On Tue, Mar 11, 2008 at 8:26 PM, Garrett Rooney
> On Tue, Mar 11, 2008 at 11:54 AM, Erik Huelsmann <ehuels_at_gmail.com> wrote:
> > Well, as we already have this with wc-1.0, I felt it would be too
> > restrictive to require it upfront. Unfortunately an older version of
> > this document had a sentence stating "Not a goal per se, but nice if
> > we can achieve them none the less."; that sentence seems to have been
> > lost.
> I'm not sure what you mean by that first sentence. How does the fact
> that we do have the feature in 1.0 imply that it would be too
> restrictive to require it in the new version?
It has always had its quirks, AFAIR, but the fact that we have or do
not have it in 1.0 doesn't make it too restrictive (or not) for wc-ng.
What does make it too restrictive is that we have a long list of items
which *have* to be met in order to actually improve from wc-1.0.
Making all the requirements I listed actually work is a far cry from
where we stand with our wc-1.0 code. So, in order to try to reduce the
number of hard requirements, I introduced a number of soft
> Do you mean to say that
> we don't have it in 1.0, thus we shouldn't require it in a new
> version? That seems incorrect, AFAIK we do have it in 1.0, unless you
> mean something different by "shared working copy" than I do...
Well, my point is: If we're actually going to improve upon wc-1.0,
then maybe we shouldn't start out by requiring all that ended up in
wc-1.0 (by design or by accident). Maybe wc-ng is a good time (and it
is, IMO) to think about what it actually is that the WC *needs* to
> > If we can create a design which fills in all the requirements *and*
> > can satisfy the 'multiple users per WC' requirement, I'm all for it!
> > If you don't mind me asking: do you have any recollection of which
> > committers that might have been?
> > I hope this answers your question(s).
> I believe Justin and Sander put some effort into making it work so
> that we could use shared working copies to serve up web content for
> apache.org. You need to allow multiple people to cd into the
> directories and update them.
On unix, this should be possible with the right settings on the
wc-root (like 'sticky bit'), shouldn't it? I mean with a wrapper which
sets the right umask and the right access rights, it *should* be
I'm not going to create a design which actually is going to prevent
this from happening, but I don't think it should be the primary
concern of the design... Do you?
To unsubscribe, e-mail: dev-unsubscribe_at_subversion.tigris.org
For additional commands, e-mail: dev-help_at_subversion.tigris.org
Received on 2008-03-11 22:26:49 CET