On Tue, 2005-06-07 at 10:44 -0700, Daniel Rall wrote:
> Most organizations and system administrators are very interested in
> statistics, as they provide a measuring stick by which process or
> deployment efficiency gains can be assessed.
Certainly, at work my managers are often interested in usage data about
what people are doing with the computers I help administer. But we
don't expect the software we use to collect that information for us, and
it's rare for a piece of open source software to provide that kind of
client-side logging.
Adopting the philosophy behind Karl's and Ben's proposal also raises the
question of whether we should attempt to have the server log purely
local operations, like a revert or a diff against the text base.
Organizations might be very interested in logging that stuff. But I
don't think we want to go in that direction, and similarly, I don't
think we really want to go in the direction of logging whether the
client is doing a checkout vs. an update from a working copy which
happens to be empty.
> With the client supplying the conceptual operation, this is true for a
> security-sensitive audit.
If administrators see logs on the server, many of them will assume that
the server ensures that those logs are reliable and correct. The idea
that they could be just a data collection repository for client-provided
information probably won't cross their minds.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Tue Jun 7 21:35:25 2005