Login Options for discourse.gnome.org

The discourse website currently only accepts Google :grimacing:, gitHub :face_vomiting: and LDAP :thinking: as quicker ways of creating accounts and using the discourse server than the full registration cycle.

Would it be possible to add any of these authentication types: GitLab, Fedora SSO, Ubuntu SSO/Launchpad or just Manual OpenID?

Thank you for any consideration.

2 Likes

If you can find a maintained plugin for those services, then we can look at adding it.

1 Like

What kind of maintenance contract are you looking for?

I hope you mean “can you find someone doing the work, which we can piggyback on for free” and not “can you find someone who we can ask to do the work for free”.

It’s a shame this isn’t django, I could have pointed you at the social-auth python plugins which do all this for you.

I thought exactly the same yesterday when I arrived on this Discourse for the first time.

Gitlab’s own Discourse has a “Log-in with Gitlab” option.

It might be this plug-in – it hasn’t been updated in 4 years, but it might just be because it works and there is no reason to change it. It is from Gitlab themselves after all.

1 Like

Either, though the first is probably easier. I’m not going to maintain a plugin myself which I don’t use. Ideally, it should be in this list: Topics tagged official

Bear in mind as well that any plugins also carry non-zero maintenance burden, so I would want to see there being significant demand or advantages in carrying them.

You personally wouldn’t see an openid plugin as self evidently useful?

In that official list is a maintained OpenID plugin: Discourse OpenID Connect - plugin - Discourse Meta. OpenID requires much less manual configuration than OAuth2.

GitLab works with OpenID GitLab as OpenID Connect identity provider | GitLab and probably the others mentioned do as well.

That said, I agree with the non-zero maintenance burden and don’t personally have a demand for this. But it may be nice to have, connect it to GNOME’s GitLab instance?

The problem is maintenance, no utility. A lot of things are useful, but not everything is worth it.

No, no, I understand completely. I’m the webmaster and programmer for inkscape.org and have the exact same considerations.

In our community this question would be pushed over to the community manager to decide how important it would be to the community development. i.e. even if there was no demand, would it induce demand. And what would be the maintenance costs be?

Does gnome have a community management committee to decide these things?

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.