[
Lists Home |
Date Index |
Thread Index
]
Tim Bray wrote:
> I'm a computer programmer.
With all due respect, as regards the salient point of this discussion: no,
you're not. You are clearly speaking here with the biases and reflexes of a
specification author, not a programmer. To satisfy the programmer, we have the
operational definition of URLs in rough consensus (of the largest number of
implementors of any IT technology) and running code (performing the greatest
number of operations per day). Precisely what you are looking to do is ignore
that enormous base of implementation experience and empirically observable
processing behavior, in favor of an evanescent abstraction.
> I like to have normative specifications.
Exactly. Joe and Simon offer the alternative of replicatable implementation
experience.
Respectfully,
Walter Perry
|
