00:05 rakkable left 00:06 rakkable joined
lizmat no, that will just .throw the Failure if the block exited with a Failure 00:08
00:08 rakkable left, rakkable joined 01:30 kjp joined 01:59 kjp_ joined 02:00 kjp left 02:09 kjp_ left, kjp joined 02:33 kjp left 03:40 kjp joined 03:44 kjp left, kjp joined 05:22 kjp left 07:39 librasteve_ joined 09:49 librasteve_ left 15:05 Geth left 15:06 Geth joined
[Coke] github-- again for suggesting someone wants us to buy copilot for business. 15:22
japhb :eyeroll: 17:54
lizmat maybe it's time to move away from github 17:55
[Coke] eh. it's hidden on the settings somewhere, I'm just annoyed everytime I see it 17:57
japhb lizmat: Got a good replacement that is owned and operated in the EU? (I think we've had a similar conversation, but I forget the results.)
[Coke] I'm in no rush to move, fwiw. 17:58
Don't want to impact release management, for one.
japhb Nothing says it has to be an all at once move 17:59
lizmat japhb: still investigating....
japhb The biggest things currently keeping me on GitHub are: 1. Being able to point to my repos and activity stats when prospective employers ask if I "do any open source stuff". 2. Markdown+Mermaid+LaTeX+... for docs, rendered on the fly. 3. CI/CD that I have some basic clue about. 4. Network effects for forks/PRs 18:02
But none of those are particularly strong or something that no one else could provide, aside from maybe the network effects. 18:03
lizmat: And thank you for that!
nemokosch Codeberg is something to consider, that's where Zig moved 18:11
melezhik. Ok. My two cents - forgejo ( the same as codeberg technically ) plus dead simple ci )) instead of yaml pipelines 18:25
If of course for self hosted 18:26
I recently announced some public forgejo instance of mine to play with
japhb When you say "the same as codeberg technically" do you mean "It literally uses the same server software"? 18:36
18:40 068AAN878 left, disbot12 joined, nine left 18:41 nine joined
disbot12 <melezhik.> Yep “Codeberg.org is a public instance/platform, and the primary, showcase instance for the Forgejo software.” 18:42
<melezhik.> Forgejo itself in turn is a gitea fully open source fork 18:43
<melezhik.> Also that - www.reddit.com/r/linux/comments/zn..._platform/ 18:44
<melezhik.> Now - dead simple ci ( by my authorship ) is external ci runner for forgejo replacing yaml ( actions ) by general purpose languages 18:45
patrickb I think a move brings with it loads of work with little immediate benefit. Also I believe this is not a situation of "if only there is a champion willing to do the work, then let's go". Because many people will have to adapt. 18:53
Just to make that clear: I'm not a GitHub fan at all and for quite a while I've created new repos over on sourcehut instead of GitHub. 18:55
disbot12 <melezhik.> Yep. I invite btw everyone to try out ( beta testing ) my forgejo instance integrated with dsci - deadsimpleci.sparrowhub.io/doc/README , at least it might give a sense what an alternative may look like even without choosing dsci itself , but it’s kinda cool 😎 and I am looking for beta testers )) 19:05
<melezhik.> Source hut is good however I am not a fun of sending patches by email thought for some teams this may fits he best … 19:06
<nemokosch> Patrick made a good point. I know for myself that leaving Github will be a casual nuisance "with no benefit" (apart from breaking ties with Microslop) 19:42
20:16 [TuxCM] left 20:18 [Coke] left, [Coke] joined 20:19 [Tux] joined
japhb I think the biggest benefit would be getting out of US jurisdiction. But yeah, this sure as heck isn't a freebie operation. 20:24
[Coke] As the person who inadvertently started the conversation, I don't think we have the resources to work on a migration, even if we do it a bit at a time. 20:44
japhb [Coke]: I actually was thinking more along the lines of "Figure out a suitable replacement, make sure tools like mi6 do the right thing there (e.g. generating proper CI scripts), and let people know it's available *if they want to switch their own hosting*" We don't even have to say it's a GitHub *replacement*, but rather a *fully supported alternative*. 22:55
(And one that we've researched enough to know that it's more likely to satisfy folks unhappy with GitHub for various reasons.) 22:56
[Coke] We could do that, but even that seems like a lot of work for something people are probably pretty opinionated on their own about 22:59
... but if someone does it, sure, share it. 23:00
japhb :+1: 23:01