[
Lists Home |
Date Index |
Thread Index
]
On Sat, 25 Jan 2003 22:13:27 +0000, Miles Sabin <miles@milessabin.com>
wrote:
> As far as network protocols and software are concerned, abstract
> Resources do no work at all. What matters in a retrieval context is that
> there be a functioning server that's capable of returning a response,
> maybe with a response entity, maybe without.
>
> So, if there were no Resources, or more than one, or different ones on
> different occasions, what would break? Can you name one piece of working
> software that'd stop working if Resources were to vanish in a puff of
> existential smoke overnight?
Weeeellllll, I guess it depends on what a "Resource" is. I think we're all
comfortable with the assumption that http://www.cnn.com is a "resource" for
up to date news. Whether that is a Resource in the RESTifarian sense is
another matter. Certainly it returns a different representation every few
minutes (and for all I know it may support content negotiation and return
raw text or WML or whatever on demand). If the idea of cnn.com as a
Resource totally vanished and one had to retrieve the news with a URL that
qualified the time, e.g. http://www.cnn.com/2003/January/24/19.47 then a
whole lot of people would stop using the site, I suspect.
Still, I think there's a lot of existential latitude between "cnn.com is a
resource for news" and "http://www.cnn.com is a Resource representing the
news and one retrieves a Representation of the up to the minute news by
dereferencing it, and one can make assertions about 'news' by using
http://www.cnn.com as an identifier of that abstract concept." I can't say
that the second is wrong, but I will assert that the empirical success of
CNN.com as a website says absolutely nothing about its theoretical validity
or utility. Are the webmasters at CNN.com adhering to the URI/REST religion
by updating the site with a new representation every few minutes? I don't
think so -- they're updating the content. Empirically, websites (a concept
never discussed formally in these discussions, AFAIK) do what their
maintainers configure them to do: some can return RDF descriptions of the
abstraction their owner hopes to identify, some return static HTML
irrespective of the HTTP content type headers, some of which try very hard
to implement RFC 2396 ... but most just sortof do whatever the webmasters
feel motivated to do with them. "Resource" and "Representation" seem like
reasonable abstractions for all this, but they are VERY abstract
abstractions.
Nevertheless, while I would quibble with some of the more categorical
statements that this doesn't work at all, I have to agree with the gist of
what Miles, Simon, and others seem to be saying -- the TAG is trying to
squeeze WAY to much juice from this rather dry fruit. If they are trying to
understand the actual principles of the Web by focusing on URIs, resources,
and representations, I'm extremely skeptical that they will produce
anything particularly useful to guide Webmasters, Semantic Web researchers,
Web services theorists or practicioners, etc.
[speaking only for myself, wearing no W3C or Day Job hat]
|
