Wikipedia:Bots/Requests for approval/BJBot 5

The following discussion is an archived debate. Please do not modify it. Subsequent comments should be made in a new section. The result of the discussion was

Withdrawn by operator.

BJBot

Operator: BJ^Talk (or other sysop)

Automatic or Manually Assisted: Automatic

Programming Language(s): Python using pywikipedia

Function Summary: Process image deletion backlogs

Edit period(s) (e.g. Continuous, daily, one time run): Continuous

Edit rate requested: maxlag=5

Already has a bot flag (Y/N): N/A

Function Details: Yes this is a admin bot. No I'm not high. The image backlogs (among others) are being processed by non-approved admin bots already (often at speeds faster than my bots run unrestricted) or by admins at high speed without properly scrutinizing the image. This bot will do nothing different, it will just be approved. A bot is suited for the task as it can rapidly check if the image is still offending, if anybody has commented on the image's talk, or any other number of checks that could be done. Since this would be an admin bot it would need to pass RfA, so I would first needs to pass RfA (which I don't think is likely to happen). If this request has support here I will first run myself and if that fails enlist a current admin to run the bot in my place. Even though I'm filing this under BJBot a new account would need to be made to separate admin tasks (and possibly be ran by a different person). The code for this bot would be posted and a review period prior to the bot running should be held to catch any bugs.

While having bots deleting images doesn't seem like a great idea it is needed and already being done (and rightly so, BJBot alone has tagged ~25,000 images, BetacommandBot many times that). I think the admin bots should be following all polices in place, running on separate accounts and be approved, rather than be hidden away behind admin's accounts and not spoken of because of the shunning of bots with +sysop. I do understand the opposition at RfA for +sysop bots and I hope I can overcome that, I request that this discussion remains on the technical merits of the bot and leave the politics for RfA.

Here is an example of how images tagged with {{orfud}} could be checked and deleted:

if last run had errors
	halt
begin loop over images tagged more than 7 days ago
if image is not orphan
	remove "Di-orphaned fair use" tag
else if image description page is not in Category:All non-free media
	remove "Di-orphaned fair use" tag
else if image or image talk page has been edited since tagging
	add to list to be processed by a human
else
	delete
end loop

Discussion

*wonders why there is no drama here* – Gurch 07:14, 18 January 2008 (UTC)[reply]

If I were in bag, I'd approve this. But I'm not, so I won't. Majorly (talk) 11:54, 18 January 2008 (UTC)[reply]

How would you be able to test it without access to an admin account? I suspect the logic here is complex enough that any automatic deletion bot is going to need solid testing to have any chance passing RFA. A manual-approval bot/script would be more acceptable. Gimmetrow 05:04, 20 January 2008 (UTC)[reply]

I'll just have to check my code better and pass it off to an admin for testing. The cycles of code/test would be much longer but still doable. I think passing RFA is hopeless for me so I'm going to cross that off and go find me an admin to run the bot. BJ^Talk 05:38, 20 January 2008 (UTC)[reply]

Well, I have experience with pywiki bots, and would be willing to test out a manual-approval bot/script. Gimmetrow 05:54, 20 January 2008 (UTC)[reply]

The problem with that is the goal is to replace the adminbots already running, a manual script wouldn't do that. BJ^Talk 05:57, 20 January 2008 (UTC)[reply]

As for the testing, if BJ wants, he can freely test in on botwiki. Snowolf ^{How can I help?} 00:44, 21 January 2008 (UTC)[reply]

What checks will you have in place to make sure the image wasn't incorrectly tagged? Gimmetrow 03:14, 21 January 2008 (UTC)[reply]
- For images tagged with {{Di-orphaned fair use}} the same check BJBot is already using to tag the images. Check to see if the image is in Category:All non-free media and that it is orphaned. For the time being {{Di-orphaned fair use}} deletion is all that is going to be done, I'm going to reword the function details to be more specific. There is also a Mediawiki bug that needs to be adressed before the bot could be 100% accurate. BJ^Talk 03:45, 21 January 2008 (UTC)[reply]
  - Will you, for instance, check the image history to see if it was ever tagged with a free license (PD, GFDL, CC-BY, etc.), like a human admin would do? Gimmetrow 03:51, 21 January 2008 (UTC)[reply]

I'm withdrawing this until I get the code complete and somebody else to run it. BJ^Talk 10:50, 25 January 2008 (UTC)[reply]

The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made in a new section.