul 03:51, 25 October 2012 (UTC)

Ah, I see it, had to search around a bit to see what I was looking for. This link Category:AfC submissions by date/September 2012 is a good example. Apparently the average number of submissions per day was 219.5 for September 2012. Busy project, this one. MatthewVanitas (talk) 19:36, 25 October 2012 (UTC)

Reviewing Reviews

I've noticed we've had a significant slowdown in reviewing reviews...not sure what this can be attributed to, but when reviews aren't being reviewed, my unscientific opinion would be that quality goes down... Go Phightins! 19:09, 30 October 2012 (UTC)

I have an hour lunch break now, so I'll do what I can.... Mdann52 (talk) 13:17, 31 October 2012 (UTC)

Subject of my article is...

Moved to WT:AFC

Mdann52 (talk) 11:04, 2 November 2012 (UTC)

Backlog elimination..

AfC submissions Random submission

~7 weeks

1,365 pending submissions to update

The backlog is growing during the elimination drive.--Müdigkeit (talk) 04:59, 5 November 2012 (UTC)

It's still lower than what it used to be before the drive, which was anywhere from 600 to 1000. Legoktm (talk) 06:00, 5 November 2012 (UTC)

And... what? Dont you see the problem? If the backlog is growing during an elimination drive, what do you think will happen when it ends? --Müdigkeit (talk) 06:50, 5 November 2012 (UTC)

To be fair, we did get it to under 200 a few days ago. So I'd say a small amount of growth is fine, seeing as it was around 1100 when we started. If the growth continues after a longer period, then we might need to put in some extra effort. But I don't think it's that big of a deal now. CharmlessCoin (talk) 07:11, 5 November 2012 (UTC)

Well any increase in cleared submissions is great, actually getting this down and keeping this down is going to require some effort. As I note above, for September 2012 AFC averaged about 219 submissions per day. To keep that down we'd need, for example, a body of 20 people doing at least 10 articles a day. The Drive now has, what, 35 people, some of whom haven't done 10 total yet (not that they're obliged to, just noting). To actually get it down and hold it down on all month we either need more people, or folks who want to knock out a whole ton of reviews. It's certainly not wasted effort, we've knocked out a lot, but fundamentally we need more people in AFC, especially as a large amount (even before the Drive) of the reviews are done by a very small handful of familiar names I see all over AFC. To keep AFC clear on a monthly basis at current rate, we'd need a good 20 people willing to average 10 reviews a day for the month. MatthewVanitas (talk) 22:12, 5 November 2012 (UTC)

Time to do something about that. Does anyone have an idea?--Müdigkeit (talk) 22:31, 5 November 2012 (UTC)

If possible from a technical perspective:

1. Develop a script that detects any submissions that are blank or lacking references, perhaps filtering them automatically or attaching notices for the author,
2. Integrate the 'Pending AFC Submissions' category into the new pages feed to increase the visibility of those prospective articles awaiting review. Mephistophelian (contact) 23:59, 5 November 2012 (UTC).

Both good suggestions, I think #1 is already under discussion at WT:AFC#Subject_of_my_article_is..._2, though I don't think anyone in a position to address it has weighed in. #2 is a really interesting idea; I had been pondering if there was some way to get some help from NPP without actually sucking away their manpower, and including AFC in their list might give them a chance to head off issues before they become articles. Part of that though would be educating folks about AFC Helper Script; I can't believe I spent a week or two manually reviewing articles since I thought it'd be too much hassle to turn on the Script, when it's literally one check-box and quite intuitive.

Automatically screening out blanks would probably cut on 19 of the daily 219 AFC submissions; cutting out articles without refs might cut out easily 50+ more.

I've also pushed for some kind of control to force folks to move to WT:AFC when they add a submit template, because a bunch of articles entitled "User:JoeSmith/sandbox" just adds hassle, especially for folks who like to review specific kinds of articles (bands, bios from specific countries, websites, etc). MatthewVanitas (talk) 01:39, 6 November 2012 (UTC)

I think the second option could be our best bet for keeping the backlog down consistently. Educating about the script couldn't be that hard, as you said it's fairly simple to turn on and use. CharmlessCoin (talk) 04:11, 6 November 2012 (UTC)

I don't know if this helps, as I'm fairly new to AfC, but I'm planning on working regularly here, so just wanted to shed some positive light to the issue of the fairly constant backlog. You can count on me to contribute quite a bit :) FoCuSandLeArN (talk) 20:51, 6 November 2012 (UTC)

It may be a little late for this suggestion, but I was wondering if it would help reduce backlog if a backlog elimination drive invitation was sent to, say, every user that has made more than 40 edits. I actually didn't get an invitation, I found out about the drive after looking for my AFC that I had submitted. I was just thinking, you could get a lot more users who make solitary edits that no one knows about to hear about the drive. B. Jakob T. (talk) 00:18, 7 November 2012 (UTC)

(Clear throat) Something has to be done. Look, we're almost back to where we started from!! I've tried my best, and I'll keep trying, but I think we need to get more people to help. There are only 17 of us that have reviewed over 30 AFC's. Not to mean that the others are required to review more. I calculated it out, and if everyone who signed up would review 20 more articles, we could knock this backlog. And if more people signed up and revied AFC's, we'd be done even quicker. B. Jakob T. (talk) 23:46, 7 November 2012 (UTC)

