I think I agree with Branko on this one.
Certainly, it is true that if we implement random access on old files
in a groovy way, then you can point your Acrobat Reader (or whatever)
at an old version of a huge doc and it will be retrieved efficiently.
But if we don't spend the time on the grooviness right now, you will
still be able to point your Acrobat Reader and read the old doc, it
just won't be retrieved as quickly.
There's nothing preventing us making the change later, if it turns out
to be needed. It would be premature optimization to spend a lot of
time on it now (unless Jim thinks it's not any significant extra work;
but it sounds like it is, so I'm going on that assumption). Let's
just get things working first.
-K
Greg Stein <gstein@lyra.org> writes:
> On Tue, Nov 14, 2000 at 01:52:16PM -0500, Greg Hudson wrote:
> > > Acrobat Reader. Today. I'd suggest the time is right :-)
> >
> > Let's have a little perspective here. Being able to efficiently run
> > Adobe Acrobat Reader directly against versions of files in the
> > repository is a good thing, but it's not a goal for Subversion 1.0.
>
> Perspective? I gave a concrete example, not an exhaustive list.
>
> And I hope that people *have* realized that a person can just point their
> web browser at a Subversion repository and see the HEAD branch. Hell, you
> can apply version control to your web site, and users can continue to just
> browse through it like they always do. And that browsing could certainly
> include the storage the PDF documents.
>
> Cheers,
> -g
>
> --
> Greg Stein, http://www.lyra.org/
Received on Sat Oct 21 14:36:14 2006