> You were either very lucky or very careful; most CVS repos of any
> age have a lot more cruft in them than this.
:-) It's probably due to my aversion to branches, at least with CVS...
> What happened to the history before 2000?
Honestly, I don't know; see my other e-mail. Looking up the groff
mailing list archive, I see that for a short time of less than a year
there must have been a CVS archive at ffii.org (in Munich), but the
data is no longer available, I fear – I've written an e-mail anyway to
a person involved, so maybe I'm wrong and there is hope.
However, the last entry before that was
Fri Aug 15 08:51:47 1997 Eric S. Raymond <e...@snark.thyrsus.com>
* README, PROJECTS, NEWS, INSTALL, VERSION,
doc/Makefile. doc/pic.ms, groff/groff.man:
Prepare for 1.11 release. No code changes.
Documentation for pic added (doc/pic.ms).
Did you have access to the CVS then?
> Was the code really not under version control until that late?
I guess that the CVS or RCS archive was private to James. Hopefully,
he reponds (he *never* did to any groff question, but I know that the
e-mail address is correct and recent, since he contacted the harfbuzz
list).
> The only step not done is massaging multiline commit comments so
> they obey git conventions. Normally I do this, but given the
> particular conventions GNU projects use it's not clear I should.
>
> Here's the problem. [...]
Yes, I fully agree that the commit style I used was far from perfect,
committing a lot of stuff with a single commit which I would today
split into multiple, smaller commits.
> I'll look at this some more to see what percentage of cases I cab
> salvage, but I think we have to accept that the view of older
> sections of the history in gitk isn't going to be very pretty.
Yeah. IMHO, we have to live with that, and investing more time fix
log entries before 2009 or 2010 is probably a waste.
Werner
