Wikipedia talk:Bot policy/Archive 24

With the introduction of AbuseFilter, pretty much all of the things the current batch of AntiVandal bots can be implemented as AF rules. This also provides better fine grained control of actions/responses. I propose we put a moratorium on considering any more Anti-Vandal bots. Q ^{T C} 23:43, 17 March 2009 (UTC)

I haven't seen any serious anti-vandal bot requests lately anyway; have I missed one? I have seen (and denied) several that seem to be wanting a vandal bot, though. Anomie⚔ 03:02, 18 March 2009 (UTC)

Anything important enough to "coast through" BRFA is IMO important enough to apply WP:IAR to the entire approval process, as was done recently to add {{copyvio}} tags to several thousand stubs when it was discovered the creator copied the text from a book. I did offer to approve the bot to do the same thing for future incidents of this type so IAR wouldn't be needed, but I've received no response on that so far. Anomie⚔ 19:39, 22 March 2009 (UTC)

The GRU page needs updated. It says we allow per the Meta policy, but the Meta policy has expanded to include double redirects as well as interwiki links. Do we want to readdress our stance with the Global Policy, or should we update the Global rights page to indicate that it's allowed only for interwiki links as that is all that was agreed upon? Q ^{T C} 03:06, 30 March 2009 (UTC)

I say update the global rights page, especially since T19888 will probably result in double-redirect-fixing being unwanted. Anomie⚔ 03:13, 30 March 2009 (UTC)

Interwiki bot approval practice does not seem to follow the current policy, for good reason. I propose several clarifications to the bot policy and one extra requirement. First, understanding of target languages is not required to add reciprocal links. Second, the operator must be able to prevent an incorrect reciprocal link from being added again. This would result in fewer cases like the one described in this thread, where an article maintainer is annoyed by multiple bots. Third, I copied the policy on global bots into this section, where it is most relevant. I also removed some redundant phrases from the existing language. This is the proposed policy:

Operators of bots which create non-reciprocal interwiki links must be familiar with the languages to which they are linking. Bot operators running standard tools such as the pywikipedia framework should update to the latest version daily. If a bot adds an incorrect reciprocal link, its operator must be able to prevent other bots from restoring the link. Globally-approved interwiki bots are permitted to operate on English Wikipedia, subject to local requirements.

This paragraph would replace the two currently in the "Interwiki links" section of Wikipedia:Bot policy#Restrictions_on_specific_tasks. I don't know if any non-reciprocal interwiki bots exist or will ever exist, so that sentence may not be necessary. I also recognize that the extra requirement on bot operators to disable certain interwiki links is a controversial change that affects many current operators and could also be removed. Wronkiew (talk) 00:53, 1 April 2009 (UTC)

I support this wholeheartedly, though I'd like to suggest a couple of slight tweaks, if I may: firstly, I wouldn't use the word "reciprocal" - I know what you're on about, you know what you're on about, but we have to remember that English may not be the first language of interwiki operators and we should make it as obvious as possible. "Original" perhaps? Also, I'm a bit confused regarding the phrase "to prevent other bots from restoring the link": isn't that other owners' jobs? Obviously, the same bot should never re-edit the page and/or edit war. Anyhow, yes, I support this sensible change. - Jarry1250 ^{(t, c)} 19:00, 1 April 2009 (UTC)

Thank you for your comments. I think reciprocal and non-reciprocal are the most specific terms that describe the type of links these bots add. Would an initial sentence defining these terms address your concerns? My intent for addressing the incorrect link/edit war problem was to make the bot owners responsible for correcting the bot behavior rather than the page maintainer. I don't think pywikipedia checks page history, so operators have no way to keep their bots from re-adding incorrect links. However, if they do add an incorrect link, the operator can ensure that the link is not re-added by other bots. I have heard about a link blacklist, but I don't know if it is operational. Adding and commenting out the link may also help. Another solution might be to delete the link at the remote wiki. None of these can be done proactively, but operators should know what to do if they are notified of an error. Wronkiew (talk) 06:28, 2 April 2009 (UTC)

Here's a possible initial definition: "Interwiki bots maintain links among articles in different languages. Most commonly, a bot will search for pages to which other wikis link and ensure that the page contains links back to them; this is called reciprocal linking. If the page is not already linked from other languages, added links are called non-reciprocal." Wronkiew (talk) 06:39, 2 April 2009 (UTC)

Call me crazy but that defeats the whole purpose of interwiki bots if they can't add the links that are missing... It's not all that hard for anyone to make sure the edit war on a given page doesn't keep hapening, you go and remove the IW links from all the linked pages then a bot won't continue to readd them once removed from all pages. -Djsasso (talk) 14:31, 3 April 2009 (UTC)

Exactly. My intent here was simply to make the operator responsible for cleaning up incorrect links that their bot added, rather than the page maintainer. I had not intended to place any restrictions on interwiki bot operations, and I apologize if my wording gave you that impression. Are you concerned only with the sentence "if a bot adds an incorrect reciprocal link, its operator must be able to prevent other bots from restoring the link"? Do you have any suggestions for improvements to more clearly state the intent? If it's unclear, and cannot be stated simply, I'd be fine with dropping that one sentence and moving on. Did you have any concerns with the rest of the proposed policy? Wronkiew (talk) 21:52, 4 April 2009 (UTC)

I found at least one interwiki bot that adds non-reciprocal links, BotKung. I've also seen a few like it at bot requests. If no one objects, I'll merge in the proposed policy, minus the contested anti-bot-warring requirement, in the next couple of days. Wronkiew (talk) 07:43, 7 April 2009 (UTC)

This is done. Wronkiew (talk) 21:59, 11 April 2009 (UTC)

There are certain kinds of tool that aren't clearly defined under the present policy wording. At present, bots are basically automated tools, and tools whose edits require some element of interaction come under "assisted editing".

There are some tools that are presumably assisted editing but not clearly defined as such, where the user doesn't "interact" for each edit, because the task is very well defined for each run, requires no element of machine judgement, and all interactional data is specified in advance of the run.

What these have in common is the use of an automated tool, to perform at high speed a repetitious and non-controversial "housekeeping" task on multiple "targets". They don't really come under "fully automated bots" since there is user interaction - it's just front loaded at the point the run is executed, by the user individually checking each action will be okay.

I'd like to see under assisted editing, something to the effect: "It also covers tools that perform well-defined non-controversial actions requiring no further judgement, on a previously checked list of targets."

The fact the user has checked the proposed targets individually beforehand, is the key. Can the wording be made more watertight to prevent its abuse for mass actions though?

Hopefully this aim is sensible and not controversial. FT2 ^{(Talk | email)} 02:50, 20 April 2009 (UTC)

My understanding of the current policy is that the user has to approve each edit before the script commits it. Tasks that process a large number of simple edits at high speed, even if they are working off a user-generated list, must go through BRFA. I think the distinction is very clear in the existing language. Also, I reverted your change to the assisted editing definition for now. Wronkiew (talk) 04:29, 20 April 2009 (UTC)

It seems to me that there are several reasons for the current approvals process in this case:

1. To some people, "checked" means "I generated this list from known criteria", which fails if corner cases were not considered. For example, I've recently had requests to tag every page in Category:Energy and subcats and every page in Category:NATO and subcats with the corresponding WikiProject banners; in both cases, it was not hard to find a page a few levels down in the category tree that had nothing to do with the project.
2. High-speed edits can mean high-speed mistakes. While we can't catch every mistake someone's bot might make, the approvals process does help detect them and the requirement to manually approve each script-assisted edit (hopefully) handles those scripts.
3. Users often do not consider how their idea may be controversial, and (when we do our job right) BAG will point this out during the approvals process. Restricting unapproved actions to "human" speed gives reviewing humans a chance to object before too big of a mess is made.

Is there something in particular this addition is aimed at? Anomie⚔ 12:07, 20 April 2009 (UTC)

It's aimed at cases where a fixed well defined action is required on many pages. This arises quite regularly in checkuser or oversight work, where a vandal may post many dozens of "outing" data, all of which need the same treatment. Other examples are as a developer of the SPI process, having to change the format of the {{RFCU}} "request for checkuser" template on all SPI subpages that used that template, when it was modified to allow a second "case letter" to be added. A further case was a process update that would be of relevance to a number of users (~30 from memory), all of whom needed the same note added to their talk page so they were aware. In each case the matter may be classified as "routine", and the exact actions to be taken are listed and individually reviewed before any are processed, up front. These tend to be non-controversial types of actions that just need doing to a list of pages. What's being asked is to add that if each individual edit is checked (ie no "assumptions") and the matter is non-controversial, they may use an "assisted tool" (at "human" and not "bot" speed) to do the actual posting for them once they have reviewed all the proposed items. I think this fits into the criteria and spirit of "assisted tools", but it would be good to add it explicitly.

