Wikipedia talk:Version 1.0 Editorial Team/Index

From Wikipedia, the free encyclopedia

This talk page is automatically archived by Werdnabot.
Any sections older than 60 days are automatically archived to Wikipedia talk:Version 1.0 Editorial Team/Index/Archive 5. Sections without timestamps are not archived. An archive index is available here.
Shortcut:
WT:1.0/I
Archive
Archives

Contents

[edit] Total number of articles tracked?

The current total of 701,787 is, I assume, the total number of articles with participating WikiProject tags on them (assessed or not). I know there have been updates on this and similar figures in the Wikipedia Signpost, but can anyone here provide a graph of how the total has varied over time? When is the figure likely to hit 1 million? Is it ever likely to catch up with the total number of articles (currently 1,698,947)? Carcharoth 11:46, 22 March 2007 (UTC)

If we could answer such questions we'd make a lot of money on the stock market. :) The bot never kept information about the total number of articles each day. That info is of course contained in the history of the page, but would be a pain to extract.
To keep it simple, this project is less than a year old I think. So perhaps in a year we'll cover all the articles (unless the bot stops working for reasons of scale, that is). Oleg Alexandrov (talk) 14:50, 22 March 2007 (UTC)
I (painfully ;)) extracted the numbers from the history page. I used 1 week intervals, here's the simple graph I came up with. Yeah, a year sounds like a good estimate. MahangaTalk to me 15:18, 22 March 2007 (UTC)
Ooh. Nice. Could you stick a green solid area under the blue bit to show total number of assessed articles? Carcharoth 18:15, 22 March 2007 (UTC)
I wonder if the total number of articles (that must have been graphed somewhere else already) can be put in as a black solid line up above this solid stuff? :-) Carcharoth 18:17, 22 March 2007 (UTC)
There's a lot of stuff on Wikipedia:Statistics, but no easy way to retrieve that information. I did add the total assessed to the previous graph. That's about the extent of my excel expertise. Hope you like. MahangaTalk to me 20:05, 22 March 2007 (UTC)
Very nice to see this, thanks! I can see the effect of Kingboyk's stubs script after August 2006, too! Thanks, Walkerma 21:29, 22 March 2007 (UTC)
Bah! It's not a script, it's a finely tuned piece of low level code! <cough> --kingboyk 22:01, 23 March 2007 (UTC)
I'd also like to think that last little surge over the past month is due to WPP:BIO's assessment drive.↔NMajdan•talk 19:11, 23 March 2007 (UTC)
Actually, NMajdan, I had thought the same thing, I think that assessment drive is having a big impact. Also, the huge surge last August came from WP:BIO joining. Walkerma 21:57, 23 March 2007 (UTC)

Next step is to program the bot to generate and update this graph itself. <ow! stop hitting me with that wet trout, Oleg!> :-) Carcharoth 22:53, 22 March 2007 (UTC)

Heh. Apparently I'm not the only one who does these kinds of things... Titoxd(?!? - cool stuff) 03:15, 23 March 2007 (UTC)
Awesome. :) Oleg Alexandrov (talk) 03:26, 23 March 2007 (UTC)

That's beautiful. Could you update it every month? Pretty please! --kingboyk 22:00, 23 March 2007 (UTC)

[edit] Update

Here is an updated version. --Arctic Gnome (talk • contribs) 17:12, 12 December 2007 (UTC)

I've now also added the total number of articles. What's interesting is that the number of untagged and unassessed articles stays about the same. We've been tagging and assessing at exactly the same rate that new articles are created. --Arctic Gnome (talk • contribs) 04:39, 13 December 2007 (UTC)

Thanks! I used the last one in Powerpoint presentations - it's very helpful to have this graphic. Walkerma (talk) 05:25, 13 December 2007 (UTC)
I find the graphic to be fairly disheartening. Those yellow and red regions haven't been getting any smaller. --Arctic Gnome (talk • contribs) 05:42, 14 December 2007 (UTC)

[edit] Adding list articles to the bot

Copied over from Wikipedia_talk:Version_1.0_Editorial_Team#NEW_FL_debate, this post by User:TonyTheTiger: I would like to start a discussion about how the quality logs handle article space only. I think we could probably all agree that the quality logs would be improved by adding t(w)o more classes (FL-class and List-class) to handle articlespace contributions. This conversation would avoid all the likely pitfalls of numerous arguments about how to handle all the other spaces. Please contribute to a discussion which will hopefully lead to something being done. Can we agree to simply add these two classes to the reports?

