Re: Generic URI things

On Fri, Jan 10, 2003 at 02:26:18PM -0500, Christopher B Ferris wrote:
> > In progress, at least for Waka.  But while that's important, it doesn't
> > matter to the uniform constraint.  It's uniform, because the meaning of
> > "MONITOR" and "WATCH" is meaningful to all things with identity (I can
> > use it to monitor dogs, chickens, lava flows, weather reports).  The
> 
> No it isn't.

I assume you mean "No" to the second sentence, not the third.

> It is meaningful to resources which support that method for 
> starters.

Not at all.  The meaning of the message on-the-wire is independant of
whether the service implements it or not.

> Secondly, unless I'm mistaken, I still have no idea whether this
> is your MONITOR or mine... Will I get a lizard on my doorstep next week
> or will I be notified that the state has changed through some less toothy 
> means?

Sure, that's an issue, which is why registration (normally via some
form of standardization) is important.  Or alternately, identify the
method with a URI, ala PEP (and maybe Waka, dunno).

> Further, can you demonstrate/prove how it is that your assertion that the 
> uniformity
> of the interface with regards to all identified resources is what actually
> derives the benefits you suggest?

Well, that's the gist of a good amount of Roy's dissertation.  Section
5.1.5 doesn't say anything about the interface needing to be standardized
to realize the benefits.

> *MIGHT* it be the case that it is the 
> standardization
> within a domain that has accorded much of the benefits observed?

Obviously standardization is responsible for much of the Web's
growth, but AFAICT, it has nothing to do with the technical benefits of
the uniform interface.

It is true that the value of standardization is greater because the
interface is so general (because of reuse), if that's what you mean.

MB
-- 
Mark Baker.   Ottawa, Ontario, CANADA.        http://www.markbaker.ca
Web architecture consulting, technical reports, evaluation & analysis

Received on Saturday, 11 January 2003 23:18:11 UTC