The npm community forum has been discontinued.
To discuss usage of npm, visit the GitHub Support Community.
Request for more details about: Moving from Github issues to npm.community
Hey all. Looking forward to seeing how the new forum works out, as well as the other changes.
That said, I have a few questions regarding code:
- Currently npm/cli returns a 404 on GitHub, when is it anticipated to go live?
- Is it correct then that npm/registry and npm/www are moving to closed-source development?
Cheers, hope all is well.
- idk. We’ll have it up soon probably. Definitely before we shut down the main one.
npm/wwwhave been closed-source for a very very long time. Those issue trackers are only there for public users to post issues, so this changes absolutely nothing on that front.
Woohoo, thanks for the speedy reply.
- Ok great! Any timeframe? Edit: Just spotted it in the last paragraph of the blog post
We plan to archive the repositories on the 12th or 13th of July, 2018. There will be another post here at that time.
- No worries
To throw a new one to the mix:
- What was the rationale for archiving npm/npm and setting up npm/cli? If it was purely to abandon the issues, then that can be disabled from GitHub’s settings for the repository.
The rationale is because you can’t disable issues without making all historical issues 404. That means that if we just shut down the issue tracker, there would be no way to search old issues or reference back, or even copy-paste things that you wanted to move to this tracker instead. It would just disappear entirely. And if we locked down
npm/npm and renamed it, things would be redirected, but as soon as we created a new
npm/npm, all the redirects would disappear. So there’s not really any way to preserve that repo name :\
There’s also no way to lock github issues without also locking down the rest of the repository. We still wanna use PRs, after all.
We think it’s super important to have that old archive there for reference purposes.
Ok! that makes a lot of sense. We’re currently facing the same issues over on the Bevry organisations. We want everyone to use our discourse forum, but kept all the issues trackers open.
I guess considering we have issues for hundreds of repositories, it would not be feasible to fork every repository, so our only hope would be for GitHub to release a read-only mode.
The forum seems great! Thanks for answering all my questions so fast! Appreciate it.
So how are you planning to track actual bugs? Surely a discussion forum isn’t the best platform for that sort of thing?
Head on over to --> #bugs <–
It does work pretty well! And it’ll make it much easier to sort through the noise and raise the quality of bug reports. In face, Discourse itself uses Discourse for all its issue tracking. They’re not alone, either: even GitHub has started using it for its own platform issue tracking, and even Electron/Atom.
I have no qualms at this point about Discourse being able to handle the firehose that is npm issues, and I’m sure we won’t be the last ones to migrate. They’re really making it into a great experience!
I’m gonna reopen this for the time being, because folks are bound to still have questions once things kick in for real this week. One small beef I have with Discourse is that the doomsday autoclose timer seems to follow a topic around based on when it was originally posted, so posts like these that started off in #support get closed pretty quick.
I think for the future, it would be nice if one could comment on the articles in #announcements directly.
People with trust level 2 and above are able to do that. Basically, if you want your commentary up in #announcements, it’s necessary to have some reasonable level of participation in the wider community.
If you want to have a conversation otherwise, you can create a topic elsewhere. I think this filter (possibly bumping to tl3 later) is enough to have some level of trust about the quality and nature of comments there.
(I configured it to work that way earlier but it seems like I failed to save the setting at that time. Should be corrected now)