This seems like a reasonable proposal, but it would be a big change to the stats table for the many hundreds of projects using the bot currently - so it should be debated IMHO. I think it would be nice to be able to include FL-class and List-class tags - we even have some in WP:V0.5 and 0.7 that are "invisible" unless you look at the full listing. I have three questions:
  1. Would we simply record these in the logs and "by quality" worklists, and not put them in the stats tables? Or put them everywhere?
  2. Oleg, would this be a simple change in the code, and would you have time to do this? I know you're busy in a new job.
  3. Is there significant support from the WikiProjects for this?
Walkerma 03:54, 17 October 2007 (UTC)
Somewhat more generally: would it be possible to have the assessment tables omit any rows for which the corresponding categories do not exist, or which have no articles in them? If that were the case, then it would be possible to add support for optional classes without impacting the projects that did not use them. Kirill 03:56, 17 October 2007 (UTC)
I think what Kirill suggests, having the stats omit empty rows could be the simplest thing to do. I can implement it unless people disagree with adding FL-Class and List-Class. Oleg Alexandrov (talk) 04:16, 17 October 2007 (UTC)
  • Kirill hits the nail on the head here. Several projects have adopted a "FL = FA" criteria for assessment; that is, a Featured list would receive a {{FA-Class}} assessment. Additionally, as others and myself have pointed out in the past, lists are assessable, unlike templates, images, categories et al. As a result, I'd be opposed to adding a catch-all {{List-Class}} which would "steal" (for lack of a better word) assessments from the SSBA system. As for {{FL-Class}}, I have the impression that it would duplicate our pre-existing FA assessment class, but I would not be opposed to a merger of the two into {{Featured-Class}}. That said, these concerns would be made moot by removing unused rows/columns from WikiProject assessment tables. The question comes as to what to do with the global assessment tables, though. Titoxd(?!? - cool stuff) 04:36, 17 October 2007 (UTC)
I've already discussed there why lists are not assessable (at least not by current metrics for articles) and why I don't believe that they need to be. (Do we really need four or five levels of list assessment underneath FL?) I would say that both FL and List are needed, and their existence makes it easy to group and locate all list articles. (Especially useful since not all of them begin with the title "List of". Girolamo Savonarola 04:42, 17 October 2007 (UTC)
GA is not applicable for lists, so we can discard that one. But otherwise, you can assess a list by several parameters, such as completeness (primary criterion), adherence to style guidelines, image availability/licensing, alternate presentation methods (such as timelines and tables) and English conventions. So there are quite a few criteria that WikiProjects currently use to assess lists. Titoxd(?!? - cool stuff) 05:19, 17 October 2007 (UTC)
My hope with my suggestion was to get the article space classes cleaned up enough so that article in that namespace appear as categorized. I think the considerations about how many classes should be added only serve to complicate the issue. I think the only other consideration should be whether we should also add Disambig-class, which I forgot is also in article space. Lets just get article space entries to show up as assessed. Later we can add more classes if necessary.--TonyTheTiger (t/c/bio/tcfkaWCDbwincowtchatlotpsoplrttaDCLaM) 14:10, 19 October 2007 (UTC)
I think I we EITHER need to come up with a way to assess lists against some kind of rubric to come up with an SSBA rating, OR we need to introduce List-class. I don't really mind either way, but Tony is right that we need to bring the non-FL lists "into the fold" somehow. Walkerma 04:27, 30 October 2007 (UTC)
I think we could start with just list and FL. Classes could be added later if necessary. Let's just get our accounting right in the "articles by quality log". We may not ever agree on how to assess them. Lets just create list and FL. If there is demand to assess them later we can. What is the SSBA system? Why get hung up on a debate about assessment? Lets just take account of the article space things that we know about. No one is assessing lists. Many are using class=list. Lets get these articles counted.--TonyTheTiger (t/c/bio/tcfkaWCDbwincowtchatlotpsoplrttaDCLaM) 22:34, 30 October 2007 (UTC)
Yes, I agree - I think we need to be pragmatic here. If people are assessing lists with the Stub-to-FA rubric, that's great, and they are showing up on the logs. But many people aren't, and List-Class has become an normal class by convention if not consensus. The fact is that there are likely thousands (if not tens of thousands) of articles which have been tagged as such. And they need to start showing up on the logs one way or another. At the very least, the List class should be added for the purposes of at least keeping these articles trackable. We can argue about the merits of FL and assessing lists under the official rubric at another time, but right now, these articles need to be tracked one way or another. Every day we wait, more information (including inappropriate tag removal) is being lost to the hole in the logs. The rest of these issues, we can debate for however long it takes, but I don't think anyone is in favor of leaving these articles completely in the dark in the meantime. Girolamo Savonarola 23:22, 30 October 2007 (UTC)
Have to agree as well. There are at least 230 projects which have list-class designations, and it doesn't make sense to exclude those articles from the bot. Maybe a two level form (Featured list and other) might work best, but there almost certainly are enough projects using this grade out there to make it necessary. John Carter 23:28, 30 October 2007 (UTC)
I hope this doesn't throw this discussion completely off the topic of the FL and List-classes, but shouldn't we also add in the other classes while we are at it? Every other class is probably not showing up in the logs either and most likely projects don't want to lose pages in other classes to the "hole in the logs". I think it was stated it was possible to have the bot omit empty rows from assessment statistics tables. FunPika 23:58, 30 October 2007 (UTC)
The difference is that most of those other classes are non-article ones. Girolamo Savonarola 03:22, 31 October 2007 (UTC)
OK, I will add List-class for now, as this has least amount of opposition from what I see. Perhaps FL-Class and the other ones need separate discussion.
I hope to implement this by the end of this week. Oleg Alexandrov (talk) 04:59, 31 October 2007 (UTC)
I implemented the addition of List-Class together with the removal of empty rows in the stats (so that projects not using List-Class are not affected by its addition). Here's a sample diff. The change will propagate to all projects once the bot starts a new run. Oleg Alexandrov (talk) 21:30, 4 November 2007 (UTC)
Thanks a lot, Oleg! —Preceding unsigned comment added by Walkerma (talk • contribs) 02:36, 5 November 2007 (UTC)
An issue that has come up now is how to properly format the first column "Quality" to span the changing number of rows, see this edit. - Trevor MacInnis (Contribs) 19:37, 5 November 2007 (UTC)
This is fixed now, hopefully. It may take a day or more until the change propagates to all projects. Oleg Alexandrov (talk) 03:53, 9 November 2007 (UTC)

[edit] FL class

I would like to get article space cleaned up. I will start two separate debates. We should add FL class. I think people have been doing patch jobs with FL=FA commands. Let's just do it right. Thanks for adding list class.--TonyTheTiger (t/c/bio/tcfkaWCDbwincowtchatlotpsoplrttaDCLaM) 21:50, 6 November 2007 (UTC)

If we are going to be seperating lists into their own class, I strongly agree that there needs to be a class for FLs. I spend a lot of my time here on lists, and I'd like to be able to see which lists are complete and which still need work. --Arctic Gnome (talk • contribs) 17:08, 7 November 2007 (UTC)
I am not sure I understand why this did not happen when list class was created.--TonyTheTiger (t/c/bio/tcfkaWCDbwincowtchatlotpsoplrttaDCLaM) 02:58, 9 November 2007 (UTC)
Well, I personally thing that anything that is Featured should be listed, no matter what. So, having a Feature List class is a must.--Kranar drogin 17:54, 10 November 2007 (UTC)
If we create an FA class, we should also probably create at least one separate lower list class, and maybe turn the existing class into the equivalent of "Unassessed-class Lists?" John Carter 18:09, 10 November 2007 (UTC)
Why doesn't {{FA-Class}} not meet the desired niche? Titoxd(?!? - cool stuff) 06:04, 13 November 2007 (UTC)
I agree. Lists don't really need a separate quality class, IMO. I think just sticking lists and articles under {{FA-Class}}, there isn't much difference between the two as people think. Spebi 20:01, 14 November 2007 (UTC)
Fair enough. I guess anything under FA is a completed work and isn't really our focus anyway. I'll keep FL as an option in my projects, but put articles so tagged in with the FAs. --Arctic Gnome (talk • contribs) 07:39, 15 November 2007 (UTC)
I think there ought to be a FL-class for the bot. I don't see a reason really to make A-list and B-list options. Just my two cents. --Pinkkeith (talk) 20:55, 9 January 2008 (UTC)
I would like to see this incorporated, too. As of right now, FL-rated articles appear to be nonexistent at WikiProject Dartmouth College. Dylan (talk) 12:26, 13 January 2008 (UTC)
The reason that FA-Class is not right is because FA stands for "Featured Article" and lists are not articles. Looking at sports projects, the NFL project specifically, there are 20 some FL's yet only 3 or 4 FA's. In the stats it only shows FA so it seems that there is only 3 featured content in the NFL project when really there is like 28. We have almost 600 featured lists and are promoting something like 35 a month, which means soon FL is going to rival FA. I think the way to incorporate this is like what someone above said, where it is up to the Wikiproject to decide if they use FL or not. I would really like to see this implemented.
Gonzo fan2007 talk ♦ contribs 04:34, 19 January 2008 (UTC)
Wait, you mean that although we have the tag and class available for FL and Lists - FL's aren't added to the statistics but Lists are? Thats... strange given that {{FL-Class}} has been taken up and is being used by so many (see Category:FL-Class articles). Nanonic (talk) 21:42, 18 February 2008 (UTC)