We are nowhere near back where we started from, which is around 1200 pages (we're 495 at the moment). Plus we had articles at the top the queue that'd been waiting over a month, now maybe a week or two. Still, improvement needed. For my part, I went and posted the recruiting banner and a brief personal message (including mentioning the highly useful AFC Helper Script) at WikiProjects India, Film, Bio, and Music, since we see a lot of incoming content from their areas. I didn't see this drive widely advertised, only came across it because I was doing AFC anyway for a few weeks, so maybe posting in a few more targeted wikiprojects would help. Folks this late might not be as competitive for #1-3 prizes, but still plenty of barnstars for all. MatthewVanitas (talk) 17:54, 8 November 2012 (UTC)

What I've noticed now when looking at AfC submissions to review is that people have been picking off the easy ones - unreferenced, only referenced to themselves, Facebook or Twitter, blatant lack of NPOV, test edits, etc etc - which can be declined very quickly. The remaining ones require diligence to check the references and content, with knowledge of the subject matter being beneficial. So in that respect the drive has been of some use in ensuring the ones left in the queue are those that have a chance of passing. --Ritchie333 ^{(talk) (cont)} 10:18, 13 November 2012 (UTC)

Totally agree, the Drive is particularly good at clearing out the chaff. I've done a few of the more serious ones when they were in my field, but for things I don't understand well, and/or am not familiar with what's WP:RS or no, I don't mess with. If the chaff is addressed effectively, editors doing serious review will be finding a much higher proportion of decent/workable articles by the time they get to the front of the queue. As noted before, the chaff-clearing would be easier if we had auto-reject for blank articles (and maybe even for articles with no apparent references?), and if we had a way to force submitted articles into AFC so we don't have a plethora of "JoeBob/sandbox" entries. MatthewVanitas (talk) 16:11, 13 November 2012 (UTC)

No, don't do that! You'd put me out of work. :) I only review "winners" e.g. AFC's that are tests or have no references. I am usually at the end of the list, purging the page every few minutes for the newest AFC's which I'm hoping are "winners". But I do agree with what you said about the sandbox entries. It's a pain in the neck! B. Jakob T. (talk) 21:22, 13 November 2012 (UTC)

That's incredibly BITE-y. You should be reviewing the oldest submissions, not reviewing ones that were just submitted and not fully finished yet. Legoktm (talk) 22:43, 14 November 2012 (UTC)

If it's not finished, the editor shouldn't submit it. While there's a little "gaming the game" in picking up the freshest and worst-looking articles (ones with suspect titles, ALL CAPS in titles, etc), those do need to be reviewed at some point or another, and there's no obligation on any one editor to review the oldest first. Plus, since this is a backlog drive, targeting the obvious Declines is the fast way to reduce the backlog. EDIT: It hasn't been my experience that so many folks submit and keep improving while they're waiting (which is kind of like cutting in line anyway); we have more the opposite trouble with folks who get a decline and clear guidance, spend all of 30 seconds fixing 5% fo the problems and immediately re-submit. MatthewVanitas (talk) 17:35, 15 November 2012 (UTC)

WP:RS is a strange old beast. It's only a guideline, and not even an important one when compared to WP:V and WP:NPOV, and you only need to take a trip to something like Wikipedia:Articles for deletion/Justin Bieber on Twitter to see how contentious it can really be. Which brings me onto the way I pass AFC articles - imagine the article's at WP:AfD, what would you !vote? If "Keep", pass the article. If "Delete", fail it. If "Speedy Delete", fail and speedy it (though I tend to only give the speedy treatment to blatant drivel, attacks and copyvios). Simples. --Ritchie333 ^{(talk) (cont)} 21:59, 14 November 2012 (UTC)

That method of passing articles would ignore all but only the most egregious cases of promotional articles that already fall under CSD G11. --Odie5533 (talk) 13:03, 18 November 2012 (UTC)

I don't understand what on earth you mean by that, but the method I described is simple and easy to explain. If you would pass an AfC submission correctly, it must be notable and hence would unquestionably survive an AfD.

Meanwhile, the drive has finished and the backlog is back up to 1000+. I do notice, however, there appear to be less quick fails on the queue now, so it might simply mean that AfC is just getting more and more popular, or our collective understanding of Wikipedia policies is increasing, so it might not be all bad. --Ritchie333 ^{(talk) (cont)} 15:01, 26 November 2012 (UTC)

Okay, let's get this over with: AfC is never getting into NPF. NPF is the WMF's attempt at trying to fix problems with the infamous WP:ACTRIAL proposal which would have solved all of these issues (deleting NPP by funneling all drafts through AfC, giving us more reviewers and faster work). I am a developer at AFCH, and I can say that I can technically solve all of the blank submission issues by writing a new article wizard that forces everyone to create drafts first. Even better would be redesigning the article wizard to use JavaScript and be more intuitive. NathanBot (talk · contribs) will also clean up submissions using the proposed AFCH-B framework. I'm trying the best I can, and suggestions are appreciated. --Nathan2055^{talk - contribs} 22:50, 26 November 2012 (UTC)