Overleaf [LWN.net]

Overleaf

Posted Jun 9, 2025 14:35 UTC (Mon) by Klaasjan (subscriber, #4951)
Parent article: The importance of free software to science

With the emphasis in this article on the venerable and excellent (La)TeX, on free software, and on the collaborative nature of modern science, it may also be relevant to point out the usefulness of Overleaf (see, e.g., wikipedia).

to post comments

Overleaf

Posted Jun 9, 2025 16:01 UTC (Mon) by rsidd (subscriber, #2582) [Link] (9 responses)

Overleaf is useful. It is not free, either libre or (except in a crippled form) gratis. Recently (14 May) an outage of a few hours inconvenienced many colleagues. I do use it but am careful to keep local copies. I don't think it deserves mention in the context of this article.

Overleaf

Posted Jun 9, 2025 17:29 UTC (Mon) by leephillips (subscriber, #100450) [Link] (8 responses)

It does not. I was forced to use Overleaf to write a book and the experience is far worse in every way than using a locally installed LaTeX with Vim. Glacially slow, and instead of useful diffs a clumsy interface to track changes.

Like GitHub in relation to Git, it’s a thin layer of mercenary slime befouling a limpid core of free software.

Overleaf

Posted Jun 9, 2025 20:58 UTC (Mon) by Klaasjan (subscriber, #4951) [Link] (1 responses)

I’m sorry to hear Overleaf left such a bad impression. I should say I do find it useful, though.
And I do agree having local backups is important and that working locally in one’s preferred editor can be more pleasant.
Perhaps the Wikipedia entry is overly positive about the free nature of the software?

Overleaf

Posted Jun 10, 2025 1:33 UTC (Tue) by rsidd (subscriber, #2582) [Link]

My bad, they do seem to have an open source version that you can run locally. I don't know how popular that is.

I don't think it is as bad as Lee says. Both overleaf and GitHub have their places. But not really relevant to this article -- overleaf is widely used by scientists, as is MacOS, but as a proprietary tool for creating open science.

Github

Posted Jun 11, 2025 16:40 UTC (Wed) by Klaasjan (subscriber, #4951) [Link] (5 responses)

Since you disqualify GitHub and Overleaf in the same sentence, this makes me wonder what alternatives to GitHub you would recommend, in particular to users (scientists) that want to collaboratively develop software using (mainline) git from their local machines.

Github

Posted Jun 11, 2025 19:07 UTC (Wed) by dskoll (subscriber, #1630) [Link] (4 responses)

I migrated off GitHub to two other solutions:

Codeberg
A self-hosted Forgejo installation.

I also mirror my repos on salsa.debian.org, but that's not freely-available to anyone; you need to apply for an account and (AFAIK) be developing free software. That runs the Gitlab software, which you can use as a service or self-host.

Github alternatives

Posted Jun 12, 2025 9:34 UTC (Thu) by Klaasjan (subscriber, #4951) [Link] (3 responses)

Thanks. I should have mentioned that the request was for a non-self-hosted solution. I'll look into Codeberg.

Github alternatives

Posted Jun 12, 2025 12:22 UTC (Thu) by dskoll (subscriber, #1630) [Link]

Gitlab.com also hosts a free tier, though it says that it's for "individuals working on personal projects and open source contributions."

Github alternatives

Posted Jun 12, 2025 12:55 UTC (Thu) by jzb (editor, #7867) [Link] (1 responses)

I like Codeberg. You might also look at sourcehut, too. It is a bit more... baroque? All its features are supposed to work without JS (it works well with Nyxt, for instance), and it is 100% free software.

Github alternatives

Posted Jun 12, 2025 14:05 UTC (Thu) by liw (subscriber, #6379) [Link]

While we're listing alternatives: I work on, and use, Radicle, which is distributed, is fully free and open source software. There's also Tangled, another take on distributed, but built on ATproto, which underlies Bluesky. (For personal reasons, I only care about distributed systems for this.)