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

Re: Shredding private/confidential information?

From: Les Mikesell <lesmikesell_at_gmail.com>
Date: Tue, 28 Apr 2009 10:05:41 -0500

Stefan Sperling wrote:
> On Tue, Apr 28, 2009 at 07:45:54AM -0500, Les Mikesell wrote:
>> There are many situations where the extra pristine copies are
>> inconvenient, particularly when working with tools that like to do
>> recursive things
> In case you didn't know, wc-ng is the default on trunk now.
> Please try it out and help by providing feedback or even patches
> if you are so inclined. The more help we get, the faster it will
> be possible to have a single working copy meta-data store in your
> home directory, for all your working copies. That will solve this
> problem.

This still doesn't cover the scenario where you don't what the duplicate
copies at all. For example, I'd like a sane way to version-control the
whole /etc of unix-like systems, treating similar systems as branches
and there's not really an associated home directory for this.

>> - and as CVS demonstrates they are unnecessary.
> With CVS, people complained that they had to talk to the server
> all the time. And now people complain that they don't have to
> talk to the server all the time -- *sigh* :)

The SVN people say you should have loads of extra disk space to waste.
I say you should have a working network and not need all that
duplication. If "talking to a server" is a problem, that's a problem
that should be fixed.

>> I'd like to see an option for SVN to not use them. If you are on a
>> reasonable network the bandwidth savings from sending diffs in commits
>> won't matter (and you may have workspaces where you don't make any
>> changes anyway) and the concept of offline work is meaningless.
> Well, a centralised meta data store will hopefully address this issue
> a bit, because text-bases will be stored in a single, controllable
> location, and can be shared by multiple working copies. And maybe, maybe,
> it will be easy to just store text-base checksums instead of entire
> copies of files in the sqlite DB? Who knows.

Why do you need additional copies or even checksums of something the
server can reproduce on demand (that being its real purpose...)?

   Les Mikesell
To unsubscribe from this discussion, e-mail: [users-unsubscribe_at_subversion.tigris.org].
Received on 2009-04-28 17:06:51 CEST

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