[edit] 1,000,000

Wow, we've just passed one million assessed articles. :-) Kirill 12:36, 18 December 2007 (UTC)

And yet as per this chart, we aren't making any progress against unassessed articles and untracked articles, we're just keeping up with the new ones. Arctic Gnome (talk • contribs) 15:50, 18 December 2007 (UTC)
Partially because there have only recently been created assessing banners for several projects, like Micronesia, Melanesia, and many of the countries not covered by individual WikiProjects. I've been blindly arrogant and created separate work groups for each extant sovreign state on the planet, and am shortly going to adjust the various relevant regional Project banners to accomodate them. When we've got that accomplished, and when those banners are placed, I hope that we'll see some improvement then. It'll take awhile to adjust all the banners, though. User:Phoenix-wiki is working on the Africa project banner, and I sincerely feel for the amount of work that's going to have to go into that one. John Carter (talk) 15:55, 18 December 2007 (UTC)
So we're going to have wikiprojects without any members that exist solely for the purpose of 1.0 assessment? I guess that makes sense. --Arctic Gnome (talk • contribs) 16:21, 18 December 2007 (UTC)
Well, no. They all have one member so far, me. My obsession with userboxes scares even me sometimes. :) And these aren't Projects per se, but work groups/task forces of extant Projects. They were basically created because of Wikipedia:WikiProject Africa's somewhat untenable scope. Basically, the reason behind it is that there already are projects out there with one "central" member who does most of the orchestration and development, and the separate assessments will allow any individual an opportunity to review the relevant material and, hopefully, improve at least the core article for each nation, knowing a bit more easily what the other related content already does and does not contain. In time, of course, I hope that the tagging will draw more members into the parent projects. But, this way, at least there is a slightly better chance of some of the smaller nations, like say Marshall Islands, which is currently only at Start class, getting a bit more attention than they do today. John Carter (talk) 17:16, 18 December 2007 (UTC)

[edit] I have no idea what is going on

I just started a WikiProject (I think), at Wikipedia:WikiProject Deftones, and am in the process of tagging the articles (I think). If someone could have a look at Deftones for instance, and tell me if I'm on the right track, I'd greatly appreciate all the help I can get! I just made the jump from editing sporadically to trying to do something more substantial, so I'm learning by doing and could do with some hand-holding. Thanks in advance, Seegoon (talk) 05:21, 19 December 2007 (UTC)

[edit] Proposal to run the bot every four days

Well, the time has come again. :) The bot takes a lot to do its runs, I think even more than four days. Three days is not enough any longer, and the way things are now there are often two instances of the bot at any time stepping on each other's feet which slows things down. So, I propose to run the bot every four days instead. Comments? Oleg Alexandrov (talk) 05:59, 9 January 2008 (UTC)

Fair enough - the bot has over 50% of the English Wikipedia to go through! Projects can survive the wait. Walkerma (talk) 06:12, 9 January 2008 (UTC)
Yep, looks fine to me. Kirill 14:01, 9 January 2008 (UTC)
I made a suggestion awhile back. I wish there was a way to have the bot skip inactive projects for a period of time. I would think that would cut back (a little) on the workload.↔NMajdan•talk 14:11, 9 January 2008 (UTC)
Long term such an algorithm could be hard to maintain and may not add much (the bot goes rather rapidly through inactive projects, since it is smart enough to realize that a page does not change so it should not attempt to edit it -- it is the later which takes most time). Oleg Alexandrov (talk) 16:25, 9 January 2008 (UTC)
Agreed, anything to make sure it runs; sometime! It looks like it has stuck again at 9.15 today (again!). :: Kevinalewis : (Talk Page)/(Desk) 14:15, 9 January 2008 (UTC)
OK. The bot will run every month on the following days: 1,5,9,13,17,21,25. So it will have a 5-6 day break at the end of each month before resuming on the 1st of next month (otherwise, if it runs on 29th also, it may run too often, and that can have two bots running at the same time which slows both of them down). Oleg Alexandrov (talk) 16:25, 9 January 2008 (UTC)

That'd be fine if the bot ran fully every four days. Right now the Zebras haven't had a bot run in TWELVE days (since Jan 2). — Rlevse • Talk • 01:46, 14 January 2008 (UTC)

Go to [1] and run it. --NE2 15:03, 14 January 2008 (UTC)

Each time there is an issue or the bot gets overloaded, such things happen. Sorry. One proposal which I also raised earlier is to not have the very large biography articles project, which is largely duplicated by the smaller biography projects (which are still very large by themselves).

I now started a new version of the bot. Hopefully it will get to the zebra soon. Oleg Alexandrov (talk) 16:12, 14 January 2008 (UTC)

If there would be any way to create the biography project statistics page simply by adding together the stats for the work groups, I'd probably support it, if we got to the point that we have enough work groups to cover most of the biography articles. Unfortunately, as it stands, many still aren't assigned or even necessarily relevant to any work groups, but I'll see what I can do regarding that in the near future. John Carter (talk) 16:15, 14 January 2008 (UTC)
Maybe you can create a category for "biography articles not assigned to a workgroup", like Category:U.S. road transport articles without a state parameter? Of course it would need a change in the bot's code to look up the quality/importance for each of those. --NE2 16:19, 14 January 2008 (UTC)
If and when I get done with what I'm doing and get access to the template, that sounds like a very good idea. I think it might be workable by the end of the week. I hope. Wish me luck. John Carter (talk) 16:21, 14 January 2008 (UTC)

[edit] Did I do it wrong?

