Wikipedia talk:Bots/Requests for approval/Archive 8

Jump to navigation Jump to search
Archive 5 Archive 6 Archive 7 Archive 8 Archive 9 Archive 10 Archive 15


EdoBot removing interwikis to nonexistent pages

Hey everyone. I've been running EdoBot with the new -cleanup option of interwiki.py. This new option removes interwiki links to nonexistent pages, and is a bit more cautious than the -force option which also removes interwikis to different namespaces, and other things that should generally be checked by humans. I didn't think this would be an issue, since the bot was approved to "create and maintain interwiki links" and removing links to nonexistent pages is part of maintaining. However, the bot has been going at quite a high rate (probably because the pages haven't been processed with the new option by other bots yet - it was only added yesterday), so I thought I would stop the bot and check that everyone is okay with it before proceeding. If not, then I will be happy to discuss it and, if needed, go through re-approval. - EdoDodo talk 16:35, 14 September 2010 (UTC)

Seems like a reasonable improvement to the current task. —  HELLKNOWZ  ▎TALK 16:46, 14 September 2010 (UTC)
Bot started again since there were no objections in over a day. - EdoDodo talk 05:15, 16 September 2010 (UTC)

Bot platform

As some know I am developing a bot platform and I like to see if the project received wide support before continuing. d'oh! talk 12:06, 23 September 2010 (UTC)

Town bot

Per WP:DUCK, these contributions by User:Starzynka look like they were created using a bot (most probably AWB). The user talk page says that they were created manually and the user has authorization to create these stubs. I see no link to any bot approval. Does bot policy apply here? Ganeshk (talk) 15:20, 22 September 2010 (UTC)

Here is a related ANI incident. Ganeshk (talk) 15:30, 22 September 2010 (UTC)
Per the bot policy: "high-speed semi-automated processes may effectively be considered bots in some cases, even if performed by an account used by a human editor" and "The community has decided that any large-scale automated or semi-automated article creation task must be approved at Wikipedia:Bots/Requests for approval. While no specific definition of "large-scale" was decided, a suggestion of "anything more than 25 or 50" was not opposed." I suggest this task still get approval, or at least get discussed with relevant WikiProjects, even if it is not completely automated. - EdoDodo talk 15:39, 22 September 2010 (UTC)
I agree this should go through BRFA. It's obvious that these are semi-automated, and the use of subst:PAGENAME to fill in the "official name" and the name in the lead is inappropriate for those that are (disambiguated) (e.g.). Notwithstanding the editor's dubious assertion that they are "completely manual", cutting and pasting a boilerplate template is still "semi-automated" as far as I'm concerned. –xenotalk 15:43, 22 September 2010 (UTC)

Thanks for the responses. I have asked the user to stop creating the articles and get an approval. I think this issue must be taken back to ANI if the user disagrees. User:Александр Мотин has been creating similar stubs. They are under the impression that these do not need approval. Ganeshk (talk) 16:09, 22 September 2010 (UTC)

do you really want to argue that use of a template is "semi-automated", that might well drag in every prod and tag. is there a ruling that approval is required? Accotink2 talk 16:15, 22 September 2010 (UTC)
If by "ruling" you mean "discussion showing consensus of the community" then yes, this was discussed and agreed here. - EdoDodo talk 16:18, 22 September 2010 (UTC)
That discussion dealt with "semi-automated or automated" creation of stubs. Though there were confusion about the exact meaning of semi-automation, it is clear that the majority of editors were talking about bot or script edits. There has not yet been a discussion whether article creation by manual template copying (which is, assuming good faith, what User:Starzynka is doing) is in par with semi-automation. Perhaps this is the time to have one? —  HELLKNOWZ  ▎TALK 16:37, 22 September 2010 (UTC)
i note no discussion of "semi-automated", i take from Wikipedia:Bot policy Definitions: "Bots (short for "robots") are generally programs or scripts that make automated edits without the necessity of human decision-making.
Assisted editing covers specifically lower-speed tools and scripts that can assist users to make decisions but leave the actual decision up to the user (see Assisted editing guidelines below). Any program or tool which does not allow the user to view each edit and give an instruction to make that edit (that is, one which can edit without the operator looking at and approving the change) is considered to be a bot.
Scripts are personalized scripts (typically, but not exclusively, written in JavaScript) that may automate processes, or may merely improve and enhance the existing MediaWiki interface."
there is no assisted editing, or scripts here, rather manual cut and paste of Help:Magic words. Accotink2 talk 16:39, 22 September 2010 (UTC)
"high-speed semi-automated processes may effectively be considered bots in some cases, even if performed by an account used by a human editor" Cut-and-pasting a boilerplate template is semi-automation as far as I'm concerned - especially when you don't even correct the official name to remove disambiguators which do not form part of the official name. –xenotalk 16:55, 22 September 2010 (UTC)

Exactly. This proposal is ridiculous. These articles are manually created. A template which just displays the title of the article is certainly not remotedly "automated". The articles themselves are manual driven. People should be free to create articles were are notable and not have to ask. Silly. Starzynka (talk) 17:31, 23 September 2010 (UTC)

It's as close enough to automated as makes no odds because you don't even bother to fix the obvious manual of style errors. –xenotalk 17:33, 23 September 2010 (UTC)


(edit conflict) The edits were almost all made in under 10 seconds. The bot policy says that "high-speed semi-automated processes may effectively be considered bots in some cases"' and this should definitely be one of them. 10 seconds is a very short time, and even if a template was being copy-pasted that is still, in my opinion, not enough time to do all the appropriate web page opening and checking that manual editing implies. Also, if this really was being done manually, mistakes like this (bracket in lead sentence) would not occur (manual editing implies checking of all content, so the fact that he pasted a magic word is not an excuse). - EdoDodo talk 16:57, 22 September 2010 (UTC)
i saw your comment above; i'm afraid i disagree. Wikipedia:Bot_policy#User_scripts "The majority of user scripts are intended to merely improve, enhance, or personalize the existing MediaWiki interface, or to simplify access to commonly used functions for editors. Scripts of this kind do not normally require BAG approval." Accotink2 talk 16:58, 22 September 2010 (UTC)
I don't see how mass creation of articles falls under "improve, enhance, or personalize the existing MediaWiki interface, or to simplify access to commonly used functions". I believe that in this context commonly used functions is intended as warning/welcoming users, protecting pages, and other basic tasks. Certainly not page creation. - EdoDodo talk 17:01, 22 September 2010 (UTC)

Prop 1 is on the face of it a good idea. However there may be some edits that require a level of scrutiny that is so low that calling it scrutiny might be considered litotic. I'm thinking for example of the monthly maintenance categories, where the manual approach is open all the redlinks, then "ctrl tab, shift alt S" until your fingers fall off. It's still useful to go back at the end and make sure al the red-links turned blue, but that's about it. Rich Farmbrough, 19:31, 24 October 2010 (UTC).

Prop 2 um.. what does "wiki" mean again? Rich Farmbrough, 19:34, 24 October 2010 (UTC).

RF, the discussion on the policy change has been moved here. Ganeshk (talk) 19:43, 24 October 2010 (UTC)

Ganeshbot 5

Hello BAG team, this bot request has been open for more than a month now. The last BAG edit was on 8/30. Can someone from the BAG team please look at the consensus and provide a result? Here is a summary of the discussion from the Gastropod project team. Thanks. Ganeshk (talk) 21:42, 26 September 2010 (UTC)

Doing... - Kingpin13 (talk) 07:38, 14 October 2010 (UTC)
 Done - Kingpin13 (talk) 08:29, 14 October 2010 (UTC)
Thanks Kingpin. It was a long discussion, appreciate you taking time to go thorugh it all. Ganeshk (talk) 11:56, 14 October 2010 (UTC)

I would like to dispute Kingpin13's closing:

  • How many people joined the discussion:
    • Support provided by 15 wikipedians (14 wikipedians and User:Ganeshk).
    • Partly support (for generic articles only) by one wikipedian.
    • Opposed were 4 wikipedians.
  • This sentence from closing is completelly wrong: "because the supporters are the ones proposing this, they need to work in cooperation with the community, to reach a proposal which suits everybody (reasonably)." Why this is wrong: Opposing wikipedians provided no reasonably evidence, that anything is wrong with this bot. Their opposing is based on their personal feeling only and not on wikipedia policies. All wikipedians who edit gastropod related articles are supporting this bot and no one of opposing wikipedians edit gastropod related articles. There is a community of gastropod-related editors and there are only 4 wikipedians who does not edit such articles and who disagree with this bot without any reason. So who needs to works with in cooperation?
  • There was no one logical or reasonable opposing reason mentioned neither in the closing comment nor in the whole discussion.
  • Kingpin13 suggest to "forming a consensus before submitting another BRfA". But there existed consensus about task of this bot at Wikiproject Gastropods for over 6 months and earlier. And only one critic have appeared after such time who initiated this Request. The only PROBLEM is Requests for approval and Bot Approvals Group, who are unable to work swiftly. THERE IS GENERAL CONCENSUS, THAT Bot Approvals Group DAMAGES WikiProject Gastropods (confer with Wikipedia talk:WikiProject Gastropods#Ganeshbot 5 and all talkpages of this WikiProject). Solve it as soon as possible, please, prior it will appear at Arbitration Committee and especially before gastropods-related editors disappear from Wikipedia. --Snek01 (talk) 19:29, 15 October 2010 (UTC)
Hi Snek01, thanks for your points.
The number of Wikipedians who opposed/supported is not as relevant as you seem to presume. The BRfA was evaluated based on the strength of the arguments presented, rather than the number of users who felt personally that the bot should run or not.
My comment that you need to work with the community is not wrong. It’s perfectly true that when proposing something new, the proposers should work in cooperation, to reach a consensus. Simply denying that there is opposition to the bot, or calling the opposition “unreasonable” or based on “personal feelings” is precisely the attitude that lead to a lack of consensus. I spent a about a week on this decision (and have also loosely followed the progress of this bot since it was proposed orignally), and many hours reading through the comments about this bot. I can safely say that there was reasonable evidence provided for why users thought this bot approval should be denied. Simply claiming there isn’t, as you are doing, can be disruptive per WP:IDIDNTHEARTHAT. Because you are unwilling to listen to the concerns, or address them, you are not working in cooperation, thus you can not compromise, thus the request was closed as no consensus.
The Wikipedians working in the Gastropod’s WikiProject are expected to support this bot, since a large number of them seem to have the objective of creating a page on every single slug/snail (or similar). Since the members of the project are users who agree with this objective, using their wide support as a “consensus” is incorrect, as it is a bias sample. You also seem to have the mistaken idea that the only users’ who opinions matter are those who are interested in writing content about the said slugs and snails.
I fail to see how the slow-nature of BAG has anything to do with this. Threatening to take this to ArbCom isn’t constructive (although you are of course free to do so). In no way am I (or any member of the Bot Approvals Group) attempting to stop members of this project from editing Wikipedia. BRfA is a necessary process, and is similar to other consensus-building/judging processes on Wikipedia. It conforms to Wikipedia’s standards, and operates through consensus like most of Wikipedia. If you have a problem with this (the way in which Wikipedia is run) then I am glad I judged the BRfA as I did, since it still seems you are unable to work cooperatively. - Kingpin13 (talk) 21:34, 15 October 2010 (UTC)

Bot flag request

Hi,

I'm not sure of the process involved in requesting a bot flag for the recently-approved User:Taxobot (WP:Bots/Requests for approval/Taxobot 2). I'd be grateful if someone could help me out.

