Re: NetBSD & Fossil

[bch <brad.harder%gmail.com@localhost>]

On 7/1/15, Greg Troxel <gdt%ir.bbn.com@localhost> wrote:
>
> Gour <gour%atmarama.net@localhost> writes:
>
>> Personally, I feel a bit strange to learn CVS in 2015 considering I was
>> not
>> using when it was relevant VCS - I quickly jumped to Darcs and then was
>> using/trying {arch, bzr, hg, mtn, fossil, git} and now will probably try
>> to
>> help EdgeBSD (NetBSD clone) which embraced git…
>
> I know you may find it strange, but NetBSD has been using CVS since
> ~~1993 with no disruptions, and a number of the things you list have come
> and gone.  It's really only hg and git which have the critical mass to
> ensure long-term survival, so it's a good thing NetBSD didn't switch to
> arch or bzr.  It is overdue to move to a DVCS

It probably will, and I hope it will. I've been involved (I was the
one that proposed fossil all those years ago, and introduced drh to
Jorg, and (iirc) got Jorg involved w/ helping Tcl convert from CVS ->
fossil (resulting in his excellent conversion tool)...

That said: I think NetBSD exhibits a stoic pragmatism that precludes a
conversion being "overdue" -- NetBSD will move when NetBSD is ready; I
was present for another notable conversion from CVS to $new_scm and
the project maintainer noted that for all CVS's "flaws", it's battle
hardened like nothing else, and has a process built up around it
precisely because it's been around for so long.

I'm not saying that NetBSD should drag it's feet, or there are not
benefits to be had by moving, but (like Greg hinted at w/ noting arch
and bzr, and the comment re: stability below) it's not a "no brainer"
decision.

Personally, I hope that the sticking  points with fossil are overcome
(either by technical fixes, procedure, or acceptance) and it's
adopted; the interface and workflow are pretty desirable.

-bch


> but the clamor of "we
> have to switch to X because it's new and shiny" has to be fended off for
> a while in order to have stability.