[Reportbug-maint] Merits of VCS tools and workflows

Ben Finney ben+debian at benfinney.id.au
Thu Jul 3 14:33:08 UTC 2008


"Sandro Tosi" <matrixhasu at gmail.com> writes:

> On Thu, Jul 3, 2008 at 15:52, Ben Finney <ben+debian at benfinney.id.au> wrote:
> > I've said several times now that I would welcome *any* modern VCS
> > that doesn't have the problems of Subversion.
> 
> and I'm saying that you're not using svn because you already know
> bzr and you don't want to learn svn, period.

On what do you base that assertion? I've already told you it's wrong,
and entirely contradictory with what I have said.

> > If you are going to wilfully misrepresent my position, this
> > discussion isn't really going anywhere.
> 
> Me? I've asked a simple question, and you didn't answer it... what
> were your negative experiences with svn+branches?

A merge in Subversion has no automatic conflict resolution. Modern VCS
tools have this feature.

A merge in Subversion has no way of knowing whether a change being
merged in has already been applied. Modern VCS tools will recognise
already-applied changes without fuss.

A merge in subversion is *slow*, many times slower than with any
modern VCS I've used.

Any operation on branches in Subversion affects the central
repository, and thus all users of the repository. Modern distributed
VCSes allow creation of branches that only exist locally, making
branching and merging cheap, quick, and unobtrusive.

That's some of the issues; there are others.

These are facts of my experience with Subversion. Please stop acting
as though they don't exist.

-- 
 \       “[Freedom of speech] isn't something somebody else gives you. |
  `\      That's something you give to yourself.” —_Hocus Pocus_, Kurt |
_o__)                                                         Vonnegut |
Ben Finney




More information about the Reportbug-maint mailing list