I created the Wikipedia:WikiProject Elvis Presley in November and ran the tool. I recall, I believe, that the table was updated immediately. However, I just created Wikipedia:WikiProject Tool, ran the tool and nothing. Have I done something wrong? Lara❤Love 17:50, 14 January 2008 (UTC)

I believe you categorised the quality articles wrong. Compare Category:B-Class Tool articles and Category:B-Class Elvis Presley articles.--Peter Andersen (talk) 18:11, 14 January 2008 (UTC)
Thank you! It's working now. Lara❤Love 14:36, 15 January 2008 (UTC)

[edit] Why is the bot slow?

First several posts taken from User talk:Oleg Alexandrov.

I'm just curious - what's the main source of delay in the bot, that makes it run so slowly? Is it the editing rate, or the download rate, or something else? It seems remarkable to me that it requires three days to complete - how many edits per run does it make? — Carl (CBM · talk) 03:26, 14 January 2008 (UTC)

The bot does one query every 1 second and an edit every 5 seconds. I'll be able to answer the question about how many edits it makes after I get a complete untampered log for a recent time, which would be in five days maybe. Oleg Alexandrov (talk) 16:31, 14 January 2008 (UTC)
I'm not sure if you know about the "maxlag" method of rate limiting.[2] The idea is that you send a parameter with each edit, and if the server lag is too high the server tells you to wait and try again. But if the lag is not too high, there is no need to add an extra delay after each write anymore. I have written some perl code to do this (I needed extra functionality compared to perlwikipedia); you might want to look at it.[3]. The watchlist part seems to be buggy, and the output is ugly, but the editing and maxlag parts work. It's based on some code from my API library.[4] — Carl (CBM · talk) 03:04, 15 January 2008 (UTC)
Thank you for your suggestion. I'd like to note I have precious little time for programming recently (I do it full time at work too, and when I get some little free time at home it is not the same fun thing anymore :) If you have the time, and would like to experiment with this, I, and all the WP 1.0 participants would be extremely grateful. The bot source code is here, and I can give you commit access to the SVN code base.
Another thing to think about is to be smarter about using query.php to get history versions in a smarter way (that is now done in a batch of 10 articles -- out of a million overall articles! -- which is much slower than category fetch, which is I think in gulps of 400). According to Yurik, one could make one monster POST request for history versions (say a few hundred), which I think would speed things up tremendously. Carl, would you be interested? :) Oleg Alexandrov (talk) 06:13, 15 January 2008 (UTC)
I have a small backlog of code to write for the version 0.7 project. Once I finish that, I don't mind looking at the WP 1.0 bot code. I would need to consult with you on a few things, but I can write code. — Carl (CBM · talk) 12:45, 15 January 2008 (UTC)
That would be very much appreciated. Thanks! Oleg Alexandrov (talk) 06:07, 16 January 2008 (UTC)

[edit] Stats problems

The stats for Wikipedia:WikiProject Unionism in Ireland seem a little off and the bot appears to be ignoring anything higher than Start class - see Wikipedia:Version 1.0 Editorial Team/Unionism in Ireland articles by quality statistics and Wikipedia:Version 1.0 Editorial Team/Unionism in Ireland articles by quality. Could someone with the technical know how take a look and see if the project assessment is set up correctly? Timrollpickering (talk) 10:11, 19 January 2008 (UTC)

The FA, GA categories are empty, that's why they don't show up. I think there was agreement a while ago that empty categories should not show up in the stats. So just add some articles there, and hopefully the stats will reflect that. Oleg Alexandrov (talk) 04:28, 20 January 2008 (UTC)

[edit] wikiproject breakbown

would it be difficult to add a simple "number of articles" column to the table in Wikipedia:Version 1.0 Editorial Team/Index? It would be really interesting to see the breakdown of the different wikiprojects.. then you could easily tell the 'big' projects from the 'small' projects. thanks! 131.111.8.97 (talk) 13:21, 23 January 2008 (UTC)

The disadvantage would be that all the table monitors yet is the number of articles tagged, which has less to do with the size of the project as the amount of tagging its done. Also, generally, if any party were to want to do that, all they would have to do is compare the totals for the various projects. However, at this point I can't see how useful it would be. Several projects now take the assessments of a more focused project as their own, by adding that group's categories to their own. The France project now takes the assessments from the Wikipedia:WikiProject Africa/French Africa work group, for instance. Those articles very likely would not appear in the list of articles tagged for France. Also, and I am speaking from experience here, several projects have much greater scope than the amount of tagging they have been able to do to date. The WikiProject Biography is probably the one project which hasn't tagged the biggest number of relevant articles, simply given the scope of that project. Any such numbers as you are proposing would probably be both time-consuming for the bot to generate, and less than productive in terms of actually measuring the real scope of a project rather than its amount of tagging to date. John Carter (talk) 14:16, 23 January 2008 (UTC)
i agree it wouldn't measure the size of the project, but how much tagging has been done - which is nevertheless an interesting statistic, isn't it? and i ask because i didn't think it would actually take that much of the bot's time to do (i.e. a single edit to the page regularly).. unless i am mistaken. 82.6.96.66 (talk) 14:55, 7 February 2008 (UTC)

[edit] Table update

The fact that the bot doesn't get as far as the update to this table, doesn't mean that the time period should be adjusted to reflect the bot "failures". I as much should be reflecting the run frequency and the update intention. :: Kevinalewis : (Talk Page)/(Desk) 16:34, 23 January 2008 (UTC)

