While browsing the groups page, I noticed that a lot of the popular groups are “headless”. What I mean by this is that the admins don’t seem to be active in the group (or on the site at all). I made use of the API to find all such groups, and it turns out it’s over a quarter of all the groups. Here are the top 10 (the users are admins, and the time stamps are for the last game ended):
Kyu Level Study Group # members: 6551
rising.star 2016-03-24T19:59:01.555038-04:00
Double Digit Kyu Room # members: 2637
Tchoup None
Deutsche Gruppe # members: 488
V1rtu4ltr3e 2015-01-05T16:25:30.911251-05:00
OGS Mini Tournaments # members: 356
Polish Go Players # members: 285
prozz 2019-03-12T17:38:11.513563-04:00
Marek 2013-04-10T16:25:16.511623-04:00
led 2019-01-04T13:30:46.241857-05:00
И Го и жизнь похожи, хорошо играть нужно и там и там! # members: 237
GO Beginners # members: 222
shaolin 2014-01-02T14:48:40.902277-05:00
Australia # members: 199
DavidMitchell 2014-07-09T07:28:29.153354-04:00
RENGO (Pair GO) # members: 198
LinuxGooo 2014-11-06T13:32:37.331742-05:00
Rengo-Tournament-Organizer 2014-07-24T16:32:41.586333-04:00
Meepsie's Community # members: 180
Kuroneko 2018-06-02T19:44:52.755348-04:00
Also, if anyone has tips on how to handle rate-limiting for the API, it would be greatly appreciated. Right now I just sleep for 15 seconds before making a limited API call, but I wonder if there is a faster, more elegant solution.
AFAIK users are not allowed to own more than two or maybe three groups, so that’s perhaps an issue.
That’s the reason why I had to relinquish adminship of the Cat League, because I needed to own another group for something (I can’t actually remember why… might have been something do with the Teaching Ladder revival project).
It might be good to have some kind of a group review.
eg. what are the different types of groups, first of all?
National / subnational
Strength-based
Other
etc.
And then, which ones are active? Which are very small? Should admins be contacted and asked whether they want to retain the existence of their group?
I don’t doubt that there are lots of group with inactive admins (so this is certainly a real issue worth discussing), but I would like to point out that it is possible to be active on OGS without actually playing games of your own
No doubt! Kugsu League is one example! Although in that case, there are still other “traditionally active” admins.
I thought that last game played was a pretty good stat for a back-of-the-envelope calculation because non-playing users are relatively rare. Let me know if you think there is a better publicly available metric
Aw, I hope you still have adminship of the Bug League
Your idea of tagging groups with characteristics is nice, and I think that would make browsing groups easier (I certainly think the groups page can be improved: Issue #1398). However, I don’t really see how categorizing groups plays into this- maybe you can expand on how that’s related to the OP?
I was thinking yes, but not necessary. Someone who hasn’t logged in in a very long time probably doesn’t care too much about that one OGS group they created years ago. In the case of public groups, I don’t believe there is sensitive information that would be in danger in the event of an admin switch.
