Re: irt: Re: Core statement on version control systems

[Andrew Cagney <andrew.cagney%gmail.com@localhost>]

On Wed, 3 Dec 2025 at 17:55, Greg A. Woods <woods%planix.ca@localhost> wrote:
>
> Again, thanks very much for writing up your thoughts!
>
> At Sat, 29 Nov 2025 23:18:04 +0000, David Holland <dholland-tech%netbsd.org@localhost> wrote:
> Subject: Re: irt: Re: Core statement on version control systems
> >
> > Pull requests are, though, a terrible vehicle for doing real work:
>
> Perhaps the guides written for the FreeBSD conversion to git will help
> offer a view that can simplify some things.
>
> https://docs.freebsd.org/en/articles/committers-guide/
>
> It's obviously not simple though.

All I can see are trees, am I in a forest?
For instance:

> Then you should have separated fetch and push URLs as the most efficient setup:

they use a public readonly https endpoint when pulling, but then
git+authentication when pushing; but why?
The comment hints at performance.  But I wonder if it is because it
lets one pull down updates without needing to go through
authentication.

There must be a big picture work flow somewhere, I'm just not seeing it.
I also wonder if some of the operations stem from SVN which would
happily re-order history when it came to pushing changes - some
command sequences mimic that.

Anway, the first step is core needs to decide a general workflow.  And
hopefully it can be implemented with a few primitives.