Lunzer, I may share the Fossil comments, as I'm an avid user of it. Paraphrasing Conway's Law[1] culture and infrastructure reflect each other and I think that Git reflect the bureaucracy of Linux Kernel development with its fork and PR by default, while Fossil considers a small group of developers who mostly know each other [2] and has a more lean/agile approach.
[1] https://en.wikipedia.org/wiki/Conway%27s_law [2] https://fossil-scm.org/fossil/doc/bsd-vs-gpl/www/fossil-v-git.wiki I program mostly solo projects and when I hopefully I will pass to projects with few well trusted developers. Seeing from the quality of SQLite and Linux, PR's presence or absence are not a warranty over code quality, but for sure PRs are sign of the believe in quality through bureaucracy and self-restrain. Of course commit message as XKCD are pretty useless (and funny ;-)), but in my case they (+diff) have been working kind of well. I have seen similar behavior on non solo projects like Fossil and SQLite. But I'm not an active Leo code contributor. So I was just giving my opinion but in the end, core contributors should choose what works best for the developers. Cheers, Offray On 21/08/20 3:01 p. m., [email protected] wrote: > @offay, I've seen similar comments on the Fossil forums. > > I don't have faith in developers to write "good commit messages". You > need only see this comic to understand my feelings: > https://xkcd.com/1296/ . Developers (in general) are lazy, and this is > not entirely caused by "laziness", but these days more often due to > lack of established best practices and lack of time. Developers will > perform the least amount of steps to get code into production. PRs, > while being bureaucratic, put a hard stop in front of developers which > forces them to think much harder about their proposed changes and how > they will be used and perceived by others. While this slows down > development, even in small teams, it is a net win for code quality. > > My biggest complaint with PRs with git is that the PRs and not wholly > encapsulated within the repo. This is bad for privacy, bad for > custody, and bad for archivability. But better documentation and more > deliberate contributions are worth these trade-offs, if a better > system comes around I'm always interested. > > On Friday, August 21, 2020 at 2:14:02 PM UTC-4 [email protected] wrote: > > > On 17/08/20 11:28 a. m., Edward K. Ream wrote: > > The days of cowboy commits are coming to an end. > > > > In future, I plan to create a PR for all my work. A PR is a good > > record of what has been done, and it should help prevent unwanted > > merge conflicts. > > > > I think separate PR's for all work makes sense for all of Leo's > devs. > > What do you think? > > > I dislike them. I think they introduce an unnecessary bureaucracy in > most projects with small/solo developers and that good commit > messages + > actual diffs can be good enough for most project as commit > documentation. But of course, each project and its developer > community > have different styles and ways to work together. > > Cheers, > > Offray > > -- > You received this message because you are subscribed to the Google > Groups "leo-editor" group. > To unsubscribe from this group and stop receiving emails from it, send > an email to [email protected] > <mailto:[email protected]>. > To view this discussion on the web visit > https://groups.google.com/d/msgid/leo-editor/e3fcfb9e-c9fc-4795-a8bc-a59add33207fn%40googlegroups.com > <https://groups.google.com/d/msgid/leo-editor/e3fcfb9e-c9fc-4795-a8bc-a59add33207fn%40googlegroups.com?utm_medium=email&utm_source=footer>. -- You received this message because you are subscribed to the Google Groups "leo-editor" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/leo-editor/eaac3f26-9fba-b964-de46-181dc772233c%40riseup.net.
