Re: gEDA-user: polygon regression in pcb+gl

[Colin D Bennett <colin@xxxxxxxxxxx>]

On Fri, 18 Feb 2011 19:12:36 +0100
Kai-Martin Knaak <knaak@xxxxxxxxxxxxxxxxxxx> wrote:

> Colin D Bennett wrote:
> 
> >  The git history is a litte muddled...
> > 
> > When you say it's a little muddled, do you think that if it wasn't
> > rebased constantly that the history would be clearer?
> 
> Probably. But I am not the one who has to deal with the the
> development tree. 

True!  I was just wondering if you ran into the same issues I just
found with rebasing.

> > It also makes it nearly impossible to maintain your own
> > branch based off the pcb+gl branch since it will constantly be
> > diverging.
> 
> On the other hand, it lets the branch track changes of the main
> trunk. It offers the improvements in main when I use Peters version.

Well, rebasing is only one way to get changes from the mainline
branch (git HEAD).  You can also "merge" from the parent (HEAD), and
for any published branches, merging is generally recommended [1].
Merging doesn't rewrite history like rebasing does. However, if it
makes life easier for Peter C to use _rebasing_ rather than _merging_,
that's great.  I just have been maintaining my own trivial branch
(2-line patch that still hasn't made it into git HEAD *grumble*
https://bugs.launchpad.net/pcb/+bug/699498) based off the pcb+gl
branch, but that made the downside of rebasing evident.

Anyway, nothing against Peter C's work or his choice of methods to
manage his branch.  It's all tradeoffs.  I think rebasing makes some
things quicker but in the long run is messier since you lose track of
the exact changes you originally made.

That's all I have to say except to end with thanks to Peter Clifton for
creating and maintaining the pcb+gl branch, and I hope it goes to
mainline soon.

[1] "The perils of rebasing"
<http://progit.org/book/ch3-6.html#the_perils_of_rebasing>.

Regards,
Colin


_______________________________________________
geda-user mailing list
geda-user@xxxxxxxxxxxxxx
http://www.seul.org/cgi-bin/mailman/listinfo/geda-user