Re: yet another sidetrack on what a URI identifies

On Fri, 2003-01-24 at 11:59, Roy T. Fielding wrote:
> Please tell me if I'm missing something, but the part of Dan's formalism
> that describes these issues is summarized in
> 
>     http://www.w3.org/Architecture/state.html
> 
> which uses the term message instead of representation (because he isn't
> distinguishing between message control data and payload, yet).

Hmm... I think I do distinguish messages from representations and from
control data and payload...

"Note that a message is not just the bytes sent; it's the event of
sending the packets. We don't say that a message is "sent twice";
rather, we say that two messages are sent with the same bytes. (This is
subtly different from the defintion of message in the HTTP 1.1 spec.
@@hmm... for this reason, use "event" in stead of message? I don't think
so...)"

As you can see, I haven't completely finished thinking about
which terms to use for which concepts...

>   That
> looks pretty close to REST, right?

I believe so, though I'm not familiar with any formal
characterizations of REST. (sorry if it's in there;
haven't read the whole dissertation.)

>   Or is that different from the Larch
> example
> 
>     http://www.w3.org/XML/9711theory/HTTP

I think it's the same...

> which I am afraid doesn't make much sense to me because it refers
> to entire messages in comparisons rather than elements within
> those messages

In the larch trait

 Message is an actual communications event
 HTTPFormat is a byte sequence (or similar syntactic data structure)
	that has the header fields and body
 Literal is a byte sequence with a mime type

e.g. to get the body out of a hunk of HTTP syntax...

      content: HTTPFormat -> Literal


>  (and I think something is wrong with the charset).

Yes, I was a bit fast and loose there... try the plain text version
  http://www.w3.org/XML/9711theory/HTTP.lsl

> I don't follow how that defines caching in terms of shared memory,

It doesn't. Oops; did I somehow imply that it does?

>  but
> I'm not sure if Dan was using it as an example of a REST formalism or
> an example of a web page formalism.

Neither; I was just using it as evidence of more than talk
when working thru web architecture issues.
Sorry if that's a distraction.

-- 
Dan Connolly, W3C http://www.w3.org/People/Connolly/