Quick comments on the above - 1/ automated generation of "targets" isn't contemplated. Individual checking of each item is intended, before they are added. The tool acts purely as a saving on the repetitious activity, not a substitute for checking each case. 2/ See #1, 3/ Agree, this would appear to be an argument for speed restriction (as with all assisted editing) though, which is already in the policy. FT2 ^{(Talk | email)} 14:07, 20 April 2009 (UTC)

The only example you give where having to get bot approval would be troublesome is your oversighting example, and even there I think it likely that a bot tasked to generically "revert (if necessary) and oversight a specific list of revisions determined by an authorized oversighter" has a fair chance of being approved and would be more likely to correctly handle any edge cases. As for the rest, we already have bots approved to manipulate templates without having to worry about any random user's poor regex skills, and we already have many bots approved to deliver notices to talk pages. And I doubt doing either with AWB in manual mode (or the equivalent using any other script falling under the current assisted-editing criteria) would be excessively onerous, if it came down to that.

In general, well-defined non-controversial tasks by proven bot ops are precisely those that can easily sail through BRFA with little delay anyway, while ill-defined tasks and those by unproven bot operators are precisely those that should have more attention paid. Anomie⚔ 16:43, 20 April 2009 (UTC)

I used to use a tool I wrote called TINA (admins can view the page User:Ameliorate!/TINA). It worked by displaying a list of each changes it was going to make, as an example:

1. Somepage: {{sometemplate|someparam=foo}} --> {{sometemplate|someparam=bar}}
   
2. Anotherpage: {{sometemplate|someparam=foo}} --> {{sometemplate|someparam=bar}}
   
3. Someotherpage: {{sometemplate|someparam=foo}} --> {{sometemplate|someparam=bar}}
   

So in a list of 1000+ edits it was easy to see any corner-cases and exclude them. As far as I was concerned this was acceptable under "... but do not alter Wikipedia's content without some human interaction" as there was human interaction, just it was done en-mass at the beginning. I never received any complaints about it, so I would support any change to explicitly allow this. ~ Ameliorate! 04:52, 20 April 2009 (UTC)

Back when you brought it up, I thought it displayed a bit more of a diff than that... Anomie⚔ 12:07, 20 April 2009 (UTC)

I think that this should be in the bot tasks category. Even though the script is presenting all the diffs for confirmation before it makes any edits, no allowance is made for feedback by other editors. If someone disagreed with the change and it was running unattended, the only way they could stop it would be to have your account blocked. Wronkiew (talk) 17:11, 20 April 2009 (UTC)

It stopped if anyone edited my talkpage. ~ Ameliorate! 23:24, 20 April 2009 (UTC)

That makes you a very responsible bot developer. I see your point, but I would consider the process you described to be unattended editing rather than assisted editing. Wronkiew (talk) 06:03, 21 April 2009 (UTC)

This is exactly the issue when trying to clarify the difference between a bot and a script. How/why is it different to manually approve 1000 changes in advance than to approve them all in real time as the script runs? Is there a difference between making a list of pages in userspace and using Twinkle BatchDelete on them versus making a list in a file and using a Python script to delete them? Mr.Z-man 06:36, 21 April 2009 (UTC)

Yes, there is a difference. If you review a list of articles and then start the edits, the delay between the review and the edit may become long and intermediate edits become likely, which may change the edit from the form initially reviewed. When the edit is generated "live" and reviewed at the moment before the edit is made, the delay should be small. Deletions in particular can cause difficulties if done in large batches - I recall one case where an image was up for deletion for some deficiency in source or copyright info. An editor fixed it, then the image was deleted a few minutes later by a script run by an admin who had reviewed the pre-fix version. As I recall, the list of images was fairly lengthy, so the delay between review and actual deletion was quite long; the servers may have also been backlogged and delayed script work. Gimmetrow 01:51, 28 April 2009 (UTC)

I've occasionally run semiautomatic scripts where I reviewed all the edits ahead of time, then the script made the edits for the next hour or so while I did something else (I was still online and available from my talk page in case of problems). The difference between what I was doing and what you describe is that the script was set up in such a way that any edit to a page between my review of that page and the script editing it would have resulted in an "edit conflict" (which would have prevented the script from editing that page at all and logged it), even if the edit happened half an hour before the script got to it, as long as it happened after my review. This is why we do not need a hard and fast rule here, otherwise you will encounter border cases that do not fall into your narrow definition of a semiautomatic script, but are in fact semiautomatic. In my opinion when the review happened should be irrelevant, so long as the reviewed version of the page to be changed is the same as the version that is in fact changed.--Dycedarg ж 02:11, 28 April 2009 (UTC)

I could see it being semiauto if the edit is reviewed, then sometime later, immediately before the edit is actually attempted, the page is downloaded again to verify it's the same as the formerly reviewed version. Is that what you're doing, and could we count on other scripts doing that? Gimmetrow 00:22, 29 April 2009 (UTC)

What the script I was using did was as follows: The timestamp of the revision I'd reviewed was compared to the timestamp of the latest revision of the page at the time the script was about to edit it, and if they did not match then the script didn't edit. So it accomplished the same thing. The asynchronous editing function in Pywikipedia does this automatically from what I can tell by looking at the code, so any semiautomatic script built using that would do it. As for other scripts, you can only count on this kind of checking if the person writing it was responsible enough to implement it. I would not have any problem with the bot policy specifying that this must be the case for these kinds of scripts; I would have thought that it would be a basic rule of common sense for this sort of thing anyway.--Dycedarg ж 05:09, 29 April 2009 (UTC)

Unfortunately, common sense is far from common. Anomie⚔ 12:23, 29 April 2009 (UTC)

Indeed it is. So how about a minor change to the policy inserting something along the lines of: "Edits to be made by semiautomatic tools may be reviewed en masse before a script makes the edits autonomously, so long as the pages are checked to ensure that no edits were made to the page in the intervening time between the review and the actual edit, and either the editor is available to stop the script or the script can be stopped by someone else (such as by editing the editor's talkpage) in case something goes wrong."--Dycedarg ж 21:05, 29 April 2009 (UTC)

And then someone will come up with some new corner case and we're back to square one. I think it would be best to not try to define such things and just handle them on a case-by-case basis. Mr.Z-man 21:21, 29 April 2009 (UTC)

Seconded. Anomie⚔ 22:45, 29 April 2009 (UTC)

Please see the discussion here. Wronkiew (talk) 06:41, 27 April 2009 (UTC)

This is a 'mandatory' notification to all interested parties that I have accepted a nomination to join the Bot Approvals Group - the above link should take you to the discussion. Best wishes, -- Tinu Cherian - 10:58, 1 May 2009 (UTC)

Per the required "spamming" of venues, I would like to bring attention to my nomination for the Bot Approval Group, which may be found at Wikipedia:Bot Approvals Group/nominations/Nakon. Thanks, Nakon 01:23, 17 May 2009 (UTC)

I think it would be good to clarify long-standing practice that if source code has been released by an author and the author is able and willing to run it on a specific project, they have the right of first refusal. Any objections to clarifying this? --MZMcBride (talk) 22:39, 27 May 2009 (UTC)

Yeah, I don't think this would be useful. It's not really a right. In some controversial cases, a task may require a particularly cautious and polite bot-runner, who may not necessarily be the person who wrote the code. Consider that in the date delinking case, we may find ourselves under the following restriction: "The Bot Approvals Group will require that the operators selected to perform any date delinking have a history of being able to handle complaints well, and willing to pause their bot when problems have been identified." This may well not be the author of the code, and I don't think giving the author "right of first refusal" would help the project. It's a good tradition, but a bad explicit guarantee. – Quadell ^(talk) 03:59, 28 May 2009 (UTC)

I don't think etiquette should be encoded in policy, although IMO an essay might not be out of place. Anomie⚔ 11:09, 28 May 2009 (UTC)

BTW, it would probably be a good idea to bring up the "Can you remain WP:CIVIL and respond promptly even if your talkpage is flooded with many repetitive uncivil complaints?" question on any "enforcement" BRFA. I'll have to try to remember that. Anomie⚔ 11:09, 28 May 2009 (UTC)

Discussion about policy subcategories for several pages, including this one. As far as I know, this doesn't make any difference, except as a help to people trying to browse policy. - Dank (push to talk) 03:14, 9 July 2009 (UTC)

Hello there. Just to let you know that I (Kingpin13) have been nominated for BAG membership. Per the requirements, I'm "spamming" a number of noticeboards, to request input at my nomination, which can be found at Wikipedia:Bot Approvals Group/nominations/Kingpin13. Thanks - Kingpin¹³ (talk) 08:01, 20 July 2009 (UTC)

WP:BOTPOL is and has been in the "behavioral" policy template. I don't think it makes any practical difference whether the page is in the conduct cat or the enforcement cat ... I had guessed "enforcement" since all of the conduct policy pages seem to be primarily about user conduct. If there's no response here, I guess I'll update the behavioral template in a week or so. - Dank (push to talk) 18:30, 23 July 2009 (UTC)

The election, using SecurePoll, for Audit Subcommittee appointments has now started. You may:

• Visit the election "home page" for an overview;
• Review the candidate's statements: Dominic • Frank • Jredmond • KillerChihuahua • MBisanz • Tznkai;
• Or go straight to your personal voting page: here to cast your votes.

