[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: SVN Obliterate.

From: Karl Fogel <kfogel_at_red-bean.com>
Date: Sun, 17 Feb 2008 20:18:37 -0500

Gavin 'Beau' Baumanis <gavin_at_thespidernet.com> writes:
> I have been doing a little reading of what I can find out about the svn
> obliterate proposal.
> And subsequently have a few questions, please.
>
> ( And of course, if this has already been started - I have obviously
> missed it, could someone please point me in the right direction?)
>
> Is the obliterate functionality supported / required by the community?

A lot of people seem to want it. On the other hand, none of those who
fund Subversion development have found 'svn obliterate' sufficiently
compelling to give it the funding it would need. (Of course, a lot of
work here is unfunded, or not directly funded, too. So please don't
take this to mean that all features require funding -- that's
certainly not the case.)

> Is it supported by those performing the development of SVN?

Yes, in theory: we'd all like to see it happen, but we have other
priorities right now, mainly the 1.5 release.

> Is there any anticipated start date with respect to starting the
> required work?

No.

> There seems to be a lot of 1/2 started discussions - but no "official" one.

That's precisely the problem, yes :-).

> Subsequently, (if the answers to my above questions are yes), what is the
> process for starting things off?
>
> I assume we would start with asking for use cases and ensuring that
> we have all use cases discussed for acceptance / rejection. The
> gathering and vetting of use cases alone is going to be a fairly
> large undertaking - let alone any "actual" design / development work
> being completed - or even started for that matter.
>
> While I can't think of any off the top of my head - I am certain there are
> going to be many use cases that can be solved by better education of what is
> available as opposed to the introduction of any new feature.
>
> Lastly, in an effort to be as pro-active as possible and to limit
> the effort of people already involved - I will most happily and
> enthusiastically take on the task of "Use Case Coordinator", for
> want of a better term.

Thank you for recognizing that this is what is needed. Be careful,
though: a lot of people have been down this path before you, but were
not specific enough to be useful. They say things like "obliterate
should let you remove a path", without specifying the behavior any
more precisely than that. Each use case will lead to a proposed
Subversion behavior, and we need to know *exactly* how that behavior
would work. For example, how does 'svn log' on the repository look
before and after the obliteration of that path? Is the data removed,
or merely unlinked? What about copies? Etc, etc.

If you take this on, be prepared to spend 90% of your time asking
people to be more specific. That's what it's going to take. That's
more or less what I've been trying to say in these comments...

   http://subversion.tigris.org/issues/show_bug.cgi?id=516#desc17
   http://subversion.tigris.org/issues/show_bug.cgi?id=516#desc19
   http://subversion.tigris.org/issues/show_bug.cgi?id=516#desc71

...especially the last one.

In issue #516, there are some interesting ideas about how to fund the
work, and one useful step might be for you to set up a pledge drive at
fundable.org. Set a high number, like $50,000 US or something, and
see if you get pledges meeting that. If you do, that might attract a
developer to come along and drive this to completion.

> I don't want to step on any toes and sincerely welcome any required
> education!

It sounds like you've got the right intentions :-), just be prepared
for the process to require a lot of attention to detail.

We'd all love to see this get done, clearly a lot of users want it
(but my priority right now is the 1.5 release, so I'm not volunteering
for issue #516). The earliest 'obliterate' could get into Subversion
would be the 1.6 release. You may wish to wait until 1.5 is out, or
at least in release candidate stage, before starting this discussion.

-Karl

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe_at_subversion.tigris.org
For additional commands, e-mail: dev-help_at_subversion.tigris.org
Received on 2008-02-18 02:18:49 CET

This is an archived mail posted to the Subversion Dev mailing list.