Thanks

Martin (Smith609 – Talk) 05:29, 24 October 2010 (UTC)

Particularly, in view of the earlier problems with Anybot, it would be desirable to first handle the issue raised a few hours ago at User talk:Taxobot. I have not looked at the merits of the issue, but it needs a response. Johnuniq (talk) 06:30, 24 October 2010 (UTC)

Probably erroneous approval for a form of spell-checking (SmackBot)

Resolved
Wikipedia:Bots/Requests for approval/SmackBot Task XI

In 2006, SmackBot (talk · contribs) was approved to (among other things) "Correct caps in headers" but this seems to be a form of typo-fixing or spell-checking contrary to WP:BOTPOL#Spell-checking. In particular, this has caused problems for suchlike "==Prime Minister==" [1] and "==World Series of Poker==" [2]. Should approval for this line-item of the linked BRFA be revoked? (or restricted to a well-defined set like ==External Links== → ==External links==, ==See Also== → ==See also==, etc?) –xenotalk 14:35, 25 October 2010 (UTC)

I don't think this is a form of spell-checking, Xeno, but applying a general rule to headers. Problem is, it is too error-prone (seen the examples you gave). May I suggest the latter solution, allow it to do 'standard' headers (there are possibly more, but these are obvious). --Dirk Beetstra T C 14:43, 25 October 2010 (UTC)
That would be fine, but whatever rules Rich is using now aren't appropriate for unattended editing. –xenotalk 14:49, 25 October 2010 (UTC)
A list of exclusions would probably be too tedious to compile. So the other option of making a list of common accepted corrections is preferred. I don't oppose these small changes, as long as they are almost never controversial. —  HELLKNOWZ  ▎TALK 16:49, 25 October 2010 (UTC)
It appears this was human error rather than a poorly programmed regex, so I'll mark this resolved. –xenotalk 18:25, 25 October 2010 (UTC)

Waiting period

Is it normal to have to wait weeks in between approval periods? Perhaps there's a need to expand the BAG. Magog the Ogre (talk) 07:48, 11 November 2010 (UTC)

Just be aware that my PHP code is crying itself to sleep because it believes it lacks purpose. It wants only to perform the action for which it was created. Magog the Ogre (talk) 02:51, 14 November 2010 (UTC)

Review

Rhetorical question apparently not aimed at bot operator closes case without reply from WP:BAG. Regards, SunCreator (talk) 22:52, 19 November 2010 (UTC)

If you intend to continue with the request, feel free to re-open it. I waited 6 days for an answer before closing it, after your comment that seemed to say that you didn't intend to continue. Anomie 05:14, 20 November 2010 (UTC)

Taxobot 5

Wikipedia:BRFA#Taxobot_5 is a relatively simple task but has been awaiting approval for almost a month. It's holding up template development so it'd be nice if I could run it over the weekend. Would a member of the team be able to evaluate it, please? Thanks, Martin (Smith609 – Talk) 18:51, 26 November 2010 (UTC)

Super, thank you! Martin (Smith609 – Talk) 19:05, 26 November 2010 (UTC)
The trial's now complete so if anyone gets the chance to give it the nod, that'd be great. Thanks (-: Martin (Smith609 – Talk) 21:42, 27 November 2010 (UTC)

Need BRFA application for functionality changes?

In the near future I'll be working on some changes to how SuggestBot handles its regular users, and am wondering if it will require to file a BRFA or not. Currently users can get regular suggestions by signing up on a list, and then we process the list about once a month and post suggestions to their user talk page.

On the Norwegian Bokmål Wikipedia we have a system in place that uses template inclusion for signing up, and which allows inclusion on sub-pages in userspace so those who prefer can get the suggestions posted there (e.g. "/Miscellaneous/Suggestions"). We'll also shortly run the bot automatically there, greatly reducing the cost of maintenance and making sure users actually get these recommendations regularly.

Do these kind of changes require me to file a BRFA for an added task to SuggestBot? Also, let me know if there's anything I've missed. Cheers, Nettrom (talk) 13:33, 22 December 2010 (UTC)

Since the actual task is almost identical and only the sign-up method is changing, I don't think it will be a problem. - EdoDodo talk 16:17, 22 December 2010 (UTC)
Agree that changes to non-mainspace for your task can be made without any new BRFAs. —  HELLKNOWZ  ▎TALK 17:11, 22 December 2010 (UTC)
Thanks for the responses! I'm happy to hear that a new application isn't needed. Cheers, Nettrom (talk) 19:38, 23 December 2010 (UTC)

Orphaned BRFA

So there's Wikipedia:Bots/Requests for approval/YouMeBot just sitting and not linked from anywhere and I'm not sure if it's a serious request or not. It was created by the bot account itself which has been indef-blocked for its username and no bot owner has been identified. VernoWhitney (talk) 22:13, 22 December 2010 (UTC)

Easily fixed. Anomie 02:49, 23 December 2010 (UTC)

Stupid comments by community members

It's obvious from my post that I don't really know what is going on with the bot, but I thought I would try commenting and learning. Big mistake. I've been put in my place.[3]

I would hope that part of the process would be community members can learn about it. There is no reason for this. --Kleopatra (talk) 18:28, 29 December 2010 (UTC)

I see nothing at all there that is "putting you in your place"; at worst H3llkn0wz's tone might be read as a little defensive. It seems to me that H3llkn0wz just explained where you were mistaken and gave some more detail as to why he thinks more input should be sought. What specifically do you take offense to in H3llkn0wz's message? Anomie 19:24, 29 December 2010 (UTC)
Yes, defensive and ignoring my request for a link, but shoving the same policy at me, again, without linking. I'm a little overwhelmed lately by how much policy is defensively quoted without links to the actual policy. When I eventually find the policy, it's usually not what is quoted, though. However, a link, sometimes, would be nice; for a policy administrator, I consider it necessary. But, those in the know on wikipedia seem often defensive when a link to policy is requested.
I am not, however, interested at this point, and I shall unwatch this page. Apparently it is also resident evil to change one's mind about posting a comment, so I was stuck leaving this up without withdrawing it.
WP:BRFA doesn't get many comments on bots to begin with. I think being defensive and ignoring input or questions is not a good way to get comments. --Kleopatra (talk) 19:42, 29 December 2010 (UTC)
WP:BOTPOL is the policy, particularly the parts "In order for a bot to be approved, its operator should demonstrate that it: [...] performs only tasks for which there is consensus" and "The request will then be open for some time during which the community or members of the Bot Approvals Group may comment or ask questions. The decision to approve a request should take into account the requirements above, relevant policies and guidelines, and discussion of the request." Anomie 20:56, 29 December 2010 (UTC)
OMG. The question is about this, "Not actually changing the output is by default against bot usage..." Now, I am unwatching this page. Please don't not answer my question again, as there is no point. When someone asks a question and the response is defensive non-answerment, there's usually a reason. Namely, the answerer was defensive already. It might be easier and less defensive all around to provide links when first asked. Thank you. --Kleopatra (talk) 21:45, 29 December 2010 (UTC)
Mass changes that don't affect the content of a page and don't have a compelling reason otherwise are usually not done. It's a waste of resources, and it clutters up watchlists. The bot policy mentions cosmetic changes, it's along those same lines. Gigs (talk) 21:53, 29 December 2010 (UTC)

Centralized List of Minor Fixes

Hey BotOps, I was wondering if anybody knew a place where there was a full list of cosmetic changes to be done to an article. The types of changes that do not merit their own bot task, but should be done if an edit is being done anyway. I know WP:AWB has an impressive library of general fixes, is there a place where those are listed? Tim1357 talk 16:02, 2 January 2011 (UTC)

Do you mean WP:GENFIXES? -- Magioladitis (talk) 16:13, 2 January 2011 (UTC)
There were some bots that dealt with longer lists of common fixes, like piping and stuff, that AWB doesn't have a full repository of. I was wanting to compile such a list of non-controversial fixes myself; including if they need human oversight or not. Perhaps on a project page. There's also Wikipedia:WikiProject Check Wikipedia and smaller ones, like Wikipedia:WikiProject Fix common mistakes. I'm sure there's more. I wonder if a list can be made and then approved by BAG as a list of changes that don't need approval or discussion. —  HELLKNOWZ  ▎TALK 16:33, 2 January 2011 (UTC)
I wrote User:Magioladitis/AWB and CHECKWIKI. -- Magioladitis (talk) 16:35, 2 January 2011 (UTC)

OgreBot -> need permission for task?

Currently OgreBot is approved to delink images that have been OKed by admins and approved users and prepare them for a batch deletion. As part of the process, I'd like to have the bot move the talk pages associated with the file to the new name; e.g., [4] (note that file talk pages for commons content are not deletable as G8). It's something I'm currently doing by hand. Would this be something I would need to get approval for, or would it fit broadly enough into the current task? Magog the Ogre (talk) 03:09, 11 January 2011 (UTC)

I don't think this fits under the current task directly, as this is a completely different action. If you fill a BRFA, it can be resolved fairly quickly. —  HELLKNOWZ  ▎TALK 10:41, 11 January 2011 (UTC)

Can you shut down Snotbot?

It looks like the robot is mindlessly placing top level tags on all articles that have a backlog in patrolling. IMHO no robot should be placing top level tags on articles. Worse than that it "giving orders" based on non-existent policies. Thanks. Sincerely, North8000 (talk) 01:27, 19 January 2011 (UTC)