The election closes at 23:59 (UTC) on 8 November 2009.

For the Arbitration Committee,  Roger Davies ^talk 07:39, 1 November 2009 (UTC)

There's only one day to go! The Audit Subcommittee election, using SecurePoll, closes at 23:59 (UTC) 8 November. Three community members will be appointed to supervise use of the CheckUser and OverSight tools. If you wish to vote you must do so urgently. Here's how:

• Visit the election "home page" for an overview;
• Review the candidate's statements: Dominic • Frank • Jredmond • KillerChihuahua •

MBisanz • Tznkai;

• Or go straight to your personal voting page:

here.

For the Arbitration Committee, — Rlevse • Talk • 17:02, 7 November 2009 (UTC)

I see that User:PseudoBot deleted a redlink here - http://en.wikipedia.org/w/index.php?title=February_17&diff=322604317&oldid=322604294 - with the edit summary "page linked does not exist".
This would seem to be a straightforward contradiction with the editing guideline WP:REDLINK.
Is it desirable for Pseudobot (or other bots) to do this, or am I missing something here?
(I'm definitely not discussing the desirability or otherwise of the specific article Christopher Colella.)
Thanks. -- Writtenonsand (talk) 16:48, 10 November 2009 (UTC)

Yes and No. Blanket removal of red links is a BAD thing. But the bot is correct in removing these specific red links, It has a specific scope and only works on specific date related articles where red links should not exist. if you take a look at its contribs you will see it removes cruft while leaving the good things per related policy on those pages. β^_command 16:53, 10 November 2009 (UTC)

As Beta says; the bot is restricted to a very small set of pages with a tightly-defined format. In general, appropriate redlinks are no bad thing; the issue here is that individuals without an article are not candidates for the Births and Deaths section of the year pages. Note that the bot is removing the entry, not just de-linking it. Pseudomonas_(talk) 16:56, 10 November 2009 (UTC)

I agree with the above, the bot is limited to a small scope and, within its scope, it is taking correct action. And, yeah, a bot that de-linked all redlinks would be a terrible idea. Useight (talk) 17:18, 10 November 2009 (UTC)

Thanks to all for response on this question. -- Writtenonsand (talk) 14:28, 11 November 2009 (UTC)

Dear colleagues

This is a reminder that voting is open until 23:59 UTC next Monday 14 December to elect new members of the Arbitration Committee. It is an opportunity for all editors with at least 150 mainspace edits on or before 1 November 2009 to shape the composition of the peak judicial body on the English Wikipedia.

On behalf of the election coordinators. Tony (talk) 09:24, 8 December 2009 (UTC)

This is due notification that I have been nominated to become a member of the Bot Approvals Group. My nomination is here. @harej 05:46, 29 December 2009 (UTC)

I've proposed a small change to the {{bot}} template which I think makes it more consistent: see here. Olaf Davis (talk) 10:12, 9 January 2010 (UTC)

Is the a list of bots with administrative rights? --Apoc2400 (talk) 15:59, 5 January 2010 (UTC)

Well I can give you a list of flagged bots with administrative rights, which I compiled by ctrl-f for "administrator" on this page. –xeno^talk 16:06, 5 January 2010 (UTC)

User:AntiAbuseBot ‎(bot, administrator) (Created on 8 January 2009 at 07:43)
User:Cydebot ‎(bot, administrator) (Created on 7 April 2006 at 01:24)
User:DYKadminBot ‎(bot, administrator) (Created on 26 October 2008 at 04:04)
User:MPUploadBot ‎(bot, administrator) (Created on 11 October 2008 at 21:06)
User:Orphaned image deletion bot ‎(bot, administrator) (Created on 26 September 2009 at 08:30)
User:Orphaned talkpage deletion bot ‎(bot, administrator) (Created on 15 May 2009 at 11:26)
User:ProcseeBot ‎(bot, administrator) (Created on 19 January 2009 at 01:32)
User:Yet Another Redirect Cleanup Bot ‎(bot, administrator) (Created on 3 May 2009 at 01:55)

Thanks. --Apoc2400 (talk) 23:45, 5 January 2010 (UTC)

There is also Category:Wikipedia adminbots. Tim1357 (talk) 03:14, 11 February 2010 (UTC)

Let's take this opportunity for a moment of silence for RedirectCleanupBot. Useight (talk) 03:55, 11 February 2010 (UTC)

Hey Wikipedians, I am here to advertise my nomination to be on the Bot Approvals Group. Take a look if you have some time. Tim1357 (talk) 02:24, 16 January 2010 (UTC)

I am currently standing for BAG membership. Your input is appreciated. ⇌ Jake Wartenberg 02:53, 26 January 2010 (UTC)

I have been nominated for Bot Approvals Group membership by MBisanz, and I am posting a notification here as encouraged by the bot policy. If you have time, please comment at Wikipedia:Bot Approvals Group/nominations/The Earwig. Thanks, — The Earwig @ 03:39, 3 February 2010 (UTC)

I have accepted MBisanz's nomination of myself for membership of the Bot Approvals Group, and invite interested parties to participate in the discussion and voting. Josh Parris 03:02, 11 February 2010 (UTC)

Looking at Wikipedia:Bot owners' noticeboard#Reporting a bot and Wikipedia:Bot owners' noticeboard#New interwiki bots behaviour: problems seems to exist consensus on discuraging the contemporary use of -auto and -force parameters with interwiki bots. What do you think? Shoud we add it to "Restrictions on specific tasks"? -- Basilicofresco (msg) 11:03, 6 March 2010 (UTC)

It's already there, but it could be restated with different phrasing; those affected by said rule will often not have English as a first language. Josh Parris 11:19, 6 March 2010 (UTC)

Those operators who do that should have their bots blocked on sight, for disruption. β^_command 13:53, 6 March 2010 (UTC)

Recently the AIV helper bots have been logging out and editing via IP accounts. This has been causing quite a bit of confusion (repeated posts to noticeboards etc.), and runs the risk of the IP addresses being blocked as unapproved bots. I'm not aware that any bots running on the toolserver have edited logged out, but I think it's against toolserver policy (and would be much more serious, since we don't want toolserver addresses blocked). I propose we make it explicitly clear in the bot policy that bots may not edit while logged out, and any that are know to should have an assert function added. Best, - Kingpin¹³ (talk) 14:55, 9 March 2010 (UTC)

Support that, seems common sense. — Martin (MSGJ · talk) 14:57, 9 March 2010 (UTC)

Support from me too. SGGH ^ping! 15:20, 9 March 2010 (UTC)

More support. Josh Parris 16:18, 9 March 2010 (UTC)

  Done No opposition, and is in line with current policies. - Kingpin¹³ (talk) 13:24, 16 March 2010 (UTC)

The Toolserver rules require that bots edit logged in. However, sometimes flukes happen, and personally I'd rather see a page updated with an IP than have the edit or action rejected altogether. --MZMcBride (talk) 17:01, 20 March 2010 (UTC)

Personally I think the idea of assert is to get the bot to log in again (when it realises it's been logged out), and then retry the edit, that way we get the page updated, and it's not done from an IP address - Kingpin¹³ (talk) 19:05, 27 March 2010 (UTC)

For the sake of clarity, in the following sentence:

"Note that higher speed or semi-automated processes may effectively be considered bots in some cases. If in doubt, check."

