Post Ab1x3wV3VTv2LLKaQK by penguin42@mastodon.org.uk
(DIR) More posts by penguin42@mastodon.org.uk
(DIR) Post #Ab1lqUhyQQI6t9l4HQ by stsquad@mastodon.org.uk
2023-10-22T11:50:16Z
0 likes, 0 repeats
It seems the topic of #email vs #forge based workflows in #floss software is a topical subject so I thought it an opportunity to generate a blog post: https://www.bennee.com/~alex/blog/2023/10/22/comparing-forge-based-and-email-based-workflow-for-open-source-projects/#comparing-forge-based-and-email-based-workflow-for-open-source-projectsYou can comment by replying to this #mastodon thread.
(DIR) Post #Ab1lqVWJPGYbPHJGk4 by penguin42@mastodon.org.uk
2023-10-22T12:01:17Z
0 likes, 0 repeats
@stsquad I think some of the cons of the email based systems are that you can easily lose a patch with no one reviewing or merging it; most of the forge based systems are better at tracking patches in flight. The forge based systems should be doing line-by-line review most do don't they?
(DIR) Post #Ab1mPKUAQOgJJpuXdw by stsquad@mastodon.org.uk
2023-10-22T12:07:35Z
0 likes, 0 repeats
@penguin42 yeah this is were tooling helps. #patchew helps tame the mess of searching your email feed to find series missing reviews. When I talk about line by line I mean at the commit level. In most forges I end up reviewing stuff on the PR diff view which is fine if PRs are small and self contained. We do suffer somewhat in #qemu by having some mega-cleanup series which is a lot to wade through.
(DIR) Post #Ab1x3wV3VTv2LLKaQK by penguin42@mastodon.org.uk
2023-10-22T14:06:59Z
0 likes, 0 repeats
@stsquad Yes, the forges seem to be quite opinionated about whether you should be working at PR or commit level; which is a shame since it varies by task. Geritt is quite nice at the individual commit level but has very little at a PR layer. I find Gitlab the opposite - flexibility would be nice. I didn't find patchew helped me that much.