See Wikipedia:Bots/Requests for approval/Snotbot / Wikipedia talk:NPP#Back of the unpatrolled backlog; the bot is approved for this task. –xenotalk 18:25, 19 January 2011 (UTC)
The bot was approved after more than a week with no objections on the assumption that Wikipedia_talk:NPP#Back_of_the_unpatrolled_backlog was enough for rough consensus and that the task identifying new pages that otherwise fall off the radar is useful. "Top-level" tag did not seem like a problem (at least I am unaware of any agreement that bots aren't allowed to do this). As far as I see it, the bot is not "mindlessly placing top level tags" as it has clearly defined unambiguous rules when to place the said tag. —  HELLKNOWZ  ▎TALK 18:41, 19 January 2011 (UTC)
I am the operator of the bot. The bot places an {{New unreviewed article}} template on the top of any pages which have existed for more than 30 days and haven't been marked as patrolled by a new page patroller. The only pages that are modified are unpatrolled articles which no longer appear on Special:Newpages. The bot should never tag articles which are still displayed in the backlog at Special:Newpages. If it is doing that, then please let me know. The {{New unreviewed article}} template is perfectly appropriate, because the article is new (31-32 days old in most cases) and the article is unreviewed. The template is not intended to imply that your article is necessarily deficient in any way, just that the article hasn't been independently reviewed by anyone yet. SnottyWong chat 19:17, 19 January 2011 (UTC)
Out of curiosity, why does this need a tag (and a rather large one at that)? This isn't really useful information to a reader. It seems like the same thing could be accomplished with something less visible. Mr.Z-man 21:03, 19 January 2011 (UTC)
That was a a part of what was behind my complaint. The first half of my complaint was that robots shouldn't be placing top level tags on articles, not that such was violating any specific rule/agreement regarding that. Further to that I have never seen robots doing that nor being allowed to do that.
The second half of my complaint was that the robot's tag is "giving orders" implying non-existent policy. Implying that primary author on an article can't remove a top level tag invented and worded by one person and placed by a robot. Sincerely, North8000 (talk) 04:41, 20 January 2011 (UTC)
FYI, I was largely responsible for the initiative that launched the creation of this bot. One needs to understand that this encyclopedia concerns three large groups of people: those who read it, those who contribute the articles, and those who keep it free of crap (there is often some overlap). Maintenance templates on articles also serve three distinct purposes:
  • They add that article to an appropriate hidden category that is watched by editors who are interested in improving articles that have been improperly created.
  • They warn the authors that the articles require attention.
  • They warn the readers that what they are reading may not be entirely accurate, and in the worst case scenario may even be a blatant advert, a hoax or a personal attack page.
All these aspects should be borne in mind before criticising the work of others just because they affect one user's contribution(s). Any reader or editor is more than welcome to consider joining one of the many projects that help to maintain quality. Kudpung (talk) 07:57, 20 January 2011 (UTC)
The bot should be adding a |bot=yes, |source=bot,|bot=Snotbot or similar parameter to the templates. Even if the parameter doesn't do anything presently, it then allows changes to be made at a later dates to only the tags placed by the bot, or only the tags placed by human editors. This way we we can easily switch between invisible and visible bot-tags, something we can presently do. This is an idea I would have suggested before, but I stopped following that discussion after Snottywong started coding instead of me. - Kingpin13 (talk) 08:08, 20 January 2011 (UTC)
North8000: FYI that template was not created solely for this purpose. It was an existing template which was already being applied to articles regularly for other reasons (e.g. for articles created by the New Article Wizard). Instead of inventing a completely new tag or new category that no one would know about, we decided it was better to use the existing infrastructure. Also, some patrolling tools (like twinkle) already automatically remove that particular template when articles are patrolled. Making up a new hidden category that gets added to the bottom of the article would be more difficult to maintain, because many patrollers wouldn't even know what it was and would neglect to remove it when patrolling. Kingpin, I think the bot parameter is a good idea, I'll add it to future applications of the template. SnottyWong soliloquize 14:44, 20 January 2011 (UTC)

← I think the most important question is: are the new page patrollers giving these tags priority when they are patrolling? i.e. ensure they've all been removed before moving on to the regular special:Newpages patrol? –xenotalk 14:49, 20 January 2011 (UTC)

The bot has only been running for a few days, I'm sure most new page patrollers are not even aware of what it's doing yet. I've added this brief mention at WP:NPP, but most established new page patrollers aren't reading that page very often. Eventually, people will probably start to take notice and work on clearing these articles. SnottyWong soliloquize 16:44, 20 January 2011 (UTC)
Although nobody really addressed either of my 2 points, (and this bot is setting a bad new precedent on both of those,) I do sincerely appreciate that folks took to respond and to try to explain things, and also apologize if my tone anywhere was impolite. Sincerely, North8000 (talk) 00:50, 21 January 2011 (UTC)
What am I misunderstanding? The request and approval was to place a hidden category,[5] The bot is also putting boilerplate at the top of articles, e.g. [6] leading to [7]. Snotty has said that the consensus changed [8] but was that discussed as part of the approval? Please clarify. Thincat (talk) 15:55, 21 January 2011 (UTC)
As an FYI, this is not the first bot to be approved to place tags at the top of articles. If you don't think that the wider community (i.e. outside of those who commented at WT:NPP) would approve of this task, I suggest you initiate a discussion at some central venue (WP:VPP, perhaps) and see if the community shares your trepidation. –xenotalk 16:06, 21 January 2011 (UTC)
Well, there seem to me to be two issues: (1) would/does the wider community approve this and (2) was bot approval given for tagging these ariicles? For me the latter question is the more immediate. By the way, when I looked for an example tag transclusion (not bot action) the first example I found had been hanging around for over six months (Mack McCarter) so hopes for timely action may be pious. Thincat (talk) 16:20, 21 January 2011 (UTC)
(2) It appears that the operator took it upon themselves to change the functionality during the trial without explicit approval (which is suboptimal) but they noted the change in functionality prior to final approval. As regards (1), if it becomes evident during the discussion at a central venue that the community does not want visible tags to be placed by the bot, then the bot can start putting hidden categories, listifying the pages needing review, or something of the sort. –xenotalk 16:23, 21 January 2011 (UTC)

I think there is some misunderstanding all round. As up to 300 pages per day are escaping unpatrolled, untagged, (including attack, spam, hoax, and non English pages, etc) into the Wikipedia, a suggestion was made in October at the VP to extend the 30-day backlog period. The request was turned down with the reasoning that although it would be technically possible, the volunteer new page patrollers should simply work harder; some Wikipedians (well, me actually) felt there could therefore be another way of addressing the problem. This project is in its infancy but although what has been achieved so far is one small step for improving the system of new page patrolling, it's already a huge leap forward towards better quality control.
First and foremost, at this stage, no new page patrollers are even supposed to be working on the pages tagged by Snotbot, and they probably won't be for a while to come. To explain clearly:

  1. New Pages are tagged as new on creation (but only the ones made by the wizzard get the ugly banner) and get a 'mark this page as patrolled' button put on them.
  2. They enter a page called Special:New Pages where they are highlighted in yellow until the 'mark this page as patrolled' button gets pressed, or an edit is made that auotmatically marks the page as patrolled
  3. Special:New Pages stays live for 30 days, after which, all still unpatrolled new pages are allowed to sneak into the Wikipedia through the back door.
  4. These pages are unpatrolled because they are in such a poor condition, many patrollers lack the experience to know what to do with them. They often lack all identifying marks such as cats, tags, stubs, talk page templates, etc.
  5. These pages are far too many to be manually patrolled by the experienced editors and admins who nevertheless occasionally have a stab at the bottom of the 'unpatrolled' list.
  6. At last, with Snotbot, we now have a way of getting these pages traced, and tracked.
  7. What happens to them next is anybody's guess, but we're working on that too.
  8. Nobody cares if they now have an ugly template on them placed by Snotbot - the creators, all SPA, have long since moved on from Wikipedia leaving us to clean up their mess, and the content of those pages is a lot uglier than any banner that might be on the top of therm.

Consensus has not changed, the use of the word consensus was inadvertently used by Snottywong instead of saying 'method', but in prìnciple nothing has changed from the original project. It was a change to a technical issue, such as for example instead of going from A to B via Y, we now go from A to B via Z. This bot is not doing anything to your articles, and cannot possibly be setting any bad new precedents. However, while I'm writing this, another 10 pages have joined the list, adding to a total of anything up to 100,000 pages by the end of the year that we are at least now able to know are in need of urgent attention, and where they are. --Kudpung (talk) 17:04, 21 January 2011 (UTC)

So, the patrolling process is broken (huge backlog) and this does nothing to help that. My note /advice is to reduce the scope of the patrolling review to serious problems that apply to brand new articles. (notability, advertisements, etc.) Reviewing for other things is very laborious and premature. North8000 (talk) 18:40, 22 January 2011 (UTC)
Uhh, so what you're saying is if we ignore the problem, maybe it will go away? Our readers appreciate being warned about potential issues in an article, and I dare say would like to know if an article has been peer reviewed (since the editor who originally came up with the content is unlikely to be checking for OR, NPOV, etc. And of course it's much more difficult to review one's own work), and those editors who are keen to review work need some way to find articles which need to be reviewed, or their eagerness and talents are wasted.. - Kingpin13 (talk) 19:13, 22 January 2011 (UTC)
No, I was giving a suggestion for fixing the problem vs. just decorating it. :-) Sincerely, North8000 (talk) 19:17, 22 January 2011 (UTC)
Well it's not possible for the bot to fix the problems, but it can direct other's efforts towards places where there are likely to be problems to prevent editors from wasting energy searching for problems rather than fixing them.. I fail to see the problem with that. - Kingpin13 (talk) 19:18, 22 January 2011 (UTC)
I was just responding to your semi-sarcasm, (i.e. my point wasn't dumb) not arguing the point further. North8000 (talk) 18:47, 23 January 2011 (UTC)

North, just decoration? Your own comments are disparaging those users who are right now in the process of doing something about the backlog - please check the timeline by kindly giving a moment to read the preceding threads. You'll se it's in it's first week of operation of the first of many steps still to go. It has to start somewhere, or are you just insensitive to the efforts of those who are prepared to give up their time and get such a project underway? IMHO, just throwing negative comments around a workgroup forum is not the best way to encourage those who can and do seek, and implement improvement. Kudpung (talk) 19:12, 23 January 2011 (UTC)

Again, I was responding to your sarcasm which implied that what I said was dumb, that was my only topic and intent in the last post. Sorry to anyone if that was not clear. North8000 (talk) 20:39, 23 January 2011 (UTC)
Nobody is implying you or your views dumb, they're simply disagreeing, it's not the same thing. I fully understand that your suggestion is made in good faith, and we simply have differing views as to what's the best method for dealing with article issues. It would be nice if you extended the same courtesy towards me, and not presume that I'm being sarcastic when I'm not being: my suggestions too are made in good faith and are entirely serious. - Kingpin13 (talk) 20:51, 23 January 2011 (UTC)
So, what do you call "Uhh, so what you're saying is if we ignore the problem, maybe it will go away?"....(mis)restating my position in a way this is obviously ludicrous?" That was the ONLY thing that I was responding to. North8000 (talk) 23:23, 23 January 2011 (UTC)
That's the impression I got from you suggesting that reviewers should only check for those errors which are easy to spot, and should ignore other errors. However, since that's apparently not what you meant, let's not get hung up on it, I've struck my comment and apologise for any offence caused. - Kingpin13 (talk) 23:46, 23 January 2011 (UTC)
Thanks. I did suggest a narrower scope of initial review, but it was a sidebar to the main discussion. My main thought is that if articles with severe initial problems (notability, advertisements) are slipping through due to overloaded patrolers spending a lot of time with general article improvement type review, then it might be good to narrow their scope to the point where every new article gets at least a quick review for serious problems. Especially since the editors already know that a brand new article needs a lot of work. Sincerely, North8000 (talk) 15:25, 24 January 2011 (UTC)
  • Ok, well to sum up, it appears we have one somewhat cantankerous editor who doesn't appreciate our efforts. I think that if North8000 believes this is as big a problem as he/she is making it out to be, then I would encourage North to start a centralized discussion to see if there is a consensus that visible tags should not be placed at the top of unpatrolled articles. It's true that the original request for the bot said that it would place the articles into a hidden category. During the bot approval process, the discussion about how the bot would operate evolved and we reasoned it would be more efficient to add them to the existing new article tracking infrastructure provided by {{New unreviewed article}} rather than create a whole new redundant tracking system. As xeno points out, I did mention this during the bot approval process in an attempt to ensure that the change wouldn't be problematic. In any case, if the templates are widely perceived as inappropriate for some reason, there is already a very easy way to hide the templates. With the exception of the first few days of bot activity, the bot includes a "bot=yes" parameter on all templates. So, the {{New unreviewed article}} template would just need to be trivially updated to hide the banner when the bot parameter is present. I'm not saying we should do this now, since I (like seemingly everyone else here) disagree with North8000 about the inappropriateness of the banners. Basically what happened here is that North created a new article, it wasn't patrolled for 30 days, the bot tagged it, and North had some kind of emotional reaction rooted in his perceived ownership of the article. I was alerted to the complaint, and I tried to resolve it as quickly and civilly as possible. I'm sorry that this issue upset you so much, North, but I can't see what more anyone could have done to help you. Again, in my view, a single editor's complaint shouldn't be enough to force an approved bot to change its behavior. I don't see any point in continuing this discussion any further, but please do let me know if a more centralized discussion on the topic becomes necessary. Thanks. SnottyWong squeal 03:45, 24 January 2011 (UTC)
"Basically what happened here is that North created a new article, it wasn't patrolled for 30 days, the bot tagged it, and North had some kind of emotional reaction rooted in his perceived ownership of the article." perfectly summed up Snotty. Which also proves wthout a shadow of a doubt that the the new system is doing exactly the job it was intended to do: getting SPA to pull their socks up and come back and clean up their articles ;) For these reasons, experienced editors often scrutinize the editing activities of new editors and SPAs carefully in a discussion to discern whether they appear to be here to build an encyclopedia (perhaps needing help and advice), or alternatively edit for promotion, advocacy or other unsuitable agendas. (from WP:SPA). Can we shut down disruptive criticism? Kudpung (talk) 14:02, 24 January 2011 (UTC)

