[01:10:13] !log tools.stashbot Updated to 5c3e0a8 and updated venv packages (T340675, T343157) [01:10:20] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools.stashbot/SAL [15:55:09] One interesting point from the glamwiki unconference at WCNA - does the proliferation of tools (each new tool gets a new account instead of a mega tool that houses multiple things) make it harder to share maintenance of tools? [15:55:43] (mentioned by musikanimal) [16:00:21] yeah, my understanding is WMCS encourages different tools for different functionality; but for instance, GLAM could be a collection of tools as they serve a similar purpose, and could be community-maintained rather than limited to whoever first wrote the tool. I almost envision it like wiki pages; someone starts a tool, but we collectively maintain it (of course with some consideration to security, bad actors, etc) [16:04:01] hmm. I've thought that it's easier to trust someone enough to give them access to a single thing, versus giving them access to large chunk of somewhat unrelated tool [16:04:49] also, "we collectively maintain this thing" unfortunately tends to mean "no-one maintains this thing" [16:05:18] toolforge has functionality to give an entire tool access to another tool, but the UX on that could be improved a lot [16:11:54] I think having multiple people listed as maintainers who are inactive is still better than 1 maintainer who is inactive, even if the end result is that all maintainers are inactive [16:12:53] fair [16:19:01] !log tools.sbot switch renamecats.py script to run under Python 3 [16:19:04] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools.sbot/SAL [16:20:54] ^ me, I also filed T350953 [16:20:58] T350953: Steinsplitter's tools need a new maintainer - https://phabricator.wikimedia.org/T350953 [16:31:56] "I've thought that it's easier to trust someone enough to give them access to a single thing" <-- I think the GLAM community is a bit different in that most people already know each other, or certainly they know everyone's real name + institution, etc. so trust isn't as big of an issue as in normal wiki stuff [16:32:17] and +1 to tools being used as ACLs as being underused