Alva, this is again right on the mark. Even with high-level programming
languages like Lisp/Prolog/Mathematica/Perl we still need good data
structures and algorithms. Without the latter, correctness or
performance cannot be achieved. We need analogs of data structures and
algorithms for infrastructure: what are the logical structures that one
sets up (via configuration) when trying to satisfy top-level
requirements on functionality, security, performance, reliability. How
does one convince oneself that the resulting composition does satisfy
the requirements? Surely, these questions are of daily, pressing
importance to every system administrator. If not, it would be very
useful to hear why. For context, consider the single-point-of failure
example in my post two days ago. Regards. -- Sanjai
Alva Couch wrote:
I see the same pattern in this thread that I've always seen.
Tool builders think the tools will solve the problem, while no one else
is convinced. The reason for this seems to be that there is a third
thing, other than theory or practice, which is a factor: how the
system administrator conceptualizes the problem of configuration
management in his or her own head.
--
Sanjai Narain, Ph.D.
Senior Research Scientist
Information Assurance and Security Department
Telcordia Technologies, Inc.
1 Telcordia Drive, Room 1N-375
Piscataway, NJ 08854
732 699 2806 (T)
908 337 3636 (M)
[EMAIL PROTECTED]
_______________________________________________
lssconf-discuss mailing list
lssconf-discuss@inf.ed.ac.uk
http://lists.inf.ed.ac.uk/mailman/listinfo/lssconf-discuss