I recommend adding something such as "even if performed by a human editor". CopaceticThought (talk) 19:05, 20 March 2010 (UTC)

I presume you mean "even if performed from an account used by a human editor". Personally I think this is pretty obvious already, but don't mind adding this to make it clear. - Kingpin¹³ (talk) 19:45, 20 March 2010 (UTC)

Hi, if a bot is not adhering to the requirements and the owner says they don't care (the owner is using it to add a request to lots of talk pages quickly, just as a time saver, not related to the bots usual activity), I'm unclear what the next step is? 0¹⁸ (talk) 16:23, 14 April 2010 (UTC)

In general the next step would be to get some input at Wikipedia:Bot owners' noticeboard or WP:ANI if you think it requires administrative action. In this case I don't think Jarry is saying he doesn't care, as such, anyway I'll go have a chat with him about it. However, if you're going to put yourself in the category of WikiProject members you should be aware that you may be subject to bot notifications about said project, unless you take the time to opt-out. - Kingpin¹³ (talk) 16:37, 14 April 2010 (UTC)

Maybe this policy just needs updating. Something about once you have a bot approved, you have Carte blanche to automate edits if you think that they are generally acceptable. 0¹⁸ (talk) 16:45, 14 April 2010 (UTC)

.... Anyway, I've messaged Jarry - Kingpin¹³ (talk) 16:51, 14 April 2010 (UTC)

Seems Jarry is happy to get approval or use a different bot if he needs to notify project members again in the future. - Kingpin¹³ (talk) 17:13, 14 April 2010 (UTC)

I have accepted Kingpin13's nomination for membership in the Bot Approvals Group, and per the instructions invite interested parties to participate in the discussion and voting. Thank you, –xeno^talk 19:24, 17 April 2010 (UTC)

For some reason, we keep having bots editing while logged out. See the current ANI discussions, for example. One of the toolserver IPs was blocked for a day because of this. Given the large number of admins, most of whom aren't digging bots much, it would be simpler to deal with this type of issue if every bot prefixed its edits with its Wikipedia account name. With this approach, if a bot is editing while logged out, at least it's trivial to find the misbehaving bot and notify its owner. I see above that the toolserver policy forbids bots from edditing while logged out, but nevertheless that happens often enough, and the toolserver itself is not governed by en.wp policies, and enforcement of their own policies seems rather lax. Pcap ping 07:30, 17 May 2010 (UTC)

Well, only a small fraction of the bots edit logged out, and it's really only difficult to catch when they are clones. So I'm not sure I would support a policy effecting all the bots. Plus, it would be no more difficult to implement than for the operators to just fix the logging out problem. The problem is hardly any of them are active. --Kingpin¹³ (talk) 07:46, 17 May 2010 (UTC)

Bots which deal with replacing internal links (such as double redirects) should check to verify that their source information has been stable for at least 48 hours.

In light of this discussion, I propose that we implement the above text (or some variant thereof) into the configuration tips section.   — C M B J   23:13, 13 June 2010 (UTC)

Should it be more generic than just "replacing internal links"? Maybe something like this:

Bots performing operations that are difficult to find for reversion later, such as bypassing redirects, should verify that their source information has been stable for an appropriate period of time. The bot should also regularly publish logs of its actions to a subpage in their userspace with sufficient information for the relevant changes to be quickly found (e.g. "* [http://en.wikipedia.org/w/index.php?diff=prev&oldid=12345 Example]: bypassed redirects "foo" → "bar", "baz" → "baz (disambiguation)"").

Although, I note in this case it wouldn't have made much difference for DASHBot, as the problematic redirect had been in place for almost 2 months. Had Xqbot done so, it would have been safe. Anomie⚔ 23:29, 13 June 2010 (UTC)

As with most blatant vandalism, the source information that caused DASHBot's erroneous edit(s) was reverted within 24 hours—so it is clear that this policy will save us a lot of headaches in the future. I do like the idea of making the clause more generic. I also think that we should include some sort of conditional waiver, so as to not bureaucratically hinder future applicants from creating bots designed to combat vandalism in real time.   — C M B J   01:08, 14 June 2010 (UTC)

Feel free to figure out a good conditional waiver, I used the RFC 2119 definition of "should" without thinking about it. Anomie⚔ 01:28, 14 June 2010 (UTC)

While this suggestion has merit, Anomie is correct in saying it wouldn't have applied to DASHBot in this case. DASHBot acted based on Associative thinking which was stable for 2 months. The "culprit" was Xqbot which changed Associative thinking based on the vandalism to Magical thinking. Xqbot would have had to have this implemented to avoid the problem. As such, I'm notifying Xqt of this discussion. -- JLaTondre (talk) 02:44, 14 June 2010 (UTC)

Thanks for notifying me. In past I had some request according that bot script due to some vandalim problems. The first changes was made in pyrev:7499 which displays the target page with the bots comment. Then I found some vandalised pages using a redirect link by the toolbar. This was fixed in pyrev:7875. Now I am implementing a sort of vandalism checking if there is a redirect on a regular page which gives a mess trying to fix redirects to this nonsensical redirects. The fix is comming soon and I consider tagging a message to its talkpage (or I guess there is a project page for tagging potential vandalized pages, isn't it) for such cases which are definetly wrong. For the remaining case when a page becomes an redirect other than moving it, and the editor is an IP, I would also presume a vandalism and leave it (maybe for 48 hours ant tagg a notice if you wish it). -Xqt (talk) 16:55, 15 June 2010 (UTC)

I spoke to Dab on #wikimedia-toolserver ^connect, who said he would send the announcement in about 2 hours. Anomie⚔ 16:15, 18 June 2010 (UTC)

Thanks Anomie (also for fixing the discussion archive, the timer on my internet café was running perilously close), looks like DaB kept his word and sent it out (thank you DaB). So I guess the 14 days starts now. - Jarry1250 ^{[Humorous? Discuss.]} 21:26, 18 June 2010 (UTC)

Just a little poke here - it appears to me that this block hasn't been carried out yet, and it's been sufficient time, so can someone with the proper bits implement the block? — Gavia immer (talk) 17:55, 7 July 2010 (UTC)

Will the blocking admin please also edit MediaWiki:Blockiptext to remove the toolserver from the list of sensitive IPs? Anomie⚔ 18:08, 7 July 2010 (UTC)

Actually, we're both too slow. Jarry1250 did it yesterday. Anomie⚔ 19:06, 7 July 2010 (UTC)

That's what I get for checking the individual IPs and not looking for a rangeblock. Such is life. Thanks, Jarry. — Gavia immer (talk) 19:10, 7 July 2010 (UTC)

The policy uses the word "should" quite a lot (47 times in the current version). I interpreted the majority of these (39; the others seem to be figures of speech using the word in other ways) along the lines of RFC 2119: basically a requirement, but if you have a really good reason not to (or your bot predates the requirement) then the rule can be bent. However, I've seen others wanting to treat these as suggestions to be ignored on a whim. Opinions? Anomie⚔ 03:40, 15 July 2010 (UTC)

You are invited to join the discussion at Wikipedia Talk:What Wikipedia is not#How is Wikipedia a gazetteer? How is Wikipedia not a gazetteer?. patsw (talk) 12:18, 19 July 2010 (UTC) (Using {{Please see}})

My understanding of the policy on this page, specifically this sentence:

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.

is that I do not need any specific approval for what I want to do: A simple Ruby script to pull some information off certain pages and compile it for my own use. However, because I wanted wiki markup, I did try to access the "action=edit" page (but not post back to it) and got a warning about not having my User-Agent set. I'll fix the issue about the User-Agent, but I want to make sure I'm not violating any policy. - Regards, PhilipR (talk) 02:31, 2 August 2010 (UTC)

If your script doesn't make any edits, doesn't request so many pages that it puts an undue load on the servers, and doesn't do anything else disruptive, probably no one will even notice. For details on the user-agent thing, see meta:User-Agent policy. Also, you should use the API or action=raw rather than trying to screen-scrape the edit form. Anomie⚔ 03:11, 2 August 2010 (UTC)

Hello everyone! I have been nominated for the bot approval group and would like to invite you all to participate in the discussion at Wikipedia:Bot Approvals Group/nominations/EdoDodo. Thanks. - EdoDodo ^talk 02:46, 17 August 2010 (UTC)

Original discussion

Manual high-speed mass article creation/modification using a single manual technique (such as, pasting boilerplate templates) should be treated as semi-automated edits.

Given the heated issue, I suggest a simple and direct proposal. If we have not fully established what "semi-automated" stands for, then perhaps we can decide if high speed use of templates falls under semi-automation. Alternate solutions and proposed rewording are welcome.