The instance of the bot from January 14 is still running, doing the biography project, and and consuming 1GB of memory. Even if the biography folks are not ready for it, I suggest we remove the main biography project from the index to save server resources and not strain the Wikipedia database. Would make it much more certain it would actually get to updating the table too. Oleg Alexandrov (talk) 17:02, 23 January 2008 (UTC)
That being the case, you might want to contact User:Kingboyk, who probably understands that project's banner better than anyone else. He might be able to add a few parameters to make it easier to accomplish the above. John Carter (talk) 17:03, 23 January 2008 (UTC)
I used to, but I've not been very active. Kirill and Oleg can do just as good a job, almost certainly better. --kingboyk (talk) 22:59, 23 January 2008 (UTC)
Actually, the instance of the bot from January 18 was in the same situation, stuck fetching the biography articles and consuming another 1GB of memory. I killed both of them, since there's yet a third instance of the bot running from January 21 and the total server memory is only 3GB. So, it won't get to updating the main table again. In short, the bot does not scale to cover the biography project with its half a million articles. Oleg Alexandrov (talk) 17:07, 23 January 2008 (UTC)
(Edit conflict) Could we perhaps persuade the Bio project to accept us running the bot only once a fortnight, or once a month? That would allow all of the other projects to get a turnaround in 3 days or so, except during the time period that the Bio update is running. The wording could be amended to spell out the new arrangement. Walkerma (talk) 17:12, 23 January 2008 (UTC)
What server is this running on now? Do we need to clone the bot to run on a faster server, at least during the WP:Bio run? Walkerma (talk) 17:12, 23 January 2008 (UTC)
The server is the one you got for me, I think it is in France. The server is powerful enough, the issue is that there are just too many articles in the biography stub category, it takes a very long time to retrieve it page by page and sometimes the bot even does not get to the end of it (the pointers to the next page in the category can get messed up in the database if you keep on accessing the category for very many hours). In short, the hardware is not the problem, it is just getting huge categories does not scale. Oleg Alexandrov (talk) 04:00, 24 January 2008 (UTC)
Alternately, and I don't know enough to know if this is feasible, might it be possible to create a Category:Biography articles not yet assigned to a work group, and then run all the various subprojects and that category in succession, maybe creating a way to "add together" all the subproject article ratings at the very end for the main biography banner's statistics? John Carter (talk) 17:17, 23 January 2008 (UTC)
More workgroups would help. Ideally, many more so that the Project resembles to the bot lots of small to medium WPs. Alas, there hasn't been much action on that front.
In the meantime, of course if it's causing problems Bio will have to be run seperately and less often. --kingboyk (talk) 22:59, 23 January 2008 (UTC)
Well, I guess then the best thing to do is create more work groups. There is an extant religious figures proposal which I can try to add to the template eventually, but I think we could probably add more as well. I'm starting a section on the Biography Project talk page to solicit ideas of further reasonable Biography subprojects. Anyone is free to suggest any that they think would be useful. John Carter (talk) 23:04, 23 January 2008 (UTC)

One intermediate solution would be to not count or list the stub and unassessed biography articles (for the main project). Those are mostly assessed automatically anyway, and there is no big need I would argue to know those numbers. That would speed up the bot very much. Oleg Alexandrov (talk) 04:00, 24 January 2008 (UTC)

[edit] WP 1.0 Bot behavior with sub-categories

Since the talks of adding additional classifications to the bot have been ongoing for some time with no consensus, I have a question regarding the behavior of the bot that could be a temporary solution. Does the bot go through articles located in a sub-category of a particular classification? For example, if Category:FL-Class medicine articles was a sub-category of Category:List-Class medicine articles, would the bot consider the articles in Category:FL-Class medicine articles as List-Class? If it did, then projects using the FL-Class assessment could at least get the proper statistics with List-Class and FL-Class merged, and still keep separate categories for featured and non-featured lists. What is the current behavior of the bot? If it does not consider sub-categories, could this feature be added? Only one sub-level of categories would need to be processed, so complicated scripting checking for loops would not even be necessary. --Scott Alter 08:35, 24 January 2008 (UTC)

As of now the bot looks three levels deep into categories, from the base Category:Wikipedia 1.0 assessments, to individual projects (biography, chemistry, etc.) to individual classes of articles (FA, GA, etc.). I would be reluctant to add one more search level unless indeed very much needed (it would make the code maintainance more complex too).
One solution is to have all FL-Class articles show up also in List-Class (there's a trick to do that with templates I think).
About whether to include FL class, again, technically it is very easy, I am just not sure if there is consensus on that or who is in charge of judging if consensus exists. (Does this project have a boss? :) Oleg Alexandrov (talk) 19:18, 26 January 2008 (UTC)
Um, I'm guessing, considering you're the one who has the ability to do it, and most of the rest aren't, you are in this instance? John Carter (talk) 22:41, 10 February 2008 (UTC)
I don't have an opinion about the merits of having the FL-class in or not. Perhaps Walkerma who's been coordinating the release of WP 1.0 articles can comment on this. Oleg Alexandrov (talk) 05:32, 11 February 2008 (UTC)
Is there likely to be any significant practical impact to bot operations from adding another level? If so, then doing this for FLs really wouldn't be worth it.
Aside from that, would it be possible to have FL mapped to FA for the general statistics? In other words, for a project where FL categories existed, the statistics table would have two rows, one for each of FA and FL; but for the overall table, the FA and FL rows would be merged, minimizing the inaccuracy due to most projects' lack of such ratings. Or would this be too complicated? Kirill 06:04, 18 February 2008 (UTC)
Merging FA and FL in the general stats would not be hard, but adding one more search level would be more tricky and would make the code more convoluted. I'd be inclined to do the former, so allowing FL for the projects which use it but adding FA and FL in the global stats table, if that's what people want. Oleg Alexandrov (talk) 16:02, 19 February 2008 (UTC)
Rather than making the code any more complex, why not just add the FL-Class to the table and treat it as its own class (which probably only requires 5 extra characters: "FL",). Then, the projects that use (or want to use) FL can do so, and keep separate stats on lists from articles. If a project does not want to use FL, then it simply won't appear in their stats table - like any other unused class. There does not seem to be any blatant opposition to including FL. Many people seem to support inclusion, and a few are indifferent. In the thread above, one person questioned why not just combine them with FA, which I think others have already addressed. With several people supporting FL, a few abstentions, and no opposition, I'd say we have consensus. --Scott Alter 19:08, 19 February 2008 (UTC)
I still believe that it really is not necessary to have two classes to address the same thing (a finished page). I'm much more inclined to rename FA-Class as Featured-Class, and combine both, which would also prevent the code of the WP:1.0 bot from becoming even more convoluted. Titoxd(?!? - cool stuff) 03:28, 24 February 2008 (UTC)
As far as I can tell, Category:FL-Class articles was created back in October 2007, so it seems that FL-class is here to stay. The sticking point is whether to bring it within the WP 1.0 bot assessments or not, and it seems Oleg was happy to do this. I support this class being tracked separately because FA and FL are different processes, and at the moment, when trying to work out stats for one or the other, many categories mix them up and that makes things harder. Have a look at Category:FA-Class biography articles and tell me how many are lists and how many are articles. Carcharoth (talk) 22:23, 24 February 2008 (UTC)
Like Oleg said, one way a project can add extra ratings is to duplicate them into two categories. For example, in the mathematics project there are 'B+ class' articles. These are also put in the GA-class category by the talk page template, so the WP bot counts them as GA class in its statistics. But at the project level there is a more specialized bot that knows how to sort them out. Compare project tables vs. the WP 1.0 table (minor differences in counts are because they are generated independently at different times). — Carl (CBM · talk) 03:40, 11 February 2008 (UTC)
Sorry I missed this post before. I don't have strong feelings, since for our article selection process we give FAs the same number of points as FLs. I can see some moderate value in the idea, but it does add yet another line to the tables. It would at least prevent the perennial question, "Why does WP:1.0 have more FAs than WP:FA (Answer - because of the FLs). The idea of FL was raised quite recently, but a common response was that most projects are happy just to tag such lists as FA-Class. Since that time List-Class has become part of the landscape, so maybe the rules have changed a bit. My perception is that a few people really want it, but many people have a mild opposition to the idea, and it's hard to decide on a consensus in such cases. My view: If a significant no. of WikiProjects want it, even a minority, then we should allow it as an option. Those who don't want it won't see the line in their stats tables anyway. Let me ask a few more people. Walkerma (talk) 04:29, 17 February 2008 (UTC)
I'm not really sure we need this level of granularity. {{FA-Class}} indicates that the article has attained the highest quality classification possible for that type of page. If FA-Class is causing some confusion when applied to FL's, we can always rename it it {{Featured-Class}} and achieve the same effect. Titoxd(?!? - cool stuff) 09:28, 18 February 2008 (UTC)
Well then, since the requests for FL keep on coming, and since the opposition is not very strong, perhaps we could add FL-Class. Barring some strong feelings on this in the next several days I could work on it. Then, if people want, I could also implement Kirill's suggestion to merge FA and FL in the global stats. I hope however that people won't want FA and FL to be kept merged or separate on a per-project basis, that would be too much to manage, so we either go with FL split from FA on all projects, or on none of them. Oleg Alexandrov (talk) 06:13, 21 February 2008 (UTC)
I'd say Support, overall, there is definitely a groundswell of strong support for the idea from some projects. Walkerma (talk) 06:40, 21 February 2008 (UTC)
...but a common response was that most projects are happy just to tag such lists as FA-Class. This is very much like the Ford Model-T's color. "Any customer can have a car painted any color that he wants so long as it is black." At best, labeling FLs as FA-Class is a hack. WikiProjects use the hack because there were no other alternatives available until just recently. However, the FL-Class has now been added to the official assessment scale, and several WikiProjects are already adopting the FL-Class to their assessments. The bot should be updated to reflect this addition of this new standard class. --Farix (Talk) 15:57, 22 February 2008 (UTC)
I implemented FL-Class, see here. It will take a few days until the bot will start again, then it will add FL-class articles. Oleg Alexandrov (talk) 04:23, 27 February 2008 (UTC)