Why don't you read what I actually said instead of imagining the above baloney? What happened at the article itself was too minor to worry about. I brought up two issues:

  • (IMHO) robots shouldn't be placing top level tags on articles
  • the robot's tag is "giving orders" implying non-existent policy. Implying that primary author on an article can't remove a top level tag placed by a robot

If you want to continue to evade those questions, that's no big deal. I was ready to leave this thread a long time ago. But invention of false things about me constitutes personal attacks. Please stop that. North8000 (talk) 14:39, 24 January 2011 (UTC)

In response to your two issues above:
  1. Your humble opinion is duly noted. There is no precedent for barring bots from placing top-level tags.
  2. The same "orders" are given by the Article Wizard on all pages created by that process. We didn't create the template which gives the "orders", it has been used for other purposes for well over a year. And I dispute the notion that it is "ordering" you to do anything. It is certainly strongly recommending that the article creator doesn't remove the tag (for obvious reasons, since the article's creator is not an unbiased reviewer), but it is not preventing you the creator from doing anything, or threatening the creator with punishment if they remove the tag. Therefore, I think your criticism is unfounded.
Anything else? SnottyWong prattle 15:27, 24 January 2011 (UTC)

Thanks, that finally addresses what I brought up initially. And, as an aside, I never had any complaints about your handling of this, except / until you recently wrote "rooted in his perceived ownership of the article" and Kudpunbg repeated it. This is false twice over. But lets move on. Thanks are due to everyone trying to make Wikipedia better, even when I don't agree with some of the particulars. Sincerely, North8000 (talk) 15:37, 24 January 2011 (UTC)

Query

Kumioko (talk · contribs) is running thousands of edits a day on average using AWB to tag pages as part of WP:WPUS. Shouldn't he be doing so with a bot flag, given the volume and rate of his editing? Imzadi 1979  05:34, 22 December 2010 (UTC)

Given that just looking through his latest 50 contribs I see that his edit rate was up to 11 tags in a minute at one point and apparently mistagging some recently (see e.g., User talk:Kumioko#WikiProject United States? and User talk:Kumioko#Please stop DC-tagging WMATA stations for time being) which indicates they're apparently not paying attention to each edit I'm inclined to agree that this falls under WP:BOTPOL. VernoWhitney (talk) 16:44, 22 December 2010 (UTC)
Thanks for letting me know about the discussion VW.
  1. First let me mention that most of the "mistagging" was jsut asking me to calrify a little better, which I have tried to do. Not actually mistagging. The situation with the WMATA stations is a little different, that was a difference of opinion. That came about because I was adding the District of Columbia project and associated articles to WikiProject United States. I was just converting so there was no need to double check wether they were valid at that time. Also, the user that brought that up inappropriately removed the US/DC tag from articles that clearly fell into the scope of US/DC (there were some exceptions though such as stations in Maryland and Virginia.
  2. To the original issue of my edit rate and needing a bot. The speed of my edit is directly preportional to the amount of manual checking I need to do. For example if I am taging a page for a File, Category or template that related to US I can do it extremely fast because in most cases the page was blank and this is the only thing there. For others such as the assessments it can take several seconds to check the article and then process it but I usually grab the assessment from a project thats already there if its missing for US (I do this using Regex coding). In most cases the high edit rate is do to working on talk pages and it requires very little time to see if there is an error before I move on. I also frequently use 2 instances of AWB at the same time so when the first one is saving I am reviewing the second and I alternate back and forth. If the community want me to request and use a bot for some I can, the problem here is that many do require review and with the hanti bot zero defect sentiment lately I feel its rather better to do it manually and be able to review them than do them automated and get blocked because my bot did something it wasn't supposed to do. I freely admit and feel compelled to state in advance that if I am required to use a bot I will not be able to for many of the changes due to the manual nature and cannot guarantee that the bot will not misedit some of the pages due to the nature of the edit being made. Additionally, with the antibot sentiment I mentioned before I have no desire to run my own bot so all you will be doing is A) creating more work for others and B) slowing down my efforts in getting WikiProject United States up and running. --Kumioko (talk) 17:09, 22 December 2010 (UTC)
I at least am not talking about bot per se, just a request for approval for doing what you've been doing to ensure that there are clear definitions of what's being done. See Wikipedia:Bot policy#Assisted editing guidelines, and less importantly the section "Bot-like editing" above it. VernoWhitney (talk) 17:37, 22 December 2010 (UTC)
I am familiar with both and both actually state in some way that I don't require using a bot in this instance. Other than the volume I am not doing anything that should require approval. For further clarification I provide the following explainations:
  1. I am currently concentrating on US related items that do not already have the US banner. I have also been actively monitoring new articles and add it as they appear or within a couple days. I typically do not include articles that have another state or US related project but I have made exceptions to this on occassion based on the importance of the article or as it pertained to the US as a whole. This includes Articles, templates, categories, files, etc. There is an end in sight but it will be a couple months.
  2. I previously asked if I needed to request a bot for the creation of talk pages because the rules state creating pages needs one and I was told no. That talk pages are not included in this rule.
  3. The first "Bot-like editing" in fact states "Note that merely editing quickly, particularly for a short time, is not by itself disruptive." So to this I ask, is my editing disruptive and if it is how (so that I may adjust my practices)?
  4. To the second "Assisted editing guidelines" I have been granted permissions to use AWB which is a "tool". This in part covers the approval mentioned by the first and second sentences of this section.
  5. Regarding the third sentence. Concensus has been established that tracking the articles and items in the scope of US is desired and as further evidence of this, previous to my efforts to WPUS, the mechanism was setup for the tracking of these through the Class assessment structure of the project.
  6. Precedent has also been established for these types of edits by users with far more edits (and more edits a day, a week and a month) than I who are currently performing these types of edits with their main accounts. It also states that the edit and tool should be noted in the edit summery, which it does.
  7. As a side not this emphasizes the point I tried to make prior that the rules about running bots and the hrole of BAG/BRFA needs to be clarified. As written the rules are simply to vague and leave to much room for confusion.
In summery of the above I do not believe that bot approval is required for this ongoing task but if concensus determines it is then I will start passing off the tagging to someone with a bot or the desire to run one. I foresee the chances of mistakes are too high as an automated task without significant drawback or errors so I do not want to be the one blamed when it starts drawing criticism. This will significantly slow the tagging process but at the same time will free me to do other project related things so just let me know. I hope this helps to clarify but please let me know if you have any more questions or comments. --Kumioko (talk) 18:22, 22 December 2010 (UTC)
It's not a matter of you needing to run a completely automated bot, it's a matter of you at present acting like one: you made 4,901 edits yesterday alone and thousands more each day before that. That seems pretty much like a high speed, high volume of edits situation to me. Perhaps a BAG member could comment since they presumably deal with this more often? VernoWhitney (talk) 18:35, 22 December 2010 (UTC)
You ignored the most relevant sentence from WP:BOTPOL#Assisted editing guidelines: "In general, processes that are operated at higher speeds, with a high volume of edits, or are more automated, may be more likely to be treated as bots for these purposes". You are obviously making a high volume of edits at a high rate of speed. I'm not sure whether there are actual accuracy concerns being raised here or if it's just "Kumioko should be using a bot flag for this to avoid clogging watchlists and such". Either way, I'm leaning towards asking for a BRFA for this task that could probably be speedily approved at this point, but I'm open to comments. Anomie 18:41, 22 December 2010 (UTC)
Fair enough, just let me know what you want me to do. I certainly can't say I never made a mistake but if I did and found out about it I did try and go back and fix it. Regarding the watchlist thing I mark most of the talk page edits as minor per a request a week or so ago so most of them shouldn't show up on most watchlists. --Kumioko (talk) 18:54, 22 December 2010 (UTC)

(edit conflict) Some points of the above discussion seems to focus on whether these edits are necessary or not. This is not the scope of this talk page. It is human judgement and human decision on how to use their editing tools. The matter in question is the method used, namely, is the user (Kumioko) paying due attention during his work? Can anyone point out any problems so it can be judged if the user is indeed committing mistakes or his oversight of edits is sufficient. See, for example, where another user has filled a BRFA for a task like Kumioko's. So this definitely falls under near-bot-like. The question is really if the user is making mistakes due to the speed of editing that a closer look would not make and whether the community can live with these. —  HELLKNOWZ  ▎TALK 19:00, 22 December 2010 (UTC)

