[
Lists Home |
Date Index |
Thread Index
]
We've learned certain best practices. The tough bit
are the contexts of application and elicitation.
What makes perfect sense for the network isn't good
enough at the application layer.
Start here:
1. Parsimony: Dare to do less. (Tim Bray)
This one never fails.
See Grice's Maxims for a more complete methodology.
2. Complicated things that work are
usually made of simple things that worked.
Follow-on to Parsimony. See Extreme Programming.
The problem is lots of simple things work but not
all of them aggregate. Choose wisely.
3. Be strict in what you send and lax in what
you accept.
A good idea but it makes you the garbage collector
and you are always searching for signal. Learn to
say no and keep records of that decision. The
affordance of discovery is not a commitment to
negotiate.
Ok, it's aphorism thread time. Anyone care to add ones
you assert apply anywhere in the system rather than work
at one layer and fail at the next? I won't be surprised
if the only one left standing is Maxim 1 above followed
by "Nothing is certain." aka, the empty set is the
universal set.
len
From: Andrew S. Townley [mailto:andrew.townley@bearingpoint.com]
I'm not sure what the best solution is. I think we're a bit out
numbered...
|