[edit] How can I request one of these boxes?

I am a member of WP:Elements and I would like to know if it is possible to get a box like this one for the project. Thanks, Nergaal (talk) 22:01, 11 February 2008 (UTC)

Can you be specific about what box you are referring to? Oleg Alexandrov (talk) 03:56, 12 February 2008 (UTC)
basically an entry in Wikipedia:Version 1.0 Editorial Team/Index called "Chemical element articles by quality" that links the pages using the template {{Chemical Element}}. Nergaal (talk) 06:08, 12 February 2008 (UTC)
You can go to the index page and follow the link to the instructions from there. Oleg Alexandrov (talk) 15:45, 12 February 2008 (UTC)

[edit] Format of the stats table

Copied from Wikipedia_talk:Version_1.0_Editorial_Team#Assessment_graph. I also support this proposal. Walkerma (talk) 03:58, 17 February 2008 (UTC)

Is there some way to have the list row line up above the assessed row? Because, as it is, there's assessed, list, total. It looks weird. Lara❤Love 04:46, 1 February 2008 (UTC)

I have to agree with this, as well. Any project which has decided to use the class clearly is going to regard these articles as having already been assessed. Girolamo Savonarola (talk) 04:54, 1 February 2008 (UTC)
Done, see here. The changes will propagate to the other projects when the bot does its run. Walkerma, thanks for posting that request here, this is a better place for bot-related discussion. Oleg Alexandrov (talk) 16:59, 17 February 2008 (UTC)
Thanks! And how appropriate to use WP:Aesthetics as the example! Walkerma (talk) 18:14, 17 February 2008 (UTC)

[edit] Adding FL class to a WikiProject's assessment scheme

Is there a step-by-step guide to how to add "FL" class to a WikiProject's assessment scheme? Off-hand, I think the things that would need to be changed are:

  • (a) The project article talk page template would need coding for the FL parameter.
  • (b) The new FL-Class category would need to be created and categorised (where?).
  • (c) The project's assessment documentation would need updating.
  • (d) Existing featured lists would have to be switched from FA to FL class.

I think that's it. Can someone confirm that the bot will automatically pick up such changes and change the pages it updates? I can do (b)-(d) OK. Could someone help out or write a guide for how to do (a)? I would like to implement this for some projects, especially as my recent attempts to use subcategories of Category:FA-Class articles to see how many featured articles some projects have, has been hampered by the featured lists being mixed up with the featured articles. Carcharoth (talk) 01:30, 24 February 2008 (UTC)

Having done the same thing myself in the past, all I did was copy the FA class parameters in the existing banner, place the copy right below the original, and alter the FA to FL. I hope that makes sense. I can't find a specific example right now, but if you give me an example of a banner you're considering, I could demonstrate how to do it there. John Carter (talk) 17:23, 26 February 2008 (UTC)
Like this? I then created Category:FL-Class dinosaurs articles and categorised and added documentation based on Category:FA-Class dinosaurs articles, and made sure the current featured list now appeared in this category. I couldn't find the project assessment documentation, so I left a note on their talk page. Now I just have to see if the bot will pick up the new category. Carcharoth (talk) 12:15, 3 March 2008 (UTC)

[edit] Proposal to remove the biography articles

Today I logged in to the server running the bot, to discover four instances of the bot running, with the older being 12 days. Here's the deal, the bot does not scale to the entire biography project. I suspect from past experience that as the bot keeps on getting pagefuls of a given category, if that category is huge and gets updated as the bot reads it, the bot may get confused and get into an infinite loop or something. Then we can't get the global stats either.

I propose we remove the biography project from the assessment, or at least, the Stub-Class biography articles, which are a third of a million and which are assessed semi-automatically anyway I think (that an article is a stub does not provide any real assessment value).