In the spirit of full disclosure and in fairness to the issue at hand let me clarify a couple things so you can make a better judgement about whether I need a BRFA. I would also like to mention that I have admittedly made a few mistakes along the way which I suppose is natural if your doing 1000 plus edits a day. But I went back and fixed them whenever they come up. They are fairly rare and usually are due to that particular page doing something out of the normal (such as having importancelow instead of importance=low which caused me to add importance=low again for about 10 articles which I fixed and then added some regex code that would watch for the bad parameter values). With that said here is a better description of the edits I am doing to help you decide which route to take, which at this point seems like a BRFA: It is quite long and I can provide the code/regex if requested were applicable.
I perform the following actions as primary edits:
  1. Add WPUS if missing to US related articles and usually were there is no other US related project present (this is done particially with regex)
  2. If the assessment is missing from a WPUS tag and certain other projects have an assessment I make the WPUS assessment mirror that (For example if WPShips has B WPUS will reflect B). I only do this for about ten of the more common, active and more reliable projects such as (Wisconsin, NRHP, WPBiography, WPMILHIST and WPLAW). This is also donen with regex.
  3. I remove duplicate blp=yes parameters from the WPBiography banner (Done with Regex)
  4. I Move WPBiography to top billing of WikiProjects if living = yes (Done with Regex)
  5. I move some Templates around if out of place according to Wikipedia:Talk page layout. (Mostly done with regex but some manual find and replace
    1. Image/map/diagram requested goes below banners. (Done with Regex)
    2. Article history related items, Skip to talk, Talk header and a few others go above (Done with Regex, some built into AWB now though)
    3. If comments appear before Wikiproject banners I move the banners above them (Done with Regex)
  6. I remove some deprecated parameters from Wikiproject banners (like nested) (Some Regex & some standard find and replace)
  7. I replace WikiProjectBannerShell1 to WikiProjectBannerShell|1
  8. Standard AWB built in talk page related changes
  9. I move blp=yes from the bottom of the wikiproject banner shell to the top directly before |1 if it appears at the bottom of the banner (Done with Regex)
I perform the following edits only if I am making a more significant edit
  1. Fix redirects for Talk page templates according to logic I wrote using a combination of C# and regex here. This list was accrued based on a list from Rich F, Magioloditis and myself but consolidated into one list for us all to use. I continue to refine it as new templates are created, deleted, changed etc. or as I am notified.
  2. I deleted some uncommon and uneeded fields from several WikiProject templates if they are not being used (i.e. = blanks) or if certain ones =no. For example if |portal3-link= on WPMILHIST is missing or no I delete that empty unneeded parameter. This makes it easier to update in the future, reduces the clutter on the page, is in keeping with the instructions for that WikiProject (they state if the parameter is not needed don't add it) and reduces the amount of space the the page takes for each consecutive historical save.
  3. I delete the spaces before and after == in the section headings
  4. I fix some broken bracket syntax (some Regex & some manual find and replace)
  5. I fix some broken or irregular HTML formatting (some Regex & some manual find and replace)
In summary, although there are some things here that could be done with a bot, many require watching. In the case of the high speed edits of late, if you were to review them, virtualy none required more than a glance to see what had change before hitting save. Please let me know if you have any comments or questions. If you have any suggestions for further improvements I will take those too. --Kumioko (talk) 19:41, 22 December 2010 (UTC)
A few questions:
  • What is "WikiProjectBannerShell1"?
  • Is there consensus anywhere for this "fixing" of talk page template redirects?
  • Is there consensus anywhere for this cluttering of the diffs by taking spaces out of section headers? Particularly since the "Post a comment" button includes these spaces, this seems particularly pointless.
  • What is the "broken or irregular HTML formatting" you fix? I hope not pointless edits like "<br>" → "<br />" or vice versa. Some things aren't even worth a minor edit while doing something else to the page.
Anomie 20:39, 22 December 2010 (UTC)
No problem. Answered in the order asked:
  • {{WikiProjectBannerShell}} is used to combine multiple projects into a group for a variety of reasons. Basically though they reduce the amount of space they take up on the talk page when you have more than 3 or 4 banners.
  • Concencus was garnered when the decision by the community was made to "standardize" the 1300ish WikiProject templates (except 3 (WPUSRoads, MILHIST and another one I don't remember at the moment) to all start with the standard WikiProject) instead of the myriads of things they used to be. This makes doing many things easier including identifying what projects are there (some names are and where more meaningful than others), fixing problems when programming bots (take a look at the code I linked to above. In the current schema if you want to capture all the projects and their redirects to do a bot change to Wikiproject banners you need to have much much much more code to do things).
  • On the header spaces issue again I only do this as a minor edit but I don't think its cluttering the diff at all. And honestly I never liked that it includes spaces by default and rather thought of it as one of those little flaws in the programming that knowone has ever taken the time to fix (like the painstaking process for renaming/moving images for example) In most cases its very obvious what I did and can be identified at a mere glance. Removing these does a couple things. It makes the talk page look less cluttered and easier to read, it again makes it easier to program changes, it takes up less space when saving (if you have 1, 000, 000 pages and they all have 10 versions in history and they all have 20 spaces in the headers, thats a couple GB of space saved when we delete the spaces. Not a huge issue I grant you because GB are cheap in the grand scheme of things but it all adds up and it is a reason I do it.
  • Although the breaks are one of them I am looking more for things like blockquote, center, big, strike through, bold, and a few others. Im not just looking to make sure they are closed but that the actually have content. If there is an opening and closing for center for instance with nothing inside it I delete it (and usually check the result in my contributions to make sure it doesn't scew something. --Kumioko (talk) 21:03, 22 December 2010 (UTC)
  • Oh, I know what {{WikiProjectBannerShell}} is. But you talked about changing WikiProjectBannerShell1.
  • So you're just bypassing redirects to WikiProject banners, and not other banners such as {{talkheader}}?
  • It's considered by some (including me) to be "cluttering the diff" because the diff is full of lines showing that you removed that whitespace. It's particularly a problem on active pages where some people may keep up with ongoing conversations by looking at the diff since their last visit; the real edits by people in the discussions are swamped by useless whitespace-only changes. It's a matter of opinion whether removing the spaces makes them "less cluttered" or "easier to read", as some may find the markup cramped and harder to read without those spaces. You're also right the GB are cheap, and IIRC the way revisions are stored may mean it takes more space for the revision changing it than it would to leave them alone. And none of that really addresses the question of whether there is consensus for this to be done on a large scale.
  • So you do uselessly change BRs to add or remove that slash? As for paired tags with no content, do you also look at the article history to see if it's the result of vandalism?
Anomie 22:02, 22 December 2010 (UTC)
Again commented in the order received
  • Thats because sometimes the template is broken and says WikiProjectBannerShell1 instead of WikiProjectBannerShell|1 with a pipe in between
  • No if you take a look at the link I proveded above you will see at the bottom that I also standardize some other templates but again only if I am doing something else more significant and for the same reasons. If {{Talk header}} is accidentally placed below the wikiprojects (you might be surprised how often) I (and now AWB) move it to the top above the WikiProjects. So for things like this its easier to program something like this: "{{\s*(Talk[ _]+header)\s*([\|}\n])" than this {{\s*(Talk[ _]+header|TalkHeader|Talk[ _]+page|Talk[ _]+page[ _]+header|Talkpage|Talkpage[ _]+header|Talkpageheader|User[ _]+talk[ _]+header|Usertalkheader)\s*([\|}\n])" so I standardize the format first before I perform the other logic. Remember I would have to say {{\s*(Talk[ _]+header|TalkHeader|Talk[ _]+page|Talk[ _]+page[ _]+header|Talkpage|Talkpage[ _]+header|Talkpageheader|User[ _]+talk[ _]+header|Usertalkheader)\s*([\|}\n])" several times because different edits are doing different things. I have 6 different code groups just to move {{Talk header}} so I would have to have that string of talk header redirect combinations at least 6 times if I didn't standardize it once up front before doing the others.
  • So in the case of this one, if one group says they like the spaces and another group says they don't like the spaces and there is no community concensus which is preferred, and it doesn't hurt anything either way, which way do you go? Do you leave them be and be forced to add extra codeing to every combination of edit, over articles you know (at least in this cae I know) you will be routinely editing over and over potentially, or do you get rid of it on the first pass so the next time you have less "clutter" and code needed to make the change. Again I do this more or less up front so that all the follow up code doesn't require the spaces. If now I need to add code to account for the spaces then I am going to need to go back through and make quite a few changes because I do this edit early on in the order to prevent from needing to account for it later a few hundred times. Again its a difference of coding for this ==([ ]*)External links[ ]*==(.*?)==[ ]*References[ ]*==(.*?)== or this ==External links==(.*?)==References==(.*?)== (this is one of the 200+ that I use to reorder sections. I only use it here because it makes a good example). Individually its not a big deal but when you have a couple hundred edits and need to do it over and over again, it slows down the time it takes for the program to process the articles and it makes the coding a lot longer and more complicated. --Kumioko (talk) 22:54, 22 December 2010 (UTC)
  • Again these changes being useless is subjective. The Wiki software doesn't care what format teh breaks are in and will accept them many different ways. The catch here is that the information in WP including these breaks is no just in WP. There are dozens of mirror sites and even face book and some of these mirror sites do not like the malformed breaks. Thats were the problems lie. It could be argued that we shouldn't care about that but, if we can do a simple edit to fix it while we are there anyway so that the data displays correctly on these other (we'll call them customer sites for lack of a better term) than why shouldn't we. Regarding the vandalism comment I had thought about that a while back and randomly check but have never yet found one that was due to obvious vandalism although admitedly it is possible. --Kumioko (talk) 22:54, 22 December 2010 (UTC)

Mostly in reply to Anomie, my interest here is the accuracy issues (as in the two examples brought up on his talk page that I linked out above) which I think just comes down to needing a clear definition of which categories (or other definition) are being tagged and maybe someone more in-the-know about AWB has some tips to avoid double-edits like this followed by this. Nothing that's a big deal except for in the context of the sheer number of edits. VernoWhitney (talk) 19:06, 22 December 2010 (UTC)

