On Dec 29, 2006, at 9:08 PM, Sam Hartman wrote:
ideology is a fine reason in my mind to fail to have consensus.
Hmm. I have to think for a moment on the definition of ideology.
I'm thinking first of a particular discussion we have going on in
tsvwg. Some are asserting that where the law of large numbers
applies, one can detect the onset of congestion by measuring the
behavior of traffic around a guard band. I am asserting that where
the law of large numbers does not apply, the behavior of the kinds of
traffic in question don't lend themselves to that technique and
something more like IntServ is appropriate. Others are stating that
they believe that they believe that external measurements of traffic
behavior in a guard band is adequate even in domains where there is
only capacity for a single digit number of simultaneous sessions,
such as BRI ISDN. I have a statement in an internet draft that says
that intserv signaling is necessary on access links, and did not
agree to "bless" guard band signaling on 64 KBPS links because I
think that is just plain wrong.
Is my argument "I won't put it in because I think it's demonstrably
incorrect" an example of "ideology"? If so, then I agree with you.
The general usage of the term (http://en.wikipedia.org/wiki/Ideology
gives an article, the Online Dictionary gives two definitions: "The
body of ideas reflecting the social needs and aspirations of an
individual, group, class, or culture," or "A set of doctrines or
beliefs that form the basis of a political, economic, or other
system") makes it less about engineering correctness than about
conformance to a social viewpoint. If you're discussing the latter,
then I have a quibble.
Which did you mean?
_______________________________________________
Ietf@xxxxxxxx
https://www1.ietf.org/mailman/listinfo/ietf