You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@subversion.apache.org by Tom Lord <lo...@regexps.com> on 2002/10/12 19:15:34 UTC
another application
Another application for svn is as an alternative implementation of
`arch' "revision libraries". A revision library is a repository-like
entity that's optimized for individual file access -- svn is a good
fit making different trade-offs from the current approach.
I'm not trying to say, with these `arch' examples, that arch should
somehow "subsume" svn -- I'm just pointing out that there's enough
immediately useful functionality right there, already, to stabilize,
declare 1.0, and thus hopefully increase the user community size --
not worrying particularly about issues that feedback from the user
community can then drive (and help with):
*) the cli and the future of client ui's.
*) performance tuning / scaling
*) convenience features for admins
The priority of 1.0 doesn't _have_ to be capturing the hearts and
minds of people looking for the perfect free rev ctl system for
software development -- it's ready to go for the kinds of back-end
apps I'm pointing out and I think the CVS-like-client for end-users is
a goal worth reconsidering, perhaps taking distributed rev ctl. into
account.
-t
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: another application
Posted by Gustavo Niemeyer <ni...@conectiva.com>.
> I have thought about something that I've never seen mentionned here
> since I read this ML : I think SVN will really gain popularity the day
> "public" svn repositories are available on the net. There are dozens
> of community websites, ranging from dinosaur SF.Net to small french
> TuxFamily.Org.
[...]
> Now, why not ask the admins of those community website what they think
> about SVN, whether they plan to switch to another SCM, and if not,
> why, and what could be done to make them switch.
http://sourceforge.net/docman/display_doc.php?docid=768&group_id=1
----
Does SourceForge.net intend to provide Subversion revision control
services?
At this time, Subversion is still under heavy development and has not
yet reached the level of stability and maturity needed to provide
production-grade Subversion services from SourceForge.net. The
SourceForge.net team is continuing to evaluate the progress of the
Subversion development process and will consider providing Subversion
services once a suitable level of maturity has been reached.
--
Gustavo Niemeyer
[ 2AAC 7928 0FBF 0299 5EB5 60E2 2253 B29A 6664 3A0C ]
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: another application
Posted by Greg Stein <gs...@lyra.org>.
On Sun, Oct 13, 2002 at 08:41:48AM -0700, Tom Lord wrote:
>
> > I think SVN will really gain popularity the day "public" svn
> > repositories are available on the net.
>
> That's a very hard goal. I don't think it should be a 1.0 goal to
> prepare for that deployment. Here's some reasons why:
I believe that 1.0 is targeted for exactly that kind of high capacity
deployment. If people simply want a "workgroup size" capable version of
Subversion, then they can use an Alpha or a Beta release. It is more than
stable enough for those kinds of deployments (as evidenced on our new "who
is using SVN pages).
As I said before, if you have a bunch of ideas that you'd like to build on
top of SVN, there is nothing stopping you from doing so *today*. There isn't
much of a need to "wait" for a 1.0 release to begin your work.
Note that when we hit Beta, we'll probably branch Subversion at that point.
The branch will shoot for the 1.0 release, while the trunk will "open up" to
the post-1.0 work. The exact timing of the branch will be up for grabs, but
it probably should happen around then. The very notion of Beta, in our book,
implies serious stability. I can't see holding up a billion ideas while we
eke out the last teeny bugs. The balance, though, is distracting people with
the post-1.0 trunk, such that the beta bugs aren't removed. We'll see how it
goes.
Cheers,
-g
--
Greg Stein, http://www.lyra.org/
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: another application
Posted by Tom Lord <lo...@regexps.com>.
> I think SVN will really gain popularity the day "public" svn
> repositories are available on the net.
That's a very hard goal. I don't think it should be a 1.0 goal to
prepare for that deployment. Here's some reasons why:
1) performance
Big public repository hosts have controlled hardware costs and
heavy work-loads. Such deployments are among the strictest
performance requirements svn will _ever_ have to face. My strong
intuition from reading the dev list is that the architecture is up
to the challenge -- but the implementation will take longer. No
surprise there at all -- it's essentially a really exotic file
system implementation and those can take quite a while to properly
tune.
