| 24 Nov 2025 |
 raitobezarius | I want to land Flakes migration early in the release window to give ourselves the chance to make this part of 2.95.0 | 23:04:32 |
| raitobezarius | Hopefully once this is done, this should also unlock easier fixes to the various regressions of Flakes | 23:04:49 |
| raitobezarius | (and also let people stack up on new features that we don't want in core) | 23:04:56 |
 piegames | Okay, then how about: We kick the current deprecations down the road another release cycle and focus on more Flaker and npins stuff in the meantime, to take off infra load. This would however require reverting the already merged commits | 23:05:44 |
| raitobezarius | How do you see that doing so will take off infra load? | 23:08:31 |
| piegames | less need for merge queue, no need for community builders (AFnix), no need for you to manually play merge queue | 23:09:08 |
| raitobezarius | Flaker is not a dependency to achieve the Flakes extraction FWIW | 23:09:10 |
| piegames | no it is a dependency for the deprecations | 23:09:28 |
| raitobezarius | I will always have to play the manual merge queue, whether you send deprecations down the road or not, fwiw | 23:09:37 |
| raitobezarius | Community builders are being manifested as we speak right now, it's hard to cancel the operations :P | 23:09:59 |
| piegames | oh nice | 23:10:08 |
| raitobezarius | The only thing it does is reduce the importance of an automatic merge queue now | 23:10:12 |
| raitobezarius | having fun with CloudFlare servers | 23:10:24 |
| raitobezarius | 
Download clipboard.png | 23:10:28 |
| raitobezarius | So, it's really up to you | 23:11:48 |
| raitobezarius | I don't mind playing the merge queue for you because I already do it to some extent, whether you put more load or not, does not change my life too much | 23:12:03 |
| raitobezarius | aloisw you have this chain https://gerrit.lix.systems/c/lix/+/4346/7 sitting for a while, do you want to merge parts of it or is there a plan? | 23:45:20 |
| raitobezarius | cc ma27 | 23:45:26 |
| 25 Nov 2025 |
 aloisw | In reply to @raitobezarius:matrix.org
aloisw you have this chain https://gerrit.lix.systems/c/lix/+/4346/7 sitting for a while, do you want to merge parts of it or is there a plan? It might be nice if https://gerrit.lix.systems/c/lix/+/4563/3 could be merged, as it was already approved (without code owner permission though, so not submittable) and fix a user-facing issue. The rest mostly makes sense together and is still blocked. | 05:20:38 |
 Qyriad | In reply to @commentator2.0:elia.garden
Qyriad: permission to open a dm with you (to chat more easily and maybe even sync about the machines stuff) go ahead | 08:19:42 |
| piegames | raitobezarius: re the merge yesterday, it's fine, I hadn't seen that the commit with the actual deprecation has not been merged, and the others are okay for me if they are for horrors (we talked about the design before so it should be) | 09:49:01 |
| raitobezarius | blocked on code reviews, right? ok for this one | 10:25:39 |
| raitobezarius | at least, trivial smokechecking yesterday in the night did not reveal any problem fyi | 10:26:09 |
| aloisw | In reply to @raitobezarius:matrix.org
blocked on code reviews, right? ok for this one There are disagreements in https://gerrit.lix.systems/c/lix/+/4333, and I am still unsure who is more correct here. I do not see a need to rush the merge before they are resolved. | 11:50:21 |
| raitobezarius | i pinged ma27 to take a look again fwiw | 11:50:36 |
| 26 Nov 2025 |
|  ellie (the hot one) changed their display name from ellie to ellie (the cooler one). | 00:56:10 |
| ellie (the hot one) changed their display name from ellie (the cooler one) to ellie (the hot one). | 00:56:58 |
 jade_ | well, it was lix people including me being extremely pissed about updating it to a newer cppnix api version during the first cves that caused this to get fixed and for nixos-option to become a normal shell script :) | 05:34:49 |
| jade_ | it was uhhh argh. may have lost it already. was a prerelease of lix 2.94 about a month old. i think it happened while querying the s3 for whether paths were present. | 05:35:58 |
| jade_ | absolutely not, i couldn't stand the perf regressions in cache query performance of 2.92 and 2.93 series and was extremely happy when we fixed the perf issues in 2.94 HEAD :) | 05:37:07 |
