W3C home > Mailing lists > Public > public-awwsw@w3.org > April 2008

RE: network endpoints

From: Williams, Stuart (HP Labs, Bristol) <skw@hp.com>
Date: Wed, 16 Apr 2008 16:55:02 +0000
To: Jonathan Rees <jar@creativecommons.org>, "public-awwsw@w3.org" <public-awwsw@w3.org>
Message-ID: <9674EA156DA93A4F855379AABDA4A5C611CE7DD846@G5W0277.americas.hpqcorp.net>

Hello Jonathan,

> -----Original Message-----
> From: public-awwsw-request@w3.org
> [mailto:public-awwsw-request@w3.org] On Behalf Of Jonathan Rees
> Sent: 16 April 2008 14:29
> To: public-awwsw@w3.org
> Subject: network endpoints
>
>
> One thing I would like to add to my diagram: A box for "network
> endpoint" with the meaning of a real-world source of
> awww:representations (e.g. a "web page" operationally defined by the
> process of sending an HTTP request specifying a particular resource-
> name string to a particular server, using the Internet, and so on).

Oooohhh, I'd really rather that you didn't - not unless you really really have to.

I'd really rather see the detail of the infrastructure of the web as something  under the hood - outside the domain of discourse - which supports and abstraction of a world of resources perceived through exchanges of representations. I'd prefer, as I said to Alan [1], to regard requests as being questions asked of the web (ie. the web infrastucture) and responses as originating from web infrastructure without having to give a detailed account of what's going on under the hood. eg. proxy's and caches - worse (and maybe this takes to a place where we have to consider the machinery) transcoding proxies or proxies that introduce/suppress advertising - but  I think we should speak of the simple cases for now... and I don't think that there is a compelling need to speak of HTTP network endpoints.

eg. discussing 404 or 40x variants becomes very elaborate - rather than for some reason web infrastructure was unable to find the resource.

[1] http://lists.w3.org/Archives/Public/www-tag/2008Apr/0057

[Aside: you might search www-tag for the phrase "hypertext dispenser" possibly hyphenated]

> This is an interesting class of things that David Booth has
> articulated on a number of occasions. Tim and others have clearly
> said that these things are *not* IRs. I think it's more useful to say
> *how* endpoints relate to the intended information resources /
> denotations than to just say that they're not information resources.

I don't see how think about them at all helps... but I could be persuaded.

> "Endpoint" would be placed near "value cloud" in the diagram and
> would be related to "value" by a relationship "responds to 200
> with ... (at time t)" and to "information resource" by another
> relationship that's analogous to "faithful to" i.e. for all time t
> any value delivered is a kr-representation of the state of the
> information resource.

Give it a try I suppose... but I think that the model will become more prescriptive than it need be about the internal structure of web infrastructure - and offer folks a larger target to say... well I model it differently... like this...

> An "endpoint" could also be related to a "value cloud" via
> composition of relationships, but that would probably be clutter.
>
> Once all of these separate entities are defined we can start to talk
> about the relationships and invariants between them - e.g. does every
> endpoint have an associated IR? Are there endpoints whose URIs do not
> denote information resources? Are there endpoints for which the URI
> denotes an information resources but that is not faithful to that IR?
> For me it is this kind of question (not necessarily these particular
> ones) that will be useful in cracking the question of web semantics.
>
> None of this is to say I understand as well as Tim does what things
> can be "information resources" and what things can't... but I'm very
> happy to know that the URI does *not* generally denote the value
> cloud and that the value cloud isn't even part of or determined by
> the information resource. (Constrained, yes, but not determined.)
>
> I'll make the change in a day or two or three if I hear no outcry.

I guess consider this a little cry out :-)

> Jonathan

Stuart
--
Hewlett-Packard Limited registered Office: Cain Road, Bracknell, Berks RG12 1HN
Registered No: 690597 England



>
> On Apr 15, 2008, at 8:34 PM, Jonathan Rees wrote:
>
> > Revised diagram based on this morning's meeting is here:
> >
> > http://sw.neurocommons.org/2008/inforesource.pdf
> >
> > (N.b. that's PDF, not PNG.)
> >
> > I've renamed 'abstract document' to 'information resource' since
> > the consensus was that they're the same.
> > The main reason to include the 'value cloud' in the diagram is to
> > help me understand how 'information resource' relates to Fielding's
> > formal definition of 'resource'.
> >
> > Let me know how