I say: go for a 1.0 that is usefully stable, then use the cachet
of 1.0 adopters to develop commercial demand for tuning and
scaling work. I think the wiki hack I described is an easy
tactical play and fits the bill. I can even think of who might
buy [*] a "development subscription" based on a 1.0 demo plus an
outline of how the tuning/extension project would be run.
[*] because it _really_ fits the business model of the company I
have in mind, plus it fits the pattern of how they have in fact
managed their costs and functionality over the years.
2) features
The goal you named (deployment on big public project hosts) takes
svn from a very small number of users to a huge number, almost
overnight. I think work on the client will suffer from the mix of
incentive-to-remain-upward-compatible plus
overwhelming-number-of-complaints.
(I've suggested elswhere to freeze the client, declare it a low
level interface, and factor client work out to other projects
post 1.0 --- I still suggest that, but it is orthogonal to the
point that, regardless of the path the client takes, a first
deployment to a large, demanding, and often rude peanut gallery
is likely to introduce paralysis.)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: another application
Posted by Greg Stein <gs...@lyra.org>.
On Sat, Oct 12, 2002 at 09:19:54PM +0200, Sebastien Cevey wrote:
>...
> either. Therefore, these big, well-known websites could be a very good
> "advertisement" for SVN.
>
> "Oh, look, they have switched to a new, possibly better SCM, let's try
> it !".
>
>
> Even future commercial uses of SVN in private companies might start
> from right here.
>
> So is there any "plans" about this issue ?
Not until we release a 1.0. If those sites want to pick up pre-1.0, then
more power to 'em. It might be interesting to do that around the Beta
timeframe. But most people seem to be concerned about coding SVN to reach
that state of "features and correctness" before really worrying about
publicity, adoption, and other similar activities.
Personally, I believe SVN will definitely follow the model of, "build it and
they will come." SVN just kind of creates a natural gravity well, and we'll
see people migrating. More and more, until one day, SVN is the natural
order, and CVS is the old beast.
Cheers,
-g
--
Greg Stein, http://www.lyra.org/
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: another application
Posted by Sebastien Cevey <se...@cine7.net>.
On Sat, Oct 12, 2002 at 12:15:34PM -0700, Tom Lord wrote:
> I'm just pointing out that there's enough immediately useful
> functionality right there, already, to stabilize, declare 1.0, and
> thus hopefully increase the user community size
I have thought about something that I've never seen mentionned here
since I read this ML : I think SVN will really gain popularity the day
"public" svn repositories are available on the net. There are dozens
of community websites, ranging from dinosaur SF.Net to small french
TuxFamily.Org.
They all provide disk space, mailing lists *and* CVS.
I can hardly imagine that SVN will truly be accepted as CVS successor
as long as most of these community websites keep using CVS. I think
they will, at least at the beginning, because, whatever you say, it
still involves some work to migrate to SVN (not to mention
SourceForge-sized examples, fortunately they are few).
Now, why not ask the admins of those community website what they think
about SVN, whether they plan to switch to another SCM, and if not,
why, and what could be done to make them switch.
Yes, these CVS repositories are only the tip of the iceberg, and there
are plenty of "private" repositories everywhere. But I can say that if
I hadn't used CVS on SF, maybe I wouldn't use CVS on my own
either. Therefore, these big, well-known websites could be a very good
"advertisement" for SVN.
"Oh, look, they have switched to a new, possibly better SCM, let's try
it !".
Even future commercial uses of SVN in private companies might start
from right here.
So is there any "plans" about this issue ?
--
Sebastien Cevey <se...@cine7.net>
Cine7 - www.cine7.net
Milcis - www.milcis.net
ICQ: 48895760
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org