Wikipedia:Bots/Requests for approval/PinpointBot

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

Request Expired.

PinpointBot

Operator: Rushyo (talk)

Automatic or Manually Assisted: Manually Assisted

Programming Language(s): PHP using bespoke engine

Function Summary: Logs vandalism using a powerful editor-generated ruleset

Edit period(s) (e.g. Continuous, daily, one time run): No main namespace edits

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

Function Details: PinpointBot Respect is designed as a collaborative vandalism-fighting tool. This first bot will be used to detect and log vandalism acquired from database dumps according to a set of rules provided by vetted users. It utilises Wikipedia's greatest feature, collaborative editing, to help deal with Wikipedia's greatest weakness, collaborative editing.

The bot will be capable of automatically vetting users to give them permission to edit rules, using regular expressions, with additional vetting from a human regardless of the automatic vetting decision. The criteria for automatic vetting will be an account in good standing with plenty of contributions and no vandalism warnings in their talk history. Additionally, the bot will check for potential 'problem code' and revoke a user's rights if any potentially wide reaching regex is found.

This will give it a very broad reach and ability to deal with vandalism that no other bot could spot that may have slipped through the net. The bot itself will perform checking on the client-side and will not make any reverts itself in this current guise and hence is no risk to article namespace at all.

If this experiment works well, with few to no false positives, a second bot could be created that uses a stable (week-old) version of the PinpointBot Respect rule library to revert recent changes automatically using tough criteria (requiring a very high weight). However that is not what is being approved here.

Discussion

Sounds like an interesting idea, but the enwiki dumps fail often (including the June dump), will this create problems? Mr.Z-man 17:25, 28 June 2008 (UTC)[reply]

Not hugely. It is not designed to deal with vandalism as soon as it appears. Rather it is designed to scour for hard to detect vandalism on pages that are not often frequented where vandalism may have been missed. If there was no dump for a long time the bot could fall back to generating it's own dump of the real content, filing that and using that as though it were an old dump. Although this would be bandwidth intensive it would not be gigantically more so than manually downloading a dump anyway and would happen very rarely, if at all. -Rushyo (talk) 18:11, 28 June 2008 (UTC)[reply]

Sounds interesting. Where can we see the vandalism it has detected? Or see the rules it would apply? ... -- maelgwn - talk 06:46, 30 June 2008 (UTC)[reply]

The latest vandalism reports will appear on the bot's user namespace whilst the rules (and the capacity to edit them) will appear on the bot's website with a 'last used' listing on the bot's user namespace. This is to ensure un-vetted users do not have the capacity to change them (read: cause havoc). -Rushyo (talk) 09:45, 30 June 2008 (UTC)[reply]

Well considering you are not actually editing outside the bot's userspace and this will be low volume work, I doubt BAG has to say anything. -- maelgwn - talk 09:55, 30 June 2008 (UTC)[reply]

Just to ensure it works fine,

Approved for trial (50 edits). Please provide a link to the relevant contributions and/or diffs when the trial is complete. —Giggy 09:57, 30 June 2008 (UTC)[reply]

Request Expired. Long development cycle, not editing article space. You are free to continue testing in the bot's user space. BJ^Talk 13:31, 15 July 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.