subversion-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Stefan Sperling <s...@elego.de>
Subject Re: subversion development pace
Date Tue, 04 Oct 2016 08:51:44 GMT
On Sat, Oct 01, 2016 at 03:41:18PM +0000, David Novo wrote:
> Hello,
> 
> We have been long time subversion users but there are some people looking to
> investigate other solutions. One of the comments being made internally is
> that SVN development has slowed down to a crawl and that innovative features
> like new backends, shelving etc are simply not happening.
> 
> Looking at http://mail-archives.apache.org/mod_mbox/subversion-commits/, it
> does seem there is a dramatic downturn in the number of commits to SVN.
> However, I am not sure if we are interpreting those numbers correctly or not.
> Is there anyone who can shed some insight as to the current pace/priorities
> of SVN development.
> 
> For example, on the roadmap in the "most wanted" features (many of which
> would be fantastic for us), many of the "most wanted" features in the issue
> tracker have not even received an update in years. How am I to interpret
> this? Is work going on but not tracked in the issue tracker?
> 
> 
> -Dave
> 
> ----------------------------
> David Novo
> President
> De Novo Software
> david.novo@denovosoftware.com
> 

Hi Dave,

Up to last year, 4 developers had been employed full time by Wandisco.
Wandisco has since changed its strategy and has stopped funding SVN
development entirely. As a result, all those developers have changed jobs.
Some development efforts (such as my work on improved tree conflict handling)
are still being funded by other, smaller, companies, such as VisualSVN and
elego. And I believe CollabNet is still funding one developer part-time.

The SVN project has always been enjoying a lot of backing by companies.
As far as I know, there had always been several full-time positions for SVN
development over the past 15 years. That is very unusual for most open source
projects. Only the most successful open source projects enjoy this level of
funding. It also explains our relatively high commit rate.

I believe, even though we're not quite there yet, that SVN is now heading
towards being a mostly volunteer-driven project.
We have always had volunteers who help out where they can, but they can only
do so much. SVN is a mature project with a huge install base, which means that
making changes to SVN is a lot more difficult today than it was 15 years ago
when SVN was a young project with lots of room for new ideas.

The good news is that SVN is not going to go away just because some companies
change their strategy. SVN is a project of the Apache Software Foundation (ASF),
because our developers wanted it to be, because we realized that individual
companies are not equipped to drive our project forever. The ASF provides
everything we need to run the project. As long as there are active contributors,
the project will go on, and anyone can participate on equal terms.
This means that your (or any other) organization is going to get as much out
of the SVN project as they put into it. Big ticket items like you're asking
for above involve more work than I believe even a very dedicated volunteer
would be willing to invest into SVN today. But if you're willing to make
things happen, you can do so. The community will be very welcoming to any
contribution, be it funding or new developers who join the project to make
a difference.

Another other option is to wait for someone else to step in on your behalf.
Or to switch away from SVN to something else. That's also OK. We don't benefit
much from users who ask us to work on new features without giving anything back
to the project. There's no great loss to our community if those people go
elsewhere. We have a lot of very happy and active users who we can support,
and in any case we will be around for a good while to do that.

FYI, next week, most of our active developers will meet at elego's offices
in Berlin to work on SVN together in the same room, rather than across the
internet, for one week. This event will be attended by about 10 developers.
In my opinion, as long as we can put events like this together successfully,
the project is heathly.

Stefan

Mime
View raw message