Ben Collins-Sussman wrote:
> On 29 Jun 2006 16:21:09 +0200, Lars Gullik Bjønnes <email@example.com>
>> "Ben Collins-Sussman" <firstname.lastname@example.org> writes:
>> | On 6/28/06, email@example.com <firstname.lastname@example.org> wrote:
>> | > I have searched the mailing list archives, and have found
>> | > to be a popular topic - this must be because many people believe
>> | > Subversion is missing a simple tag feature.
>> | Yes, and the general consensus of the subversion developer community
>> | is that this a small (but vocal!) minority, and that they are wrong in
>> | thinking that 'svn cp trunk tags/version-1.0' isn't a "true tag".
>> Which of course is just a copy and there is nothing in subversion
>> itself that makes it immutable.
> So what? I think you missed my point. Tags in CVS aren't immutable
> either. In CVS, not only are tags changable, there's *no record* of
> anyone changing them. CVS tags are *more* dangerous than Subversion
> tags. So why do people want to imitate CVS?
I might have missed some relevant mails in this thread, but nothing in the
quoted bits here indicates that anyone is trying to imitate CVS.
But besides all that, I question the utility of arguing this point at all.
Some say Subversion has "true tags", some say it doesn't. Some say
immutability is a requirement for a tag, some say it isn't. And the end of
the day, consider whether or not Subversion meets your particular needs. If
not, can it be changed to meet your needs without violating its own design
goal? If yes, suggest/implement the change. If not, shut up about it an
move on. There are many (many, many, many...) folks for whom Subversion's
tree-copy functionality is filling the need for "humanly-named" snapshots.
If your need significantly exceeds theirs, Subversion might not be right for
C. Michael Pilato <email@example.com>
CollabNet <> www.collab.net <> Distributed Development On Demand
Received on Thu Jun 29 18:05:14 2006