Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 21 Jul 2016 11:48:27 -0500
From:      Pedro Giffuni <pfg@FreeBSD.org>
To:        chat@FreeBSD.org
Subject:   What Larry McVoy (bitkeeper) got wrong ....
Message-ID:  <44e4450e-8dbb-f401-bd5c-df503f7a4ef8@FreeBSD.org>

next in thread | raw e-mail | index | archive | help
Hello;

I was in the process of preparing a port of bitkeeper and I found this:

https://github.com/bitkeeper-scm/bitkeeper

"The BitKeeper history needs to be written up but the short version is 
that it happened because Larry wanted to help Linux not turn into a 
bunch of splintered factions like 386BSD, FreeBSD, NetBSD, OpenBSD, 
DragonFlyBSD, etc. He saw that the problem was one of tooling. ..."


This actually made me laugh. Wrong problem and wrong tool to "fix"
it.

While CVS could have been seen as the wrong tool for just about
anything, FreeBSD could have used perforce for "free" and still we
would have a similar situation. The underlying issues IMHO was that
of community affinity and developer interests, not of tooling.

Back in the early days, the "splintering" of the BSDs might have been 
seen as an issue but as I see it now, it has been a blessing. I for
one have no interest in adopting the OpenBSD approach towards security,
to name just something some developers see as interesting, but I do see
the idea of different teams trying new things critical, and I do take a
lot of code from the other BSDs.

Version control, and in particular *distributed* version control,
actually makes forking easier, but curiously some of the BSDs have found 
it very difficult to get away from CVS.

I guess we should be thankful about the new tooling but if the idea was 
to prevent linux to become like the BSDs (for good or bad), I think the
linux community is to credit for that.

Just my $0.02,

Pedro.



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?44e4450e-8dbb-f401-bd5c-df503f7a4ef8>