[
Lists Home |
Date Index |
Thread Index
]
The hubris in that is in suggesting that a separate e-mail list
will succeed in putting to rest issues that have not ever rested
in the voluminous threads here and on other lists. Just how is
that supposed to work: a self-selected group sits down and by
dint of focus comes up with an answer that will be acceptable
by all? Not likely.
The rot is in the URI. It has to be solved there. The
solutions are obvious.
1. Insist that the URI is always dereferenceable, meaning,
by near equivalent syntax (case being the dull exception),
one can always attempt the dereferencing operation. The
result will be success or failure.
2. Understand that the URI as Namespace string may in one
specification indicate a syntax hack for disambiguating
aggregate names, but that this overloading by a secondary
specification increases in practice the unreliability of
the primary role of resource identification and retrieval.
That is all. Specifications that overload the syntactical
devices of other specifications will produce ambiguity and
increase unreliability. Deal with it. How?
Put something at the end of the URI or let the customer
eat cake. Decide which is best for your business.
len
-----Original Message-----
From: Robin Berjon [mailto:robin.berjon@expway.fr]
> Might it make sense to move namespace discussions to a forum of their
> own?
I fully agree, it's well past time we tried to put that to rest once and
for all.
|