Note: the effect of this won't be that large. Very many biography articles are covered within subprojects. Comments? Oleg Alexandrov (talk) 06:37, 26 February 2008 (UTC)

Would it be possible to have a separate instance of the bot just to do the Bio articles? Obviously, this would mean that the global stats wouldn't include them; but that's a minor issue. I'm more concerned with keeping projects supported log-wise for as long as possible. Kirill 13:04, 26 February 2008 (UTC)
Can we simply get a second bot, perhaps even a clone of WP1.0 Bot? It could perhaps be tweaked to allow for the huge categories. Perhaps it could be run & administered by someone from the WP:Bio project? As for global stats, it should be trivial to get a bot to add the two sets of data together - we could probably get the SelectionBot to do that task if necessary.
On a related theme, Oleg, where does the bot reside these days - did it get relocated onto Emmanuel's server in Europe? Walkerma (talk) 13:40, 26 February 2008 (UTC)
The bot has been on Emmanuel's server in France for a while. Walkerma, thank you, the bot is so greedy recently it would have ruined the university computer I ran it on.
At some point I will then fork the biography articles to its own bot, to be run weekly. I can't promise each of those biography runs will be successful, as, again, the bot seems to be confused with huge categories (and it may not even be bot's fault, the server may be feeding it incorrect data). But it would be better like what is today, I think. Oleg Alexandrov (talk) 04:22, 27 February 2008 (UTC)

[edit] Including update date

I would like to propose to include the date of when the information in the stat boxes (such as this one) was last updated. The date is already included in the edit summary, but not in the actual box. At this point I am simply looking for feedback regarding whether this is a good idea or not. Thanks.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); 15:10, 14 March 2008 (UTC)

The date will show up everywhere where a stat template is transcluded. Many people may not like that. My take is that it is simpler to take a look at the history if you'd like to see the last update date (or even simpler, take a look at the bottom of the page, there is already a datestamp on the bottom of every wikipedia article). Oleg Alexandrov (talk) 03:49, 15 March 2008 (UTC)

[edit] WP 1.0 bot trials

I've been working to update a few parts of the WP 1.0 bot code to speed things up. The goal is to get the bot to finish a complete run in two days, rather than four or five. There should be no visible change in the bot's output; the changes are all to the underlying code to access the article data. So there is no new functionality, just faster processing.

The good news is that the new code is ready for heavy testing. I have already been testing it on individual projects by hand, but at some point I need to simply turn it on and let it go. I'm planning to do that sometime in the next few days. The cgi script that allows you to rate one project at a time will still be using the older code.

The new code puts "(test code)" at the end of the edit summary. If you notice any strange behavior in an edit marked that way, please let me know. If all goes well, the new code will replace the old code in a couple weeks. — Carl (CBM · talk) 01:08, 20 March 2008 (UTC)

This would be awesomely great! Thanks! Oleg Alexandrov (talk) 04:15, 21 March 2008 (UTC)
The full trial is now underway. If anyone sees something odd, please let me know. — Carl (CBM · talk) 19:10, 21 March 2008 (UTC)
And complete, in about 44 hours, including an update of the global stats at Wikipedia:Version 1.0 Editorial Team/Statistics for the first time since Feb. 14, and including updates to the category structure. I'm going to look through the logs it made during this run for any issues; if you see anything odd in the results, please let me know. — Carl (CBM · talk) 18:58, 23 March 2008 (UTC)
Cool! Will this be the regular run from now on? Oleg Alexandrov (talk) 19:00, 23 March 2008 (UTC)
Not quite yet. There are a few minor technical issues I need to fix (for example, two or three of the 1000+ biography pages didn't get uploaded because the database locked for a while overnight last night. I can avoid that problem.). Then I want to do another test run before I make it automatic. Also, I want to give time for any other bugs to be reported. But overall I am happy with this run. — Carl (CBM · talk) 19:19, 23 March 2008 (UTC)

I ran the new code a second time; this run took 49 hours. I did fix the issue with database locks - the DB locked 3 times during the test run, but the script just waits them out. So unless anyone has errors to report, I'm ready to set the new code to automatic.

We could hypothetically run the full assessment 3 times per week, but it would be tight. Conservatively, it takes 2.5 days to complete each run. I think 2 full runs per week should be enough, since individual projects can always run it by hand. I was thinking of running the full assessment Sunday and Wednesday. Does that sound reasonable? — Carl (CBM · talk) 16:15, 28 March 2008 (UTC)

I set up the crontab to run it Sunday and Wednesday. — Carl (CBM · talk) 12:31, 30 March 2008 (UTC)

[edit] Thought I did it right

I redid the Wikiproject Ohio template to include importance. It works fine as far as I can tell. The problem is that the chart that lists the number of articles by importance and class will not update to show importance even after running the bot manually. It just says "No importance" for all of them. Could someone please take a look at it? Thanks. §tepshep • ¡Talk to me! 00:18, 23 March 2008 (UTC)

The category needed to be named Category:Ohio articles by importance instead of Category:WikiProject Ohio articles by importance. I fixed that, ran the script, and now the table is correct at Wikipedia:Version 1.0 Editorial Team/Ohio articles by quality statistics. The rule of thumb is that whatever the statistics page is called, that's what the quality and importance categories should be called. In this case, "Ohio". — Carl (CBM · talk) 19:13, 23 March 2008 (UTC)

Totally looked that over. Thanks so much for your help! §tepshep • ¡Talk to me! 03:16, 24 March 2008 (UTC)

[edit] Index page shows up blank

There is currently an issue with Wikipedia:Version_1.0_Editorial_Team/Index showing up as a blank page (with 0 bytes of content). This is because the source for that page is too large. Hopefully the bot will be fixed soon to split the data over several pages. The bot is just about to finish a full run, so even though the index is not visible, the individual projects should be correct. — Carl (CBM · talk) 00:39, 24 April 2008 (UTC)

Thanks! I saw this yesterday, but thought I just had a browser problem. Walkerma (talk) 06:16, 24 April 2008 (UTC)
I fixed this issue. Now there is Wikipedia:Version 1.0 Editorial Team/Index2 where the second part of the index resides, with the first part in the old place, Wikipedia:Version 1.0 Editorial Team/Index. This is a stop-gap measure, more thought needs to be given to how best to split a big index into subindices. Oleg Alexandrov (talk) 21:21, 26 April 2008 (UTC)

[edit] Project Quality Stats

Wikipedia:Version 1.0 Editorial Team/Biography articles by quality log, Wikipedia:Version 1.0 Editorial Team/Chicago articles by quality log, and Wikipedia:Version_1.0_Editorial_Team/WikiProject Illinois_articles_by_quality_log had not been updated for ten days until PeterSymonds ran the bot automatically today. However, although I can see the tables have updated, these pages did not.--TonyTheTiger (t/c/bio/WP:CHICAGO/WP:LOTM) 19:33, 3 May 2008 (UTC)

We ran into an issue with the index page (see the previous section). I didn't start a full run because of that, until I have a chance to integrate Oleg's fix into the new WP 1.0 bot code. I should be able to do that this afternoon, but I haven't been able to do it earlier this week. Thanks for the gentle reminder.
I don't know why running the code manually would cause the tables to update but not the log pages. If that happens with the updated code as well, I will be able to diagnose it. — Carl (CBM · talk) 09:14, 4 May 2008 (UTC)