• Support. It requires scrutiny and care of operator/editor in making sure the new material is accurate. Judging by the above contributions, even manual human editing is not beyond making mistakes. I see no principal difference between using a script and filling fields of a template — the result is the same, if only editorially cumbersome to produce. In my opinion, any almost equal edits more frequently than 5-7 per minute for more than a couple of minutes is in par with semi-automation, though not "semi-automation" per se. —  HELLKNOWZ  ▎TALK 17:18, 22 September 2010 (UTC)

While it seems a bit like instruction creep, I suppose the clarification is needed to prevent wikilawyering. –xeno^talk 17:19, 22 September 2010 (UTC)

Ditto. This is ridiculous, but apparently necessary. Anomie⚔ 19:09, 22 September 2010 (UTC)

Human editors are expected to pay attention to the edits they make, and ensure that they don't sacrifice quality in the pursuit of speed or quantity. Bot-like edits (i.e. high-speed or large-scale edits that make errors an attentive human would not) are liable to be judged on the same terms as an automated or semi-automated bot, even if the methods actually used are ostensibly manual.
We could use the shortcut WP:Turing test ;)

The dilemma above is that it is not clear whether the restriction on semi-automated article creation applies to the "manual" bot-like template substitution this user was doing. The first proposal tries to cut the knot by expanding the restriction to manual large-scale article creation. This comes at it from the other direction, and basically says we can apply WP:DUCK instead of having to argue over the method used to make apparently automated edits.

I don't have specific examples, but I do recall seeing a few cases in the past at WP:ANI and elsewhere that effectively did this, with the resolution that the offender must be more careful to not make stupid mistakes. Anomie⚔ 19:46, 22 September 2010 (UTC)

Yep, I prefer this to the above which seems altogether too specific. –xeno^talk 19:50, 22 September 2010 (UTC)

Sounds reasonable. The above proposal is also good, but a bit too specific. - EdoDodo ^talk 19:52, 22 September 2010 (UTC)

Well-phrased, support.—  HELLKNOWZ  ▎TALK 20:06, 22 September 2010 (UTC)

Support. d'oh! _talk 16:45, 23 September 2010 (UTC)

• FYI I've copied some of the discussion here from its original location, as this is the correct location to discuss changes to the policy. –xeno^talk 16:32, 23 September 2010 (UTC)

This is nothing but Wikilawyering. Bad policy drive contributors away from project by taking away freedoms to edit.Starzynka (talk) 17:49, 23 September 2010 (UTC)

Of course, there would be no need for this clarification to the policy if you exercised due diligence with your boilerplate articles. You remain free to edit, but you are not free to create countless articles that require cleanup or A10 candidates that give no more information than was already contained in the list of settlements. –xeno^talk 17:54, 23 September 2010 (UTC)

I see.Starzynka (talk) 18:01, 23 September 2010 (UTC)

I have to agree that even with all good faith and intentions, there needs to be discretion exercised. I agree with you on excessive fanboy episode lists and fictional characters filling an undue part of the encyclopaedia. And I support your work on adding material of actual real locations. But as long as it is approved by the community, as biased as it may sometimes be. —  HELLKNOWZ  ▎TALK 18:30, 23 September 2010 (UTC)

This is a really bad proposal. Firstly, good-faith improvement of the encyclopedia is not and should not be prohibited, although disruption can be prohibited. Secondly, using the bot policy to prohibit normal human editing is in error regardless. — Gavia immer (talk) 20:17, 23 September 2010 (UTC)

This isn't about "normal" human editing. This is about "being so careless or inattentive that people could easily believe you're actually a bot" human editing. Anomie⚔ 20:33, 23 September 2010 (UTC)

I have to concur that this is not about "normal human editing", as the method, speed, and accuracy are all differing. This is about when editors cannot tell apart human edition from (semi-)automated editing. Furthermore, this does not prohibit good-faith improvements; if anything, this ensures that no harm comes from changes en mass. —  HELLKNOWZ  ▎TALK 20:38, 23 September 2010 (UTC)

Thing is, I have actually been threatened with being blocked before, purely because I was editing at high speed and an editor assumed that there must surely be something wrong with that. The threat didn't go anywhere, because such edits don't violate policy - but you can understand how I might think that it is a bad idea to change policy in order to make some third party's ignorance a sufficient reason for blocking. — Gavia immer (talk) 20:58, 23 September 2010 (UTC)

This is not about giving more reasons to block, this is about giving clear references to discussion about being allowed to do what you are doing. If anyone threatens to block you, you can always point to your task's discussion and show that no objections were raised by the community. This is both for editor's and other's benefit. Editor benefits by having his tasks approved, so no objection are raised. And others benefit by having no doubts about the validity of such tasks or actions. This is what (I hope) the proposal is about. —  HELLKNOWZ  ▎TALK 21:11, 23 September 2010 (UTC)

The editor would benefit a lot more by not needing permission to improve the encyclopedia. — Gavia immer (talk) 23:27, 23 September 2010 (UTC)

How do "large-scale edits that make errors" improve the encyclopedia? Mr.Z-man 03:03, 24 September 2010 (UTC)

They don't, and remember that I have no problem with preventing actual disruption - such as large numbers of erroneous edits. I am worried that adding more policy encrustation will make editors that much more likely to report or block those making large numbers of constructive edits, simply because they don't see what the editor is doing and "there must be something wrong here". Also, bear in mind that I'm responding to both proposals here; the one above, which luckily has not gained much traction, would subject editors to the bot policy simply for being efficient. I am aware of how bad high-speed screwups can be; I just don't see these proposals as the answer. — Gavia immer (talk) 03:41, 24 September 2010 (UTC)

"large-scale edits that make errors". False. My stubs don't have errors. I just need to remove the bracket for some dabbed article names from the intro and infoboxes after creation that is all and if possible add some initial data. This is not a problem. Starzynka (talk) 10:41, 24 September 2010 (UTC)

My concern here is not just the MOS errors, but a lack of discussion with the WikiProjects and the community before creating the stubs on a large scale. The community involvement will avoid a major cleanup and waste of effort at a later time. — Ganeshk (talk) 12:25, 24 September 2010 (UTC)

It certainly is a problem if you're going around creating large numbers of articles that need further editing to remove brackets and add actual data. Anomie⚔ 15:41, 24 September 2010 (UTC)

Looking in some of your older edits, many of the stubs you created are unsourced, that's a little more important than the subst issue. I certainly hope they don't have any factual errors, considering most of them contain only 1 fact. Most of them are probably orphans as well. In any case, are you actually planning on fixing them? The point here is that these types of errors, while not severe in this case, would not have been made by an editor paying attention nor a bot that had been properly approved. Mr.Z-man 21:44, 24 September 2010 (UTC)

For the purpose of dispute resolution when there is uncertainty whether a series of edits is a done by an automated process, or a user editing manually at a bot-like speed, they may be treated as automated edits.

• The above was cribbed liberally from Wikipedia:Requests_for_arbitration/Regarding_Ted_Kennedy#Sockpuppets which is the basis for WP:MEAT. The wording is concise, and captures the problem: It's not whether or not a bot or a human makes the edits that is the problem, it is the fact that very rapid, unchecked edits can cause a huge mess if they are not done properly. It really doesn't matter if a bot does the edits, or a person does them manually but so fast that it is impossible to actually check what they are doing, its inattentive editing that is the problem. --Jayron32 05:16, 25 September 2010 (UTC)

  It's not only speed that can be at issue. The relentless performance of edits exhibiting careless errors an attentive human would not make is an issue whether it's at a high or low speed. Anomie⚔ 17:23, 25 September 2010 (UTC)

  I support adding this to the wording of Proposal 2. However, "at a bot-like speed" should be omitted. Also, it should be "treated as semi-automated edits" as this is not true automation (i.e. unsupervised bot). —  HELLKNOWZ  ▎TALK 17:30, 25 September 2010 (UTC)

Should the policy include directions on notifying anyone besides the blocked user if an account is blocked for operating as an unapproved bot, or an approved bot operating in unapporved ways? BAG, perhaps? Are there such procedures? --Bsherr (talk) 17:41, 4 October 2010 (UTC)

I don't think it's necessary to always drop a line, but if there were some reason you thought one was needed, you could post to WP:BON. If you block an unapproved bot (or an approved bot doing unapproved things), then directing them to BAG for approval is enough and then BAG will be engaged when that occurs. If they choose not to, then the bot would stay blocked, of course. –xeno^talk 17:45, 4 October 2010 (UTC)

Previous discussion

I was looking back at this discussion last night, and decided to sleep on it without knowing that the bot would archive it before I woke up. I propose the following compromise wording to be added to WP:BOTPOL#Dealing with issues:

====Bot-like editing====

Human editors are expected to pay attention to the edits they make, and ensure that they don't sacrifice quality in the pursuit of speed or quantity. For the purpose of dispute resolution, it is often irrelevant whether high-speed or large-scale edits that involve errors an attentive human would not make are actually being performed by a bot, by a human assisted by a script, or even by a human without any programmatic assistance. No matter the method, the disruptive bot-like editing must stop or the user may end up blocked.

