| 11 Aug 2025 |
 uep | The common thread / element (indecisive pun selection failure) with moving platforms, whether irc or matrix or zulip, and possibly even between rooms as v12 causes fallout, is that there are two forces and a lot of inertia in between:
- some kind of pull, reasons to set up a new platform, change required
- some kind of push, forcing users to make some kind of change
The former is rarely strong enough to draw in enough of the userbase, enough will stay behind that it just amounts to a community split and something about the new platform needs to be really compelling. The latter is always disruptive and damaging and will just push some users away.
| 23:57:47 |
| uep | * The common thread / element (indecisive pun selection failure) with moving platforms, whether irc or matrix or zulip, and possibly even between rooms as v12 causes fallout, is that there are two forces and a lot of inertia in between:
- some kind of pull, reasons to set up a new platform, change required
- some kind of push, forcing users to make some kind of change
The former is rarely strong enough to draw in enough of the userbase, enough will stay behind that it just amounts to a community split and something about the new platform needs to be really compelling. The latter is always disruptive and damaging and if it's forceful enough to have any impact will just push some users away.
| 23:58:49 |
| 12 Aug 2025 |
| uep | I'm still happy to explore zulip, without a huge amount of available effort in the next forseeable few weeks, but my expectations of it being dramatically better to have enough pull are minimal | 00:00:36 |
| uep | * I'm still happy to explore zulip, without a huge amount of available effort in the next forseeable few weeks, but my expectations of it being dramatically better to have enough pull, as above, are minimal | 00:00:43 |
 emily | (I did presume that in the event of any switch the Matrix rooms would just be closed down; maintaining two platforms of course makes no sense) | 00:26:54 |
| uep | matrix being matrix, that may not actually be possible. there are enough old rooms that have been upgraded and tombstoned, but where people are still stuck and active | 00:35:54 |
| uep | (but yes, that would be part of the plan if we were to make such a switch, even with the limitations) | 00:36:55 |
| emily | yeah, the ghost rooms are at least orders of magnitude less active though | 02:52:49 |
| emily | (and my presumption is that if every room was clearly decommissioned and sites etc. didn't point to the Matrix then they'd be less active still, because less expectation that a room that appears to be the main Nix Offtopic room should be active) | 02:53:42 |
|  @i.am.dario:matrix.org left the room. | 06:40:23 |
 Gi Are | zulip.com, or something else?i remember a library called rulip a long time ago | 10:32:54 |
 Charles | yes, zulip.com zulip. it can also be selfhosted, no idea what nixos would prefer to do there | 14:30:27 |
| emily | you do need their servers to do push notifs through the app | 14:34:14 |
| emily | but they offer that for self-hosted users too | 14:34:20 |
| emily | and it would be $0 both SaaS and self-hosted for us | 14:34:26 |
| emily | no idea which would be preferable in terms of autonomy and infra burden, but there is at least full export/import both ways, so the lock-in would be minimal | 14:34:56 |
| emily | (self-hosted is also fully FOSS and has I believe equivalent feature set; it's just getting to use their Apple/Google-approved notification servers that has any involvement from the corp attached) | 14:35:37 |
 QuadRadical (Ping) | is zulip actually used for non-corporate stuff? | 14:35:37 |
| emily | Rust uses it | 14:35:45 |
| QuadRadical (Ping) | i dont hear about it very much at all | 14:35:48 |
| QuadRadical (Ping) | hmm oke | 14:36:20 |
 @emma:rory.gay | call me a fangirl or whatever but i'd prefer to stick to matrix personally | 14:36:52 |
| @emma:rory.gay | despite its flaws | 14:37:13 |
 K900 | I like the ambition of Matrix | 14:37:53 |
| K900 | But I don't really trust it to actually ever get there, at this point | 14:38:04 |
| K900 | You basically need a full protocol redesign to fix the thing roomv12 is trying to patch | 14:38:24 |