In response to your comments VM that particular one stems from me updating the code to put the banner on a single line vice multiple lines. I will update it to not do this unless I am doing a more significant edit at the same time. Aside from uneeded minor edits like this (and even those should be rare) the error rate should be low but admitedly, wether donen manually or by bot it still won't be 100% and by bot will likely be much higher. --Kumioko (talk) 19:50, 22 December 2010 (UTC)
I'm not sure how else to say this: the discussion started because your rate and quantity of manual tagging means you already are acting like a bot under the "Assisted editing guidelines" part of the policy. This has never been about requesting you to program a bot. Period. VernoWhitney (talk) 20:14, 22 December 2010 (UTC)
I understand what your saying I really do but the truth is I try and follow the rules of editing and although I understand the need to have bots and the like I just think its rather funny that here you have an editor who devotes a lot of time editing. Thats it, if I was making 50 edits it wouldn't be a big deal but because I am making in excess of 1000 a day its a problem. The rules of editing don't change between having a BRFA and not having one it just puts more eyes on you and increases scrutiny. Its inevitable that I am going to make a few bad edits and that chance increases exponentially the more edits I do and the more difficult the edit becomes. Having the BRFA request means that, I will have been duly informed of the consequences of my actions so that, when I make a mistake, and that will happen eventually, I get an escelating series of blocks, warnings and the like eventally leading to a situation like the one that Rich is in where I have 2 or 3 editors watching every edit I make and as soon as I make a mistake I get blocked for 48 hours. I apologize if it sounds like I have a bad attitude about this but it really boils down to a lack of AGF. If I make a mistake on an edit all someone has to do is tell me and Ill do whatever I can to fix it. I believe my talk page responses will convey that. Contrary to what seems to be the sentiment doing a lot of edits is a good thing and we should be encouraging users to do so not bogging them down in beaurocracy so they slow down. If someone questions my editing accuracy then go through my edits, if you find some mistakes, especially repetetive ones let me know. --Kumioko (talk) 20:37, 22 December 2010 (UTC)
Actually, a few things do change with a BRFA: it can get you some feedback from people more familiar with doing large numbers of edits; it can get you the bot flag, which helps keep your copious edits from annoying other editors; and it gives the community a little reassurance that someone has looked at the task at some point. Also, "a lot of edits" is not necessarily a good thing, and while WikiProject tagging isn't bad and can help organize some things, it's really not on the list of edits we need more of. Anomie 20:48, 22 December 2010 (UTC)
Based on your last comment I wanted to clarify that the above list is only what I do on talk pages. I have a complately separate list of edits I make to articles but I haven't been doing them for a month or so because I am trying to get all the articles in the scope if US tagged as such so I can get a better handle on the scope of the project. Even ignoring all the articles tagged by the other 150+ US related projects there are still thousands that aren't tagged at all. Those are the ones I am currently addressing as well as supporting some of the US related projects that have either become defunct, inactive or have minimal activity (such as US counties, District of Columbia and Superfunds). I would be happy to provide the list of other edits but I recommend a separate discussion so that we don't get to confusing here. That list is also very long and I separate things into like groups (Talk page edits, Reference/Citation fixes, other). I find doing them all at once gets very confusing when reviewing differences as you mentioned above. --Kumioko (talk) 21:13, 22 December 2010 (UTC)
I also forgot to mention that I there is a certain order I do the edits in so if for example you tell me I cannot do the template redirects anymore (which is the first to process) I will have to completely rewrite my code to compensate for the thousands of redirects (multiple times due to the programming of some of the changes) to all the projects rather than just program for WikiProject Foo (+ the 3 projects mentioned above). --Kumioko (talk) 21:17, 22 December 2010 (UTC)
It would also allow you to speed up. If you separate the tasks that need review from those that don't, you could run the latter under AWB's fully automatic mode. Thst said, which of the tasks actually requires review of each edit? I agree with Anomie, removing spaces in headers does not help anything. You're not really saving any significant amount of space, as it doesn't really add up to much. I think I actually estimated it once, and found that even under an extremely optimistic estimate, removing the spaces saves something like $1-10 in disk space costs over the course of a year without considering compression. However, Wikimedia uses a diff-based, compressed storage system for revisions, so making more changes actually increases the disk space used for that edit. If you think it should be changed in the software, file a bug report and convince the developers.
Your concerns about "escelating series of blocks, warnings and the like" are completely ridiculous. For one, you've already been informed of the bot policy, here and in other places, so that point is moot. Second, if you screw up without a bot flag, you're still screwing up, and can still potentially be blocked for it. Third, if you haven't noticed, Rich is not the only bot operator on Wikipedia. But do you see the things happening to Rich happening to any other operator? Not counting Rich, I can think of only 1 or 2 other bot operators in my entire time on Wikipedia who have run into serious problems with the community to the extent that the operator got blocked. In each case, the problems go beyond simply "making mistakes." Though in some cases, they didn't even make that many mistakes, they just compounded the ones they did make by acting like an ass when it was brought to their attention. If you really do promptly fix every problem brought to your attention, you have nothing to be concerned about. Rich's issue is that he either doesn't fix the problems when he says he does, or he decides they aren't actually problems. He's being blocked now because he's used up people's good faith.
For a good example of how one can massively screw up and still walk away fine, see Wikipedia:Articles for deletion/Anybot's algae articles and Wikipedia talk:Bots/Requests for approval/Archive 4#Request for deflagging and blocking of Anybot. The bot created thousands of articles with factual errors and had a major security flaw that allowed anyone to run it. There was some drama; the articles were deleted and the bot was deflagged/blocked, but the operator was never blocked, and he's currently running other bots without problems. It was an isolated incident, not a pattern of behavior.
This has been explained before and you didn't quite seem to get it, but I'll try again. The reason for the lessening of the distinction between fully automatic bots and semi-automatic tools is because the specific tool used to make an edit is not really important. What's important is the content of the edit, and when the edits are made in large quantities and at high speeds, the communication skills and responsiveness of the editor/operator. Mr.Z-man 22:04, 22 December 2010 (UTC)
Its no problem, if a BRFA is needed then so be it. I really don't think its necessary but if I am told to use it then I will. I just foresee drama thats all. To answer your question most of the changes are ok but where I usually run into hickups is when I rearrange the templates or occassionally when doing the assessment changes. Since most editors have little interest in talk pages they seem to be particularly dishevelled. But there are several of us now (Me, Magio, Xeno and a few others as well as several bots) cleaning these up so gradually we are straightening them out. Just let me know what you want me to do. Since this discussion started I haven't done many edits till this gets sorted out and I would like to get back to it. I've only done about 500 edits today and I'm starting to feel like a slacker.:-) --Kumioko (talk) 22:19, 22 December 2010 (UTC)

Bot request submitted

It seems clear that the sentiment is that I submit this as a bot request so I have done so. The link is: Wikipedia:Bots/Requests for approval/Kumioko. I only submitted the list of edits I do on talk pages for now but once this clears I will submit the list of article space changes separately. If you want me to do this as a bot account rather than my main account I have one setup as Bob. I have already built in the stop button and I think its set to go for the most part. I would suggest that both be allowed. Manual edit from my main account. Automated from Bob (although there are few on the list I would trust with Automated and I would feel compelled to check a large number of them to make sure they were correct). --Kumioko (talk) 14:58, 23 December 2010 (UTC)

How long does it usually take to get approval or disapproval? I would like to start working again ad since I have self imposed not editing until this matter is resolved I would like to get rolling again. --Kumioko (talk) 15:22, 24 December 2010 (UTC)
Minimum a week. We like to get community feedback, you see, but very rarely do. The silence is hence useful as a measure of consensus. - Jarry1250 [Who? Discuss.] 15:39, 24 December 2010 (UTC)
Ugh, ok thanks. I guess thatll give me time to build the project Page for theh Library of Congress project and start updiating the data on Portal:United States. --Kumioko (talk) 15:43, 24 December 2010 (UTC)
I thought about it a little bit and since my stopping from editing was a self imposed thing I am going to resume but at a much slower pace. Please let me know if this is a problem. --Kumioko (talk) 18:32, 25 December 2010 (UTC)
Is there any chance someone could take a look at my bot request please? Is there anything else I need to do or any other concerns that need to be addressed? --Kumioko (talk) 14:49, 24 January 2011 (UTC)

Is there something I am missing here

I'm not trying to sound like a smart ass here but why does it take so long for a bot request to be approved? I have seen some linger for months and others fly in and out seemingly without hesitation and there doesn't appear to be much of a trend. It seems like a week or so for discussion another week or so for a test run and then if the test goes ok it should be approved. I withdrew mine because I simply got tired of waiting and since knowone was screaming for or against it I just decided there must be enough bots and mine wasn't wanted or needed. No hard feelings, its really not a huge issue since I can easily find things to do but at some point I will probably resubmit it and I am trying to learn a little more about how the system works for when I do. --Kumioko (talk) 04:10, 2 February 2011 (UTC)

Reactivating my bots

I've updated two of my bots (User:ImageRemovalBot and User:FairuseBot) to properly deal with renamed images. Since it's been a year since ImageRemovalBot last ran and two years since FairuseBot last ran, can I simply start them up again, or is there some procedure I should follow? --Carnildo (talk) 07:48, 19 January 2011 (UTC)

Asking here is procedure enough, IMO. I see no problem with reactivating User:ImageRemovalBot, although if it doesn't already you might want to consider having it log somewhere which images it removes from which articles in case of undeletion.
Regarding User:FairuseBot, have you been following WP:AN#Trying to defuse a problem with NFCC#10c removals and WT:NFC#Do we need to change how to handle #10c removals? It seems there is no consensus (anymore?) for automated removal of images under 10c where a rationale exists but the article link is "obviously" incorrectly disambiguated, e.g. the FUR links to 2001: A Space Odyssey but the image is actually used on 2001: A Space Odyssey (film). If the bot cannot avoid such cases, it shouldn't be run at this time; if it can, please post how it does so for a quick review. Anomie 18:09, 19 January 2011 (UTC)
The Dab issue should be solved fairly soon, see my thread on an WP:AN#Idea ΔT The only constant 18:11, 19 January 2011 (UTC)
I'm reasonably familiar with the general issue of "obviously incorrect disambiguation", as one of my goals with FairuseBot was to deal properly with them. The bot can't handle the specific case of 2001: A Space Odyssey correctly because the page is neither a redirect nor a disabiguation page. I can add the ability to deal with hatnote disambiguation (it's not hard, but it will cause a massive increase in data transfer), but it won't be 100% perfect: Template:Hatnote templates documentation lists over a hundred different ways of creating a disambiguation hatnote, there are probably unlisted ways, and there's always the possibility of freeform wikitext, which the bot will never handle properly. --Carnildo (talk) 21:22, 19 January 2011 (UTC)

Bot needed to populate DYK's and OTD's

Is there a bot or would it be possible to create one, for On This Day and DYK's to be updated automatically by a bot for Portal:United States (and potentially others)? Currently these 2 sections must be manually updated but I would like to make these 2 section of Portal:United States as user friendly and maintenance free as possible. Any ideas? --Kumioko (talk) 16:43, 19 January 2011 (UTC)

I think you're looking for WP:BOTR. SnottyWong squeal 19:20, 20 January 2011 (UTC)
OOps, I think your right thanks. --Kumioko (talk) 19:33, 20 January 2011 (UTC)

XLinkBot

The following discussion is closed. Please do not modify it. Subsequent comments should be made in a new section.
Related discussion: Wikipedia talk:External links#Blanket ban on Youtube links?

I propose the revocation of User:XLinkBot's editing rights. Although thew bot might do some fine work, the operator has refused to adjust the parameters it uses when dealing with YouTube videos. The recent request was made at User talk:XLinkBot#Youtube videos.

There is no blanket ban on linking to YouTube videos. Automatically removing these links has proved problematic. For bots to be approved, the operator must demonstrate several things. Of these, the following are problematic:

  • is harmless (the bot creates additional work for others when it removes links inappropriately)
  • performs only tasks for which there is consensus (there is no consensus to remove all YouTube links so the bot is editing against consensus)
  • carefully adheres to relevant policies and guidelines (please see the above concern. We have [[WP:Youtube which states "There is no blanket ban on linking to YouTube or other user-submitted video sites, as long as the links abide by the guidelines...")
  • uses informative messages, appropriately worded, in any edit summaries or messages left for users (It does not. In fact, this bot caused a misunderstanding of our policy when I was a new user since I assumed YouTube videos were not allowed under any circumstance. Certainly a disclaimer would be needed)

All the operator needs to do in my opinion to stay running is not remove links to YouTube. I believe we should err on the side of caution if the bot is editing against consensus since the policy is clear. It is not the bots job to start discussion "just in case" but to preform menial tasks that are of obvious benefit.Cptnono (talk) 23:26, 17 February 2011 (UTC)

