@lightweight I suspect that in most cases, "enterprise" editions are mostly just their specific server configs anyway, along with other stuff that wouldn't be inappropriate or useless to publish. Anything in #GitLab that the self-hosting community actually has a need for tends to get moved to the "#FOSS" edition once someone makes a case for it.
@strypey sadly, that doesn't seem to be the case with Gitlab. For example, the anti-spam user/content provisions... are only in the Enterprise version, and they don't seem to have plans to port them to the Community edition.
@strypey no - the fact that currently, there's no mechanism in Gitlab to allow admin vetting of new accounts, meaning that if you have open registration, it's easy for spam users to create accounts, and very difficult to delete them en mass.
@lightweight right, but what I'm saying is that up until recently they didn't have an in-house system for that either. They just used #reCAPTCHA, complete with non-free JS. There were a bunch of complaints about that on the Issues a few months back and they have been working on a #honeypot based alternative. Last I checked in they're testing it on GL.com, see:
If it's confirmed to work, I can't see why they wouldn't include it in the "FOSS" edition.
@strypey the main issue I have is how to delete large numbers of spam users and associated spam projects due to a year or two of open registrations and no one really paying attention to who's creating accounts... there's a new "bulk delete user" admin feature, as well as a "admin moderated account creation" workflow that have been implemented for the enterprise version, but not push to the #FOSS version.
@lightweight their practice seems to be to test new features on GL.com, where they can control for more variables, then rollout the code once they're sure the feature actually works. Also, as I say, the squeaky wheel seems to get oiled. I've seen code released after requests in HN threads. If you hit them up about needing those features on your instance, I suspect they will either reply with a timeline for that in their current plans, or add it to their release plans.
@strypey at the bottom of the thread you cited re reCAPTCHA, https://gitlab.com/gitlab-org/gitlab-foss/issues/46548 someone describes a way to remove existing spam users.... I think it'd be great (as do others) but no indication if it's been recorded and the issue's now closed...
> the issue's now closed
True, but removing spam users was not the focus on that issue. there's a whole cluster of issues related to the reCAPTCHA system and its proposed replacements, along with other user managements issues that have fallen out of those discussions. If you care to explore you can follow links from that issue to some of the others. But it seems all issues are being moved to the ee repo to reduce duplication, so it's all a bit hard to navigate at present.
@gitlab self-hosting users want access to some of the spam prevention and user management features that have recently been rolled out on GitLab.com. What's the prospects for that? a) is there already a timeline for moving these features to the free code edition? b) if not, could you be convinced to do so?
The social network of the future: No ads, no corporate surveillance, ethical design, and decentralization! Own your data with Mastodon!