Wikipedia:Bots/Requests for approval/HiDrNickBot
From Wikipedia, the free encyclopedia
[edit] HiDrNickBot
tasks • contribs • count • sul • logs • page moves • block user • block log • flag log • flag bot
Operator: ➪HiDrNick!
Automatic or Manually Assisted: Automatic and unsupervised.
Programming Language(s): php
Function Summary: Warn and report users for possible violations of the three-revert rule.
Edit period(s) (e.g. Continuous, daily, one time run): Continuous.
Function Details: This bot monitors recent changes to report possible violations of the three-revert rule, and issues warnings to users as appropriate. No script can differentiate edit warring and constructive editing accurately all of the time, but most edit wars exhibit patterns that can be easily detected. These edits will be reported to administrators to determine if a violation occurred.
Three-revert rule reports are very time-consuming for even veteran editors to prepare; there are many requirements, and many users just don’t bother. The bot will produce easy-to-read reports showing the article history from the first revert in question to the last.
All warnings issued will be non-bitey, with an aim toward informing a possibility clueless user of the rule and preventing a violation. Reports will not be filed until the bot has evidence of further editing to a page after a warning; however, warnings will not be required for administrators, IP editors, those who have been blocked for edit warring previously, and editors with more than 2000 edits.
The bot is written in php; it runs automatically every five minutes, and makes only a few edits at a time.
[edit] Discussion
Does the bot ignore admins and long time editors "editing warring" with IPs or new users? How about long time editor vs admin? Or admin vs admin? BJTalk 11:58, 9 June 2008 (UTC)
- No. Since all editors are equal before the three-revert rule, anyone that attracts the attention of the program will be reported. It is impossible for a script to tell if a given revert is 3RR-exempt, which is where administrative judgment comes in. All reports made by the bot will be clearly labeled as automated reports, and all users reported will be informed of the report on their talk page. In the 24-hour testing data set that I’ve been running, about 10% of reports are false positives. We’ll have to decide if that is too high to be useful. ➪HiDrNick! 12:10, 9 June 2008 (UTC)
- This may come down to our interpretation over policy but I think that admins should be totally exempt from getting warnings as anything to do with admin edit warring is a mass drama fest and I don't see bots helping. I also think "trusted users" (define that how you will) should get the benefit of the doubt against non-autoconfimred users or IPs. I'd love to see some data posted for us to review. BJTalk 12:19, 9 June 2008 (UTC)
- Admins won't be warned by the bot, since they are presumed to already be familiar with 3RR. I’ll post up the sample set by Wednesday so you all can have a better idea of just what I’m talking about. ➪HiDrNick! 16:01, 9 June 2008 (UTC)
- This may come down to our interpretation over policy but I think that admins should be totally exempt from getting warnings as anything to do with admin edit warring is a mass drama fest and I don't see bots helping. I also think "trusted users" (define that how you will) should get the benefit of the doubt against non-autoconfimred users or IPs. I'd love to see some data posted for us to review. BJTalk 12:19, 9 June 2008 (UTC)