Note that merely editing quickly, particularly for a short time, is not by itself disruptive.

Anomie⚔ 17:59, 17 October 2010 (UTC)

"it is often irrelevant" — remove "often" — it is irrelevant, period. Otherwise, support as before. —  HELLKNOWZ  ▎TALK 18:31, 17 October 2010 (UTC)

Do we have a policy on adminbots whose operators aren't around anymore? I ran across User:Orphaned image deletion bot tonight; it's operated by Chris G: he hasn't edited for a month, and he's had fewer than fifty edits in the last four months. I'm uncomfortable with adminbots that do anything (except for very specialised tasks, such as updating the fully-protected Template:Did you know) without the oversight of an active human editor; would it be reasonable to block this bot and leave Chris a message of "feel free to unblock whenever you want; we just didn't want it to run when you weren't around to look after it"? Nyttend (talk) 01:06, 23 October 2010 (UTC)

While it would be preferable to have active users behind the adminbots, it seems to me that the bot should be allowed to continue unless there's an immediate reason to block it, such as, it's actually malfunctioning, or something raised on the talk page requies attention before the bot should be permitted to continue. Since everyone has the ability to examine bot edits, there's no need to speculate. --Bsherr (talk) 02:25, 23 October 2010 (UTC)

*cough* not inactive, just don't edit fairly often. I do respond to requests regarding my bots (the most common being undeleting images) on my talkpage, and generally I try to respond within 48 hours. --Chris 02:28, 23 October 2010 (UTC)

You are invited to join the discussion at WT:N#What is the consensus on City articles?. patsw (talk) 01:21, 29 October 2010 (UTC) (Using {{Please see}})

I have a feeling that because anybody can make a (unathorised) bot, I believe that at some point there will be user(s) that will program a bot to spread true shit all over the project, and messing up all over and possibly spreading a virus. I think that the bots have a too low security level, and if a bot account ends up in the hands of a clever vandal, we may be faced with a major problem. I think we should take more care about the safety of our bots, to prevent possible abuse. MikeNicho231 (talk) 21:28, 19 November 2010 (UTC)

The last bot that started misbehaving was blocked 5 minutes into its edits. Unauthorised bots (which are just regular accounts/anon edits) show up on watchlists/recent changes so their edits will be highlighted very soon. Bots cannot spread malicious software directly as MediaWiki software prevents that. Currently bot operator is required to have sufficient edits, list enough function details, and run a trial before having their bot approved. What additional measures do you suggest we take? —  HELLKNOWZ  ▎TALK 21:40, 19 November 2010 (UTC)

A bot malfunctioning is not a major issue, as long as it does not behave in any malicious way. What IS concerning, is the risk of a human being programming a bot to behave in a way that may disrupt the whole Wikipedia. A person with some insight in computing will have little problem to do this. I want Wikipedia to improve bot safety in such a way that even the slightest deviation from the intended purpose of the bot causes the bot to be blocked from further work until it has been solved. MikeNicho231 (talk) 21:50, 19 November 2010 (UTC)

While the bot policy does not say "slightest deviation", a bot would be blocked if it performed a task it is not approved for. How do you propose to monitor the bot operation? —  HELLKNOWZ  ▎TALK 21:56, 19 November 2010 (UTC)

An automated script that should prevent any change to the bot tasks. If a bot is programmed into vandalising, spreading malicious software, or if an admin bot is programmed into deleting lots of pages, blocking users, and such, it could be very harmful to the whole community. MikeNicho231 (talk) 09:03, 20 November 2010 (UTC)

How would an automated script prevent changes exactly? To be precise, how is an automated process to recognise what edits are good and what are bad? —  HELLKNOWZ  ▎TALK 11:11, 20 November 2010 (UTC)

The "bot tasks" pages^{[I assume you mean BRFAs?]} here are simply human readable descriptions of what the bots are supposed to do. Changing them will not change what the bot does. The bots are actually run from various servers, where only the bot's operators can change it. -- Cobi^(t|c|b) 11:32, 20 November 2010 (UTC)

Question: does "Mass article creation" also apply to mass category creation? Rd232 ^talk 14:20, 15 December 2010 (UTC)

The discussion and resulting consensus only considered article creation; mass content creation in other namespaces (reader-facing or non-reader-facing) was not discussed. Is there a particular case you have in mind? Anomie⚔ 15:39, 15 December 2010 (UTC)

yes, a particular case inspired the question. I would have thought the same principles applied. Rd232 ^talk 20:50, 15 December 2010 (UTC)

Whilst not following the actual case, I would too have thought that this logically applied to categories, if to a lesser extent. Perhaps another tweak is in order, that while namespace plays a role, one shouldn't think this is article exclusive guide and should exercise due discretion? —  HELLKNOWZ  ▎TALK 20:57, 15 December 2010 (UTC)

There is no particular reason the same principles shouldn't apply to reader-facing content in other namespaces (e.g. normal categories like those are reader-facing, but hidden maintenance categories or non-article categories probably aren't), but there isn't a consensus one way or the other yet that I am aware of. Feel free to start the discussion at WP:VPR or WP:VPP if you want. Anomie⚔ 21:25, 15 December 2010 (UTC)

Proposal—Wikipedia:Bot policy#Bot Approvals Group should be amended so that when applying for BAG membership, editors are not required to post to WP:AN, WT:RFA and WP:VPM. ╟─TreasuryTag►Regent─╢ 20:26, 4 December 2010 (UTC)

I was genuinely astonished to discover that editors nominating themselves for BAG membership are required to "spam" various major fora. What is the purpose of this? Surely anyone interested in the bot process will be watching this page, the nomination page, WP:BON, whatever. It cannot be necessary to clog up already congested noticeboards with this sort of very niche announcement? ╟─TreasuryTag►Regent─╢ 20:26, 4 December 2010 (UTC)

(short statement in response to TT, because I'm about to head off). It mainly was a result of anger with BAG members adding themselves, and people didn't like that. So they kind of went over the top to please everybody. I don't think that requirement is needed anymore. (X! · talk)  · @897  ·  20:31, 4 December 2010 (UTC)

Support removing the requirement to post to WT:RFA and WP:VPM. I think it's fine to make an optional post to WP:AN as the nominations seem to be sparse enough to not be considered spam. Nakon 20:34, 4 December 2010 (UTC)

Keep WP:VPM, as that's what a Village pump is for. And maybe WP:AN, as admins may know relevant history of a candidate but not be watching WT:BAG. No real need for WT:RFA, IMO. Anomie⚔ 20:41, 4 December 2010 (UTC)

@Dycedarg: Interesting bit of history there. At least with the "mass forum spam" people can't claim they didn't have the opportunity to comment. Anomie⚔ 16:31, 7 December 2010 (UTC)

I remember when this was added. Although my memory on the details is fuzzy, essentially a bot got approved that shouldn't have, there was a kerfluffle which resulted in demands for the heads of all the BAG members, and what eventually resulted was a consensus that BAG had gotten to be too much of an insular group that promoted members into itself without community review or discussion, and was completely out of touch with the needs/wants/desires of the community at large. Of course, the reason for that happening in the first place (as it so often is when it comes to these things) is that that vast majority of people don't care about the bot approval group or bots in general until there's a rogue bot or some approved yet horribly ill-conceived bot, so the only people who watch bot-related pages are bot owners/members of BAG/random people who are interested, which hardly makes up a decent cross-section of the community. The easiest way to solve this was mass forum spam.

Of course, it doesn't work. The nomination which alerted you to this requirement has received all of 6 !votes, 3 of which are by current BAG members, and the other three were by editors who have run or currently run bots. All the !votes on all the BAG nominations that have taken place in the last year combined do not add to the number of !votes in a single average RFA. The community doesn't care, and all the spamming in the world won't make them care. The only thing the spam might accomplish is if someone who is utterly unacceptable runs someone might notice the spam posts and bring it to the usual voters' attention, assuming they didn't notice themselves. This might warrant keeping it at WP:VP and WP:AN, but I've personally never understood the part about posting at WT:RFA. It has nothing to do with that. This whole discussion might warrant greater community input, given that it was the community at large who implemented this, and such things generally shouldn't be undone by 5 editors on a talkpage no one reads. Of course, if you did bring it up somewhere else to gain more attention here, no one would care anyway...--Dycedarg ж 07:45, 7 December 2010 (UTC)

I think this was the bot that added protozoa or some such. A storm in a teacup, since the only real problem it created was deciding whether to delete all the articles it had created or merely most. Since deleteing them all is low cost, their must have been some benefit from the the articles saved. I also remember checking out the BRFA at the time, and it was quite reasonable. Since then we had a BRFA on a content creation vein which wanted to create a handful of pages (possibly only one long one) which was talked to death - by which I mean months, not weeks. Rich Farmbrough, 13:16, 17 December 2010 (UTC).

Everyone supports removing WT:RFA. Opinions are split on WP:VPM (3 to 2 in favor of removing) and WP:AN (3 to 2 in favor of keeping), although no reasoning for removal was given while reasons for keeping both were presented. I've gone ahead and removed WT:RFA, but I've left the other two for now pending further discussion. Anomie⚔ 15:46, 15 December 2010 (UTC)

To be fair, I think I presented clear reasons for removal. ╟─TreasuryTag►prorogation─╢ 23:33, 15 December 2010 (UTC)

"Surely anyone interested in the bot process will be watching this page, the nomination page, WP:BON, whatever."? Maybe, maybe not. Unlike RFA, hardly anyone seems to really pay attention to bot matters. An occasional post at WP:VPM and WP:AN is a minimal price to avoid claims that we do things without at least trying to involve the wider community. Anomie⚔ 02:03, 16 December 2010 (UTC)

So what you mean is, "I personally disagree with the reasons for removing WP:AN," not, "Nobody gave any reason for removing WP:AN." ╟─TreasuryTag►belonger─╢ 08:44, 16 December 2010 (UTC)

No. When I looked at the comments, I considered statements like "It's good" or "It's not necessary" to be personal opinion instead of reasoning. I perhaps incorrectly put your unsupported assertion that surely anyone interested will be watching bot-related pages into that category. But hey, now we have some of that further discussion I hoped for.

Back on topic, do you have any comment on the history presented by Dycedarg that lead to the current wide advertising of BAG nominations? That seems to indicate that at one point people did generally make your assumption and it was determined to be flawed. Anomie⚔ 12:24, 16 December 2010 (UTC)

Comment. I suppose later then never. WP:RFA is unnecessary, there is little relevance between the two. WP:AN should be kept, as BAG is "administrative" section and if anything sysops should weigh in, especially if they are aware of some relevant history. I am borderline on WP:VPM. After all the other sections, those interested should have seen the nomination. Perhaps if we need to trim, then VPM is the next in line. Rest are bot-related pages, and are fine. —  HELLKNOWZ  ▎TALK 10:19, 16 December 2010 (UTC)

This is crazy. If we are to do that then we need to unify editing requirements for humans and bots. Having a two-tier system is bad enough, non-human prejudice is to be expected. But when we then say, "Oh and by the way if we wish it, you will be classed as bot, and subject to non-human editing restrictions" the thing becomes Kafkaesque. (Or perhaps I should say more Kafkaesque.) Rich Farmbrough, 11:20, 17 December 2010 (UTC).

The text you removed didn't create any sort of unified editing requirements for humans and bots. It just lets us avoid having to argue with someone over whether their series of disruptive edits that looks like a poorly-done bot is really a poorly-done bot or is actually a human failing an impromptu reverse Turing test, by pointing out that disruptive edits are disruptive edits either way and need to stop. Anomie⚔ 12:16, 17 December 2010 (UTC)

Makes sense to me. Rd232 ^talk 13:32, 17 December 2010 (UTC)

Have to agree on this; bureaucracy aside, we seem to need this addition since some high-speed editing users seem to think they are exempt from being meticulous. —  HELLKNOWZ  ▎TALK 13:49, 17 December 2010 (UTC)

Those who quack like bots, edit at the rate of bots, make errors like bots, shall be treated like bots. Don't see a problem. –xeno^talk 14:07, 17 December 2010 (UTC)

• After reading Rich's comments and others in the old discussion, I have to agree and wonder why this needed and why is it in a policy about bots? From what I read this proposal came about because of a user creating a few articles in a short period of time and not using any bots or AWB. I am not sure what the user did, but it is possible to create a few articles in a short period of time without using a bots, for example in the past I have redirected 13 articles in the space of three minutes without using a bot, AWB or any scripts. I did that by opening multiply edit windows, check each edit to make sure all are good, and finally save them all at about the same time. So for all we know these edits made by the user in question was checked and done in good faith, and since none of the articles was deleted I believe this is the case. In any case this issue is about reckless high-speed editing by humans not bots, so I oppose keeping it here in the bot policy, there maybe another place for this issue to be clarified, but it should not be in a policy about bots. -- d'oh! _[talk] 14:34, 17 December 2010 (UTC)

  "check[ed] each edit to make sure all are good" <-- In that case, you aren't the one being addressed with this policy. –xeno^talk 14:36, 17 December 2010 (UTC)

  Yes, exactly my thoughts. This is about editors who create problems by edits almost like bots, therefore the addition to this particular policy. What better place? —  HELLKNOWZ  ▎TALK 14:38, 17 December 2010 (UTC)

  You need to go back and review the situation a little more carefully. The problem wasn't that the user created a number of articles or did it at any particular speed, it's that he created articles with errors that any reasonably attentive and meticulous editor would have corrected before saving. Your statement "So for all we know these edits made by the user in question was checked" is incorrect, because we do know quite well that either the edits were not checked or the checking was faulty. Your statement "since none of the articles was deleted I believe this is the case" is poor reasoning, as it is entirely possible that the articles were fixed rather than deleted.

  The reason this is in the policy about bots is because this is one of the places people would look for policy in a situation like this: User A makes a bunch of edits with errors that make him look like a poorly-run bot, User B accuses User A of running an unauthorized bot, User A responds "No, I did those edits manually, I swear!", then follows a long unproductive discussion trying to determine just what User A did and whether or not that really qualifies as a bot (with reference to this policy) and so on when the real issue is that User A needs to stop making these bad edits whether or not they were done by hand or with some degree of automated assistance. Anomie⚔ 15:27, 17 December 2010 (UTC)

For me it's less about what the user's doing, than how responsive they are to suggestions that they should desist. Anyone might make a lot of bad edits quickly; they need to be treated as a bot rather than a human if they've responded to criticism like a bot does (i.e. not) rather than like a human does (i.e. stop what they're doing and/or demonstrate that there's consensus for doing it).--Kotniski (talk) 15:37, 17 December 2010 (UTC)

Although there was discussion along those lines in the original proposal, nothing in the final version being discussed here says to treat anyone like a bot. It just points out that it is normally irrelevant whether disruptive editing is automated or manual, as the consequences are the same. Anomie⚔ 15:51, 17 December 2010 (UTC)

But not all bot (or bot-like) editing is disruptive. The case to be considered is where no-one actually objects to the edits themselves, but is concerned merely about the fact that they appear to be being done by an unapproved bot. For me the Turing test in this case is less to do with what the edits are, and more to do with what answer I get (if any) if I leave a query about it at the account's talk page.--Kotniski (talk) 16:45, 17 December 2010 (UTC)

Actually, do I mean that? Remind me - why do we have a bot approval policy anyway? Does anyone actually look at the source code to check it won't screw up? No, right? So all that's actually being approved is the task, not the method? So rather than having a bot policy, shouldn't we have a mass-edit policy, to help determine when any mass editing campaign is/is not appropriate? That would then be separate from the process of getting a bot flag, which would merely reduce to convincing the BAG that the task is automatable and showing them some test edits to show that the program does what it's supposed to do. If an account is making obviously silly edits then you're right, the consequence should be that it gets blocked, regardless of whether it's a malfunctioning bot, an extremely careless person or a deliberate vandal. But if it's not quite so obviously silly, then we should leave them a query first and see if they stop and discuss, thus avoiding an unnecessary block.--Kotniski (talk) 17:28, 17 December 2010 (UTC)

If there is nothing disruptive about the edits, then this new section doesn't really apply. That's not to say people shouldn't discuss whether the user is running an unauthorized bot or not.

Sometimes yes, the code is actually reviewed. It's up to BAG to decide whether a review of the code is required before approval, but anyone can ask for code (if it isn't already posted) and review code if it is. Anomie⚔ 17:51, 17 December 2010 (UTC)

Is there a policy about whether bots may convert HTML name codes, such as "&Nu;", to Unicode characters such as "Ν"? Some of the Unicode characters are hard for a human editor to identify in editing software, while the HTML name code is easy to identify. Thus, such conversions make articles hard to edit. Jc3s5h (talk) 02:12, 7 March 2011 (UTC)

The bot would first have to be approved to do this, which may fall under the heading of "general cleanup" that some bots have been approved to do in conjunction with their more substantial edits. Also, no bot may do this as the only edit to a page or as part of an edit that contains only other "trivial" changes. Whatever conversions are done must also be in line with the relevant guidelines and policies that apply to all editors. Anomie⚔ 04:47, 7 March 2011 (UTC)

{{rfctag|policy}}

Current version: Spell-checking

Bot processes may not fix spelling or grammar mistakes or apply templates such as {{ww}} in an unattended fashion, as accounting for all possible false positives is unfeasible. Assisted spell- and grammar-checking is acceptable when done with due diligence, and may or may not be considered a bot process depending on the editing rate. Such processes must not convert words from one regional variation of English to another.