  • Strong Oppose. That's like throwing the baby out with the bathwater. You don't have consensus going your way in the EL discussion so you decide to...forum shop over here? That's what it looks like. The merits of XLinkbot so heavily outweigh anything to do with Youtube that this request has a snowball's chance in hell of succeeding. You do not have consensus in your favor...this looks like a bad attempt at strong-arm tactics.
    ⋙–Berean–Hunter—► ((⊕)) 00:52, 18 February 2011 (UTC)
Would appreciate it if you did not assume the worse or make unfounded allegations. The EL discussion does not have the authority to shut off a bot and isn't about the bot. It is about if there is a blanket ban on YouTube so it is going perfectly well as far as I can tell. But yes, it is like throwing the baby out with the bathwater. The operator could have stopped this action if he would have addressed the requests made to stop editing against consensus. So it isn't my fault, is it? Cptnono (talk) 03:14, 18 February 2011 (UTC)
Has the bot reverted any of your additions of YT links? ...or you are seeing new users and IPs being reverted by the bot? I'm under the impression that the bot is only reverting those.
⋙–Berean–Hunter—► ((⊕)) 03:23, 18 February 2011 (UTC)
  • Support with comments. I think the current method of automatically removing Youtube videos on sight is a bad one since some videos are added to Youtube, for use in Wikipedia, because they exceed the 100MB limit of the Wikimedia servers. If a way can be devised to allow certain videos then the bot can be allowed to resume. --Kumioko (talk) 01:52, 18 February 2011 (UTC)
I'm disagreeing with shutting off the bot over a singular issue. The correct course of action for someone who doesn't think that YouTube should be automatically reverted should have been to try to gain consensus to have the singular entry for YouTube removed from XLinkbot's revert list...not try to shut the bot off. I didn't see such any such consensus formed after the discussion at the EL noticeboard. There should have been more effort there. This effort looks pointy. As to the size limitations, I agree with you and would favor allowing established users to upload >100MB files (such an allowance might mean pre-approval of the upload so as not to waste resources).
⋙–Berean–Hunter—► ((⊕)) 02:53, 18 February 2011 (UTC)
Perhaps but if the bot operator is unwilling to discuss it in light of the 3 issues above then there is a problem. Personally I think there are few occassions were Youtube should be used. I do think that it should be allowed though. I didn't know that there was an exception list for the bot and I bet 99% of other editors don't know that either and to me thats part of the problem here. The issue of using youtube for larger files isn't just one of access but is a software limitation requiring the programmers to do some work on their end from what I understand. Regarding the consensus that doesn't surprise me. Consensus frequently gets formed and then once the action starts and people notice the consensus is revised. Additionally with more and more government agencies using Youtube I can see the exception list getting very large and unwieldy pretty fast. --Kumioko (talk) 03:07, 18 February 2011 (UTC)
(ec)Or the bot can not remove appropriate links against consensus. Since the operator has not coded it to do so it should not be allowed to function. Why is there an assumption that the bot should be used as a safety valve instead of as a harmless method of doing mundane tasks?Cptnono (talk) 03:14, 18 February 2011 (UTC)
  • Strong oppose There seems to be some confusion. Kumioko, can you tell me just how many FedFlix folks you have who are (1) adding good links from YouTube and (2) haven't ever made ten edits? XLinkBot doesn't 'see' edits made by users who have made more than ten edits (ever), so it never reverts these. If you wanted, you could spend all day doing nothing but adding YouTube links.
    As for the specific four issues:
    1. There are occasional false positives, but when people have actually bothered to go through the bot's history, they usually come back saying that it's 90–95% correct. (Cptnono might want to go through that exercise. I suspect it would be enlightening.) So it saves the community time, by removing so many bad links. It certainly doesn't cost us more to revert XLinkBot's occasional false positive than it costs us to revert the anons' frequent inappropriate additions.
    2. Certainly there's no consensus to delete every single YouTube link—if we wanted that, we'd WP:BLACKLIST the domain—but that's irrelevant, because XLinkBot is not doing that. XLinkBot removes only links added by IPs and brand-new accounts, which have a very high rate of misusing these links, like this.
    3. XLinkBot does carefully adhere to the guidelines. There is no guideline that requires us to accept YouTube links. There's one guideline that says these links aren't absolutely prohibited (it weakly discourages them), but there's a big gap between "not actually prohibited" and "must be kept".
    4. I think that XLinkBot's messages are adequately informative. It provides suitable links for anyone who wants to read the details. I am concerned that Cptnono might not have read the user notice XLinkBot left, since it explicitly says that the bot might have been a mistake and that s/he could revert the removal. WhatamIdoing (talk) 05:32, 18 February 2011 (UTC)
  • Oppose I have monitored anti-spam pages for a long period and have seen many complaints about XLinkBot. I have investigated some of those complaints from time to time and have never found a "wrong" action by XLinkBot. Of course there will be some actions by XLinkBot that are not correct, but I have not seen one. I cannot see (in this discussion or at WT:External links#Blanket ban on Youtube links?) that any examples of bad edits by XLinkBot have been provided. The amount of damage caused by people adding inappropriate links should not be underestimated, and XLinkBot saves editors a lot of work. The advantage of using the bot is that it quickly reverts certain links added by brand-new users, and it puts a very helpful and non-bitey comment on the user's talk, and it specifically advises that the link can be re-added if wanted which will not be reverted by the bot again. If XLinkBot were not performing its work, a new user may very well add 100 or more links before someone notices and takes action. Assuming the links were inappropriate, it becomes difficult for someone like myself to respond once, say, 10 links have been successfully added by a new user because the early success provides an entitlement feeling. If link additions are reverted quickly, the user works out that pursuing their aim of adding links will not be productive. Also, many new users can add links more quickly than another editor can remove them (because you are supposed to individually check each link in each article). Johnuniq (talk) 01:27, 19 February 2011 (UTC)
  • Question as far as I know, if you re-revert XLinkBot, it won't edit war with you, right? If that's still the case, then I don't see a big deal here. If XLinkBot makes a mistake and the youtube link was indeed not a copyright violation and was an appropriate link, then just revert what it did. Gigs (talk) 02:26, 19 February 2011 (UTC)
Correct. This proposal was opened by someone who last participated in the discussion at Wikipedia talk:External links#Blanket ban on Youtube links? on Feb. 12 and he garnered no support for the idea of shutting the bot down. The thread died down on the 13th and without any further input from him there, he switched to this venue several days later with no consensus to back him up. He is attempting to state that because the community has a consensus which allows YouTube to be used some of the time, that consensus is on his side despite the fact that things were clearly explained that XLinkbot is only reverting new editors and IPs and all at the discussion agreed. He never protested but instead headed here. I note that he hasn't answered the question I placed to him above which is, Has the bot reverted any of his additions of YouTube?. His previous statement above, "The operator could have stopped this action if he would have addressed the requests made to stop editing against consensus." is unfounded. The only one making the current request is Cptnono. The only other editor who had complaints concerning XLinkbot never suggested nor backed its shutdown and after the nature of the bot was explained appears to be satisfied well enough. See User talk:XLinkBot#Youtube videos for the preceding thread prior to the one at the External links talk page.
⋙–Berean–Hunter—► ((⊕)) 03:41, 19 February 2011 (UTC)
FFS, BereanHunter. You don't need to paint me as some sort of dick since I believe the the bot should be shut down. So do you have any reason to believe that the bot should continue to operate or is your argument that I am a jerk? I actually stopped following the discussion at EL since it wasn't the correct venue to seek its removal. It had nothing to do with not getting nay support. Note that another editor here has supported my proposition so saying it is just me was untrue and does nothing but change the discussion from the bot making edits against consensus to me and whatever reasons I might have. And have I ever been reverted by the bot? No. I have seen it in edit summaries which led me to believe that YouTube was not allowed at all which is why I mentioned that it adds confusion. But if consensus is that it only reverts new users and IPs then we should consider changing the "anyone can edit" mentality of Wikipedia if we can finally admit that we do not respect the edits made by newcomers.Cptnono (talk) 20:22, 21 February 2011 (UTC)
If you want to continue the discussion, please respond to the points raised above. In general, there is no perfect solution to any problem, and some pragmatism is required. Johnuniq (talk) 22:02, 21 February 2011 (UTC)
I already did: And have I ever been reverted by the bot? No. Was there another question I missed? and But if consensus is that it only reverts new users and IPs then we should consider changing the "anyone can edit" mentality of Wikipedia if we can finally admit that we do not respect the edits made by newcomers. I think it is good that it only looks at newcomers but I think that the principle of that is not inline with our stated intentions. I do not think the bot would be approved if its sole purpose was to remove YouTube links from newcomers but that is just an assumption. And why is it acceptable for me to bring up reasoning based on precedent while Berean Hunter only brings up reasoning based on assumed motivation? So we do not need to "continue the discussion" but it was not me who derailed it. Cptnono (talk) 22:15, 21 February 2011 (UTC)

Here's my position as a BAG member: There is no reason to stop the bot at this time, as there is no evidence that the bot's heuristics are having excessive false positives, the messages left for newbies by the bot are extremely civil, and it is exceedingly easy for anyone to override the bot in any individual case. The hypothetical concerns that a newbie will choose one of the tiny fraction of usable YouTube links to add (and then be offended by the bot's polite message) is not convincing enough to outweigh the vast majority of the time that the bot's edit will be entirely appropriate. That said, however, I do encourage the operator to look into the possibility of whitelisting those YouTube channels that are run by reliable sources.

I'll leave this discussion open for the moment in case any other BAG members want to comment. Anomie 00:27, 22 February 2011 (UTC)

I don't think that whitelisting channels is possible. The typical link only seems to give you watch?v=1234567890, and occasionally &feature=channel—but not something that tells you which channel each video is from. WhatamIdoing (talk) 01:17, 22 February 2011 (UTC)
Comments from BAG is great. I am pretty sure there is no longer any chance that the bot will be turned off even though I disagree with part of its function.. Ugh... this is why I don;t like the actions of this bot. The misconception that there is a blanket ban is so prevalent. That of course is not the bot's fault. I do admit that the fact that it is often a useful bot. I also appreciate a comment from a BAG member. However, I would like some comments on if it is appropriate to assume that additions of links from newcomers is bad. Although stats might show it to be true, the idea seems against the project's principle's. If the bot is allowed to continue to make such edits, I will be sure to add it to the essay WP:VIDEOLINK that the community has come to a consensus that the bot does more good than harm. Cptnono (talk) 01:34, 22 February 2011 (UTC)
I for one hate the misconception that there is a blanket ban on "primary" sources, or that there is some sort of real difference in the way we can use a "primary" versus a "secondary" source besides meeting the requirements of WP:N, or that WP:PSTS actually says anything true enough to be in policy instead of in an essay, (if you want details on any of that, ask at my talk page) but there isn't any more that I can do about it than you can do about the YouTube misconception. At the moment, it seems to be well demonstrated that YouTube links added by IPs and the newest of users are all too likely to be unsuitable if not outright linkspam, and in case of false positives the bot is extremely careful not to bite the newbies and refuses to edit war about it. Anomie 01:53, 22 February 2011 (UTC)
@WhatamIdoing: But is it possible to go from v=1234567890 to the channel via YouTube's API, or some other reasonable method? Anomie 01:53, 22 February 2011 (UTC)

@ALL, but especially Cptnono. Anyone can edit, sure. But still we block, protect, abusefilter for edits and spam-blacklist for links, and we use a handful of antivandalism bots (of which I count XLinkBot to be one) to undo the edits which are giving a high number of cases where the edits do more harm than that they help. Anyone can edit, but that does not go for the main page. Whole IP ranges are blocked, pages get long page protections or abusefilters shut down certain types of edits. All of these have an immense impact, they block innocent editors, they disable ANY form of editing to specific pages or specific editors. And yes, practically ALL of these methods have collateral damage, most of which we will never know since the editor is blocked from doing the specific edit, but which certainly harms Wikipedia.

The antivandalism bots have complex heuristics to find vandalism, and revert that on sight. However, also there there are mistakes. Those bots sometimes revert edits which are fine and OK. As for the blocks, protects, abusefilters and spam blacklist, there are cases where the bot is mistaken. Try (as an IP) to cite literally an explicit piece of text on the page of a pornography writer making sure there are a significant number of 'bad' words in the text .. I think the chances are quite big that you will be reverted and will receive a warning.

XLinkBot works similarly, there are links which have a huge hitrate on being 'bad' - most of the links XLinkBot is reverting are plain spam, but there are also many of the others which are, really, hardly ever helpful and have special mention in WP:EL. MySpace is an example - I did a statistics on that link on 30 reverts, and I found 1 which I would not personally have reverted, though I do think it was not particularly helpful even on that one. Another couple were in a 'maybe' mode, but many were not related to the page, many were dead, many were totally superfluous. And for MySpace the problem of copyvio is pretyy minimal. We do not have WP:EL for nothing, and it is based on our core policies (e.g. WP:NOT, WP:COPYRIGHT). Some think that it is good to have as many links as possible about a subject .. but our policies and guidelines, which have been established by many editors, oppose that view. That a link is on topic and links to non-copyvio material is not a reason to link to it (it is also a reason not not to link to it...)

