In article <199509261642.MAA12620@remus.rutgers.edu>
kenn@remus.rutgers.edu (Ken Nakata) writes:

   Even if the person who believes that most sun-lamp problems are due to
   software bugs is "the" Chris Demetriou himself?

There's nothing particularly special about Chris, as opposed to any
other contributor.  He is not God.

The fact is, it's pretty much useless to talk about whether it's a
hardware or software bug, unless you're actually debugging it.  I've
only had access to the machine's console for about 20 minutes since
the project began, so I've never been able to debug it, and Chris has
not done it himself.  Based on the symptoms I'm seeing now (completely
random memory corruption), however, I believe it's quite likely a
hardware problem.

Given the limited access to lamp, it seems our only option is to just
move the services off of it.  We're working on that.