Proposed version: Context-sensitive changes

Unsupervised bot processes should not make context-sensitive changes that would normally require human attention, as accounting for all possible false positives is generally unfeasible. Exceptionally, such tasks may be allowed if – in addition to having consensus – the operator can demonstrate that no false positives will arise (for example, a one-time run with a complete list of changes from a database dump) or there is community consensus to run the task without supervision (for example, vandalism reversion with a community-accepted false positive rate).

Examples of context-sensitive changes include, but are not limited to:

• Correcting spelling, grammar, or punctuation mistakes.
• Converting words from one regional variation of English to another.
• Applying context-sensitive templates, such as {{weasel word}}.

The proposal is to clarify that bots cannot make "human-like" decisions in context-sensitive content. Currently, the passage mostly addresses spelling. However, same principle applies to tasks that are not necessarily spell checking. Since there is no BOTPOL point that addresses this, BAG is left to come up with all kinds of possible false positives, whereas it should be the operator that demonstrates that the bot is "harmless", i.e. does not do false positives.

It has been voiced and hinted that BOTPOL needs an overhaul. So let's start here and hopefully write a better page. I put some thought into the wording, so if there are specific issues, please do bring them up. I will leave this for comments first before advertising broader. —  HELLKNOWZ  ▎TALK 14:25, 6 March 2011 (UTC)

In general, I think it makes sense. Two comments though:

1. I'm not fond of the use of "problem" in "Applying problem templates" as it reads as a value judgment on the template which is outside the scope of this policy. Though redundant, it might be better to change it to "context-sensitive problematic" templates to avoid that (or find other wording). ✓
2. I always read the "Such processes" in the original "Such processes must not convert words from one regional variation of English to another" as being the assisted editing processes in the prior sentence. That is not supposed to be done with a bot or non-bot. Including it in the bullet list is fine, but I think it should also be included at the end of the assisted editing paragraph (similar to the current version).

-- JLaTondre (talk) 15:08, 6 March 2011 (UTC)

I thought changing WP:ENGVAR per WP:TIES or for consistency is fine, even with tools of "assisted editing"? At least, I never read it as forbidden. The current version was inserted here, so I cannot say for sure what it meant. I placed a clarification sentence in italics above, but let's see if anyone clarifies as to how it is actually supposed to be interpreted as. —  HELLKNOWZ  ▎TALK 15:39, 6 March 2011 (UTC)

Not sure of the specific rationale behind that addition, but we have had cases in the past of people making mass changes of regional variations. I was referring to such mass changes and I think you are correct if instead it was per WP:TIES, it would be okay. I like your italicized addition better than my suggestion. Actually, in reading it, it occurs to me that it address my concern much better and that if people wanted to get rid of the WP:ENGVAR example, I'd be fine with it. A specific example is less important that simply saying semi-automated edits still need to have consensus and changes likely to be contested should not be made. -- JLaTondre (talk) 17:32, 6 March 2011 (UTC)

I got rid of example to not get into specifics. I added "large-scale" so that this implies bots rather than being BOLD or IAR. —  HELLKNOWZ  ▎TALK 18:20, 6 March 2011 (UTC)

It works for me, as long as the italic formatting is not part of the actual proposal. ;) I don't know about the original intention of the flagged clause, but changing it to "follow WP:ENGVAR" works for me too. Anomie⚔ 15:52, 6 March 2011 (UTC)

Italic part got added after JLaTondre's comment. The "follow policies" bit is really just a reiteration of existing principles. "changes likely to be contested", though, is something BOTPOL didn't say before; so that would be new. —  HELLKNOWZ  ▎TALK 18:20, 6 March 2011 (UTC)

I think this is mostly good. The second sentence could use some tweaking. Right now it sort of sounds like if you can demonstrate no false positives, you don't need consensus, which obviously isn't the case. I'd either clarify it to note that all tasks require consensus, or just leave it out entirely, since the rule that all tasks require consensus is stated elsewhere in the policy. I'd get rid of the last 2 sentences entirely, they're not specific to context-sensitive changes and redundant to the section that it links to. Just a short note "All assisted editing tasks must follow the rules", if anything at all, should suffice. But there's no need to summarize other parts of the same policy. Mr.Z-man 16:46, 8 March 2011 (UTC)

I reworded the second sentence to make it clear consensus is needed. The thing is, the last two sentences were added exactly to re-iterate that consensus is still needed, no matter the specifics. But it is true, that the "Assisted editing" section already says "Contributors intending to make a large number of assisted edits are advised to first ensure that there is a clear consensus that such edits are desired." Per you, anomie, and redundancy to the latter section; I'll remove those two then. —  HELLKNOWZ  ▎TALK 17:14, 8 March 2011 (UTC)

I boldly added the wording to the policy. As the previous version was done by a single user without prior discussion (afaik), I believe this RfC + several responses + what BAG usually does anyway is at least more suitable. —  HELLKNOWZ  ▎TALK 10:47, 6 April 2011 (UTC)

OK, that may be enough, but to follow up point 2, how about this:

Any individual wanting to be exempt from clearing each specific mass creation task with the community would have to ask for "pre-approval" to do whatever they see fit, which the community may or may not grant. Such requests should be made at BRFA, and must be advertised at WP:AN. A pre-approval may subsequently be challenged in the same way, and is considered immediately suspended unless or until there is consensus for it to be renewed.

I'm not sure it's worth the complication, but that's what I was thinking of. Rd232 ^talk 10:57, 7 April 2011 (UTC)

I'm required by BAG policy to notify this noticeboard of my nomination for BAG member. Headbomb {talk / contribs / physics / books} 07:44, 20 April 2011 (UTC)

OK, that may be enough, but to follow up point 2, how about this:

Any individual wanting to be exempt from clearing each specific mass creation task with the community would have to ask for "pre-approval" to do whatever they see fit, which the community may or may not grant. Such requests should be made at BRFA, and must be advertised at WP:AN. A pre-approval may subsequently be challenged in the same way, and is considered immediately suspended unless or until there is consensus for it to be renewed.

I'm not sure it's worth the complication, but that's what I was thinking of. Rd232 ^talk 10:57, 7 April 2011 (UTC)

I've moved this back from the archive and have notified the participants about this proposed clarification. Cunard (talk) 08:50, 8 May 2011 (UTC)

• "..to do whatever they see fit" -> "..to run any mass article creation tasks." Let's not give leeway to wikilayering. —  HELLKNOWZ  ▎TALK 09:06, 8 May 2011 (UTC)
• "must be advertised at WP:AN"? Not sure that's a necessary step. Perhaps VP and related projects instead? After all, AN is really for admin issues and less content creation. —  HELLKNOWZ  ▎TALK 09:06, 8 May 2011 (UTC)
• "may subsequently be challenged in the same way" In the same way as what? As BRFAs? As in, post on the corresponding BRFA's talk page? —  HELLKNOWZ  ▎TALK 09:06, 8 May 2011 (UTC)
• "and is considered immediately suspended" So anyone objecting immediately suspends the task? I would think at least third opinion/BAG should weigh in before that, especially if the objection is subjective. —  HELLKNOWZ  ▎TALK 09:06, 8 May 2011 (UTC)
• Does this apply to bots, non-bots, or both? —  HELLKNOWZ  ▎TALK 09:06, 8 May 2011 (UTC)

I think "point 2" may be more complication than benefit.

I believe that the original concern was about individuals with insufficient expertise or experience embarking on mass creations (articles, and now categories visible in mainspace) that are hard to fix or undo, and that will have a negative impact with the readership in the short term.

I would think that experienced expert in these things would be given pretty easy treatment after a few passes. I think that in practice, an experienced expert would only need to state his intentions to receive quick approval. I don't think there is a current need to codify this.

To HELLKNOWZ's point "Does this apply to bots, non-bots, or both?". I read it as applying to both, but with "semi-automated article creation task" specifically applying to non-bots. I read it as an instruction to an editor at the limit of their technical ability and judgment to not proceed unilaterally. --SmokeyJoe (talk) 12:39, 8 May 2011 (UTC)