YouTube is another case. Yes, there is good info on YouTube, yes, I could, can, and will whitelist links (and every admin can adapt the youtube rule to exclude certain rules, and anyone can request that on the talkpage of the bot, on a talkpage of one of the operators, or come to IRC) which are recognisable as good links, but unfortunately the majority of them is not as such recognisable. But YouTube has a bar to pass (just like any other link that one adds; and note that XLinkBot tries to detect whether external links are used as references and not revert those). YouTube still has the problem that many useful video's are copyvios (no, I am not saying that most of the videos are copyvios, I am not saying that most of the good videos are copyvios), and that is a pretty grave problem. But there is more, YouTube requires a significant bandwidth, it requires installation of software (which not everyone has, and which is not avaible to anyone), etc. etc. WP:ELNO has a number of points which are 'ticked' by YouTube. And do note, some of the YouTube video's are plain spam, they are added by the 'owners' of the channels to get the traffic with which they make money.

This however does still not totally disallow linking to YouTube. If a YouTube video significantly adds to a page then sure, why not. And we are linking to a large number of YouTube videos. XLinkBot does not disallow editors, it, at first, friendly notifies (not warn!!) them of the common problems with YouTube (or other links), and asks them to reconsider. It will not edit war. Only when editors are persisting in adding numerous youtube video's in a short period of time to different pages XLinkBot may issue warnings. XLinkBot does not have a bot-bit, so that recent changes patrollers will see the XLinkBot reverts and may re-revert, and even when XLinkBot gets to a warning level that editors get reported to AIV, it still needs a human editor to review if the editor should be .. sanctioned. XLinkBot hardly EVER gets to that level, and I don't think that it ever has on YouTube.

However, and that is why XLinkBot exists, if links on a certain domain have a high hitrate of being unsuitable, then that is a reason to revert additions them and notify the adding editor of the problems. Many new editors have no clue that videos on YouTube are sometimes copyright violations, or are not suitable as an external link. Yes, there are many YouTube videos that are good information and are suitable as an external link, but really, the large majority of movies IS NOT. We do not need videos of birthday parties, heavy rain in a small village in China, someone playing with his volleybal, name it (and yes, those are not copyvios .. !). I did a very, very quick check of about 10 reverts, and I did find 2 copyvios being reverted. Others were indirect or just too general to be of significant use. OK, I did not look significantly into them, but if I find 2 copyvios in 10 reverts, then that is certainly .. cause for concern.

Now, Cptnono. This bot was designed to revert additions of external links by new editors and IPs which were very often (most of the time) problematic (if they were always problematic, it would be blacklisted). If there are particularly grave problems with a certain external link (i.e., a significant number are plain copyvios) then we, IMHO, should accept a bit higher error rate, others which are less grave should have a low error rate. XLinkBot reverts on specific rules per link. One could ask that if one finds a specific rule which has a high error rate, that that specific rule gets removed. If YouTube is really having so much trouble, then that rule could be removed. But it is by no means a reason to shut the whole bot down. Now, pending any statistics on a significant number of reverts for YouTube, I don't think that this calls (yet) for removal of the YouYube rules, let alone for withdrawing the editing rights completely.

Cptnono, I find 'the operator has refused to adjust the parameters it uses when dealing with YouTube videos' an incorrect description of what is being said and done (to say the least; you did not ask for adjusted parameters, you just demanded a shutdown without showing errors), and 'Automatically removing these links has proved problematic' is yet to be proven (that there are good YouTube links does not mean that the majority that is added is actually not). --Dirk Beetstra T C 11:42, 22 February 2011 (UTC)

Wow, back to me again instead of the issue, huh? "This bot should not remove YouTube videos, as it did here. If anything, it should set something like a flag—maybe using {{Youtube}}—that a human editor may check if the video violates anyones copyrights. Because, for instance, in that incident above, it did not." Is the first thing at the discussion from another editor. A request to adjust it was made and it was refused.
So no, not anyone can edit. New users are deemed problematic so their edits are reverted. Just admit it. Unfortunately, we have an IP on the bot's talk page who said: "Just another annoyance with the youtube link reversion: I changed an existing link to a removed youtube video into a working video, also on youtube. xlinkbot reverted my edit, so we've still got a youtube link, but it just doesn't work. This does not feel like a win"
So according to the standard, this bot violates it. However, if community consensus is that violating the written standard is worth it then so be it. But don;t pretend it is editing within the alloted rules.Cptnono (talk) 08:40, 3 March 2011 (UTC)
This is not the place for continued complaints regarding how the community treats YouTube links. Take it to WT:EL or WP:VPP. Anomie 12:06, 3 March 2011 (UTC)

The above discussion is closed. Please do not modify it. Subsequent comments should be made in a new section.

Bot flag request

I (and others) would be grateful if someone could flag User:Taxobot 7 as a bot account. See User_talk:Taxobot#Taxobot_7. Thanks, Martin (Smith609 – Talk) 18:26, 28 February 2011 (UTC)

Someone please flag this bot!! It's clogging the Special:NewPages log for newly-created templates... — Preceding signed comment added by Cymru.lass (talkcontribs) 08:16, 13 March 2011 (UTC)
You might have better luck asking at WP:BN, as the bureaucrats are the only ones who can actually give the flag. Anomie 13:49, 13 March 2011 (UTC)

Problem filing request for additional task, default links generated incorrect?

If I type "Wikipedia:Bots/Requests for approval/RjwilmsiBot 8" (as my next task would be) into the box at the top of Wikipedia:Bots/Requests for approval the logic to derive bot name etc. on the created page always seems to generate a red link for me, that then has to be corrected. Am I missing a trick? Thanks Rjwilmsi 17:59, 11 May 2011 (UTC)

No trick, the page generation is just wrong. It's too bad we didn't made subsequent requests use a slash instead, e.g. "Wikipedia:Bots/Requests for approval/RjwilmsiBot/8", as that could easily be made to work correctly. With the way things are, someone would have to make a hack using evil string manipulation to detect and remove the suffix. Anomie 10:50, 12 May 2011 (UTC)

Possible solution for the attribution problem with category renames?

Feel free to join the discussion at Wikipedia talk:Categories for discussion#Possible solution for the attribution problem with category renames?, where I brought up a possible solution to the problem. עוד מישהו Od Mishehu 10:08, 15 May 2011 (UTC)

Restrict Citation bot to article space

I believe during the discussion leading to the approval of the latest version of Citation bot that there was an implicit assumption the edits would be done in article space. Other spaces, such as Wikipedia space and User space, may have project pages that contain deliberate citation errors as examples of what NOT to do, or pages that compare and contrast various citation styles, or draft articles where unexpected changes would not be appreciated. But, in fact, the bot is editing Wikipedia space. I request the approval be revisited and limited to article space. Jc3s5h (talk) 14:12, 20 May 2011 (UTC)

I would have thought a bot editing templates designed for mainspace would only edit the mainspace. So I do agree it shouldn't edit other namespaces without prior discussion due to inevitable false positives. —  HELLKNOWZ  ▎TALK 16:02, 20 May 2011 (UTC)

That's a senseless/overkill restriction. There is perhaps a case for a restriction in the User/Wikipedia namespaces (aka, only run in these namespace on request rather than systematically crawl these pages), but there's no reason why the bot should be restricted from running in the Template/Project/Portal/Help/etc... namespaces (especially in the template namespace), or disallowed to run in the User/Wikipedia namespaces entirely. Headbomb {talk / contribs / physics / books} 18:00, 20 May 2011 (UTC)
I might agree on automatic runs on some namespaces, but not User: for one. Drafts are drafts for a reason, editors may not wish for bots to start treating them as articles. In general, there are no policies/guidelines concerning the use of references in most namespaces, so there is no reliable way of knowing if someone did something on purpose (giving examples of correct/incorrect use, for instance) I just assumed bots are approved for mainspace unless the operator indicates otherwise; after all we don't have bots dating maintenance citations in Wikipedia:. I do realize there wouldn't be many false positives and I am OK with these edits if they are supervised, but not autonomous. —  HELLKNOWZ  ▎TALK 18:23, 20 May 2011 (UTC)
P.S. Per clarification, I was referring to automated edits. I am perfectly fine with user initiated per-page edits. After all it is then the user responsibility to verify that what they are doing is fine. —  HELLKNOWZ  ▎TALK 18:33, 20 May 2011 (UTC)
I agree that supervised edits should be OK; the bot owner will want to test it on his/her own user space, and other users may want to invoke it to test draft articles. But I am not aware of any space other than article space where citations are required, so who cares if a citation is less-than-ideal in, say, a template, when the citation isn't required at all? Jc3s5h (talk) 18:41, 20 May 2011 (UTC)
Because several templates do make use of citations, like {{Infobox hydrogen}} or those in Category:Science citation templates. Headbomb {talk / contribs / physics / books} 18:44, 20 May 2011 (UTC)
As far as I can tell, {{Infobox hydrogen}} does not include a citation template in it's source code. In general, I would be very concerned about changing a citation in Template space for fear that there could be two versions of the citation, one that is generated by the source code on-the-fly, and a different one in the template documentation. The bot could create contradictions between what actually appears in the articles where the template is transcluded, vs. the template documentation. Jc3s5h (talk) 19:08, 20 May 2011 (UTC)
{{Infobox hydrogen}} has two {{cite book}} and one "manual" citation (which I'll fix in a second). As for "template documentation", I can't even fathom how it could be an issue. Headbomb {talk / contribs / physics / books} 19:21, 20 May 2011 (UTC)
If you look at {{Cite pmc}} you see that it consists of a bunch of logic that builds a template, and it is unlikely a bot would recognize it as a citation. It also has a separate documentation page, Template:Cite pmc/doc. In this particular case, the documentation does not contain any example of what a finished citation would look like, but if it did, a change in the documentation page by the bot would make it contradict the citations that would be built by the template. Jc3s5h (talk) 20:29, 20 May 2011 (UTC)
And that affects, what? 20 templates documentations? {{Bots|deny=Citation bot}} is perfectly appropriate for this. Headbomb {talk / contribs / physics / books} 04:23, 21 May 2011 (UTC)
I'm not sure if this request for restriction is really necessary. The request has arisen from only one incorrect edit in the last week as far as I can see. In any case the citation bot must continue to be permitted for the template namespace to support the cite pmid, cite doi templates etc. Rjwilmsi 00:58, 21 May 2011 (UTC)

Very small change to SDPatrolBot

I want a quick approval from another BAG member for this, but it's not really worth a full BRfA. I've talked on IRC about this. So just requesting that SDPatrolBot can report troublesome users to AIV, instead of just edit warring with them over the tag. - Kingpin13 (talk) 11:06, 30 May 2011 (UTC)

 Approved. Clarified details. Multiple (4+) tag removal by author results in AIV report instead of spamming warning user talk templates. Trusted operator, uncontroversial. —  HELLKNOWZ  ▎TALK 11:10, 30 May 2011 (UTC)