[edit] Bot queue

Is there any way to know what projects the bot is currently or planning to work on? I've tried to get Albums updated (haven't been since the 22nd) but the bot seems to stop working once it gets to the Gs in the 29k stub category. It's still has the fun 33k unaccessed category, so I wonder if it's just been queued or it that just too much to ask for? -- Ricky81682 (talk) 06:22, 5 May 2008 (UTC)

The bot does not provide the functionality you are asking for, and I don't know how I would implement it. The bot does not work with very long projects from the web-based interface, the server just cuts it off. For long projects it is better to wait for the scheduled run. Oleg Alexandrov (talk) 03:13, 10 May 2008 (UTC)
As Oleg says, the web interface only works for small projects, as there is a time limit for a cgi script to run. The idea of having the bot give more detailed info is good, and I'll make sure it is considered in any updates of the bot code. — Carl (CBM · talk) 10:55, 14 May 2008 (UTC)

[edit] Running the bot by hand does not work

The tool that allows people to run WP 1.0 bot on demand from the web-based interface does not work for the moment. The system administrator on the network it was running on told me it takes too many server resources. I am now looking for a new home for it. In the meantime, the only way to run the bot is through the regular scheduled run. Oleg Alexandrov (talk) 03:13, 10 May 2008 (UTC)

The tool is now at http://www.kiwix.org/~oleg/wp/wp10/run_wp10.html. The system administrator there asked me that people don't overuse it. My suggestion is that the tool should be used only for new projects or for small projects. Otherwise it would be better to wait for the standard run. Oleg Alexandrov (talk) 02:57, 14 May 2008 (UTC)
Does it keep a log of how often it is run? We can probably add some code to throttle it, so that it cannot be run too frequently. — Carl (CBM · talk) 10:53, 14 May 2008 (UTC)
It does not keep a log. I would assume it is not used a lot. If it causes problems in the future, I guess we can introduce something to restrict the usage. Oleg Alexandrov (talk) 00:10, 18 May 2008 (UTC)

I went on the link and typed in Devon. It appeared to be doing something and at the end said "Done with Category:Devon articles by quality!", but nothing appeared to have happened on Wikipedia regarding Devon. I also tried with Foobar and same as above happened. Meaty♠Weenies (talk) 00:48, 29 May 2008 (UTC)

Nevermind it works now. I've changed the name of my assessment from "WikiProject Devon" to just "Devon", but WikiProject Devon still appears on the list. Could someone please remove it. Meaty♠Weenies (talk) 01:40, 29 May 2008 (UTC)

[edit] How to add a project?

How does a given wikiproject start assessing its articles and getting indexed? I've looked around and can't find any instructions. --Padraic 19:16, 17 May 2008 (UTC)

If you visit Wikipedia:Version 1.0 Editorial Team/Index, then on top there will be a link to the instructions. Oleg Alexandrov (talk) 00:10, 18 May 2008 (UTC)
Oops, thanks. --Padraic 15:22, 18 May 2008 (UTC)

[edit] Could someone please help out...

the Wikipedia:WikiProject U.S. Supreme Court cases by putting a quality assessment box on their main project page. We aren't very good at working with the 1.0 bot. 20:08, 22 May 2008 (UTC)

Is there a trick to getting this to update? I've been assessing for a few days now but the chart doesn't reflect it. --Cdogsimmons (talk) 05:09, 31 May 2008 (UTC)
Essentially, wait until it gets to it. The bot works in alphabetical order, and you can see where it is right now here. Titoxd(?!? - cool stuff) 07:55, 31 May 2008 (UTC)

[edit] Bot Error

Hi, I've been trying to get the bot to run for the newly created Category:Lost articles by quality but it keeps giving me the same error over and over again:

Error message is: Could not get_text for Wikipedia:Version 1.0 Editorial Team/Lost articles by quality! at /home/oleg/public_html/cgi-bin/wp/modules/bin/perlwikipedia_utils.pl line 93 eval {...} called at /home/oleg/public_html/cgi-bin/wp/modules/bin/perlwikipedia_utils.pl line 82 main::wikipedia_fetch('Perlwikipedia=HASH(0x8b51fd8)', 'Wikipedia:Version 1.0 Editorial Team/Lost articles by quality...', 200, 1) called at wp10_routines.pl line 401 main::fetch_list_subpages('Wikipedia:Version 1.0 Editorial Team/Lost articles by quality...', 'ARRAY(0x86d3a64)') called at wp10_routines.pl line 182 main::main_wp10_routine('Category:Lost articles by quality') called at /home/oleg/public_html/wp/wp10/run_wp10.cgi line 53 Fetching Wikipedia:Version 1.0 Editorial Team/Lost articles by quality.}}

I don't know if it's a problem with the bot or if I did something wrong (most likely the latter) but if anyone could help it would be appreciated. Thanks, Scorpion0422 20:36, 4 June 2008 (UTC)

I was able to reproduce the problem. The page Wikipedia:Version 1.0 Editorial Team/Lost articles by quality is a redlink, and the bot is complaining about not being able to fetch it. In the past redlinks were not a problem, I don't know what happened later.
The problem is thus in the library the bot calls to get Wikipedia pages. I'll try to dig in there this weekend, and see what is going on. Oleg Alexandrov (talk) 03:02, 5 June 2008 (UTC)
Mmm, now its a bluelink. :) Somehow the bot corrected itself and updated the page. Oleg Alexandrov (talk) 03:04, 5 June 2008 (UTC)

[edit] List of assessment groups by article number?

Would it be possible to generate a sporadically-updated list of all of the 1.0 groups by some basic statistics like total number of tagged articles, number of unassessed articles, and the like? It would be useful so that we can see which projects clearly have assessment issues, as well as projects which may have a small enough scope to be more useful as task forces of other projects, or - in the case of blatant inactivity - be brought to MfD. Girolamo Savonarola (talk) 03:23, 5 June 2008 (UTC)

It would be a pain in the ass with the current framework, as it would pretty much require scraping the statistic pages for each project. It might be better to add this to the new framework being thought about. Titoxd(?!? - cool stuff) 03:43, 6 June 2008 (UTC)
If the list needed to be kept in sync with the statistics, perhaps. But the actual task is pretty simple; if I have a bit of free time this weekend, I'll put together a script that generates the needed lists. Kirill (prof) 02:11, 7 June 2008 (UTC)
Ok, the results are up at Wikipedia:Version 1.0 Editorial Team/Index/Comparison. Kirill (prof) 17:46, 8 June 2008 (UTC)