[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: David Weintraub <qazwart_at_gmail.com>
Date: Tue, 28 Apr 2009 11:54:51 -0400

On Tue, Apr 28, 2009 at 11:05 AM, Les Mikesell <lesmikesell_at_gmail.com> wrote:
>>> - 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.

It's always been a question of whether you bet on the network or local
storage. ClearCase, for example, bet heavily on the network. Many of
us had diskless workstations, so we didn't even have storage.
ClearCase's working copies were stored in "views" that were usually
kept on a "view server". The advantage is that you could go to any
machine and see your view and do your work. Plus, that extra fast
10Mbps Ethernet and server was certainly a lot faster than that poky
processor on my machine.

The best feature about ClearCase was its ability to search for already
built objects that it would "winkin" into your view. That saved so
much time and effort. Why build something when it was already built?

Now, I have a 2.1 dual gigabyte processor sitting on my desktop with
120 Gigabytes of space. Nothing too exceptional these days. The big
problem is the slow network. Back when Atria was ClearCase's company,
the only traffic on my network was email. Now, gigabytes of data
passes back and forth every minute.

Even ClearCase's "winkin" feature is useless. My system can build any
object way before ClearCase can locate it in its database and transfer
it to my view.

Subversion was designed with this modern architecture in mind. We have
access to massive amounts of diskspace and fast processors while our
servers can be bogged down with traffic. Why not put all of that disk
space to good use and reduce the amount of traffic on the network?
Heck, many users use their laptops and they don't always have network
access. Subversion is structured to allow these users to do their
diffs and reverts and check to see what has been changed without
worrying about your network connection.

However, environments keep changing, and Subversion's design may soon
become as archaic as ClearCase's original design. Look at Netbooks
with minimal diskspace. Look at the ubiquitous network. Every donut
place and coffee shop has free WiFi connectivity. With 3G and 4G
networks, you have access to your repository from almost anywhere.

Maybe the next great revision control system will rely on this
universal network access and less on local diskspace. Maybe the next
repository storage medium will be "cloud based". However, for right
now, I believe the designers of Subversion have hit a good compromise
between disk space usage and network usage.

David Weintraub
To unsubscribe from this discussion, e-mail: [users-unsubscribe_at_subversion.tigris.org].
Received on 2009-04-28 17:55:41 CEST

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