Talk:Bot policy/Archives/2016
 | Please do not post any new comments on this page. This is a discussion archive first created in 2016, although the comments contained were likely posted before and after this date. See current discussion or the archives index. |
Speed limit
Hi all,
I was wondering a bit about whether the server load issues for edits still warrant a maximum of 12 edits per minute. This rate is in the policy since day one (september 2007). I could imagine that given the years past and possibly improvements in technology higher limits could be possible.
From my experience there are users getting these rates with normal edits (maybe up to 20-30 on extreme minutes), the difference with bots being that they likely don't keep this up for an hour. Using semi-automatic tools such as cat-a-lot and visualfilechange on Wikimedia Commons users easily achieve speeds of 100s of edits per minute (and I'm guilty of those speeds as well with semi-automatic edits and categorisation). I haven't ever seen stories of this influencing server wide edit speeds or lags (but I could have easily missed something like that).
Based on this a few questions from my side:
- Are editing peaks resulting in slower edit times for other users?
- With which speeds does performance decrease?
- How is this caused/is this dependent on global edit speeds?
- If this is really an issue wouldn't it be better to have some automatic method to slow down overspeeding bots, instead of allowing them to influence performance?
Greetings, Basvb (talk) 17:19, 4 February 2016 (UTC)
Is this useful nowadays?
Open question. Is this policy useful nowadays, with most IWs on Wikidata? Should we work on a new global bot policy or rather abolish it? —MarcoAurelio 14:40, 15 February 2016 (UTC)
- How many global bots do we have actually (although, that' sright, we don't need them for interwikis)? -jkb- 14:53, 15 February 2016 (UTC)
- 13 users, with ~60 global bot flags removed this year due to inactivity. —MarcoAurelio 14:55, 15 February 2016 (UTC)
- Oh, that's less as I was afraid. Might be we should ping the users so that tey can provide a summary of the actions they do after interwikis have been moved to Wikidata. Do they see the necessity of a global bot flag or could it be anough to aply some local flags instead. -jkb- 15:06, 15 February 2016 (UTC)
- The global bot policy is not merely a global bot policy, there are non-global bots running under the policy as well. The point is to avoid having hundreds of bot policies, which would make the life impossible for hundreds of wikis. Nemo 15:09, 15 February 2016 (UTC)
- Oh, that's less as I was afraid. Might be we should ping the users so that tey can provide a summary of the actions they do after interwikis have been moved to Wikidata. Do they see the necessity of a global bot flag or could it be anough to aply some local flags instead. -jkb- 15:06, 15 February 2016 (UTC)
- All the individual parts of the policy seem still useful to me: it's still used actively in over a hundred wikis (judging from the active bots). It would be even more useful to make more wikis adopt the global policy, to further ease the life of certain bot owners who take care of certain tedious tasks. Nemo 15:09, 15 February 2016 (UTC)
- I agree that it's not only that and I suggest that we keep and enhace that. What I'm concerned is that the global bot flag and the automatic approval clauses might get useless if we don't expand the scope of the policy. Maybe we should change the system from opt-in to opt-out and have it enabled by default on newly created wikis and every other that has not explicitly opted-out of it? —MarcoAurelio 12:08, 25 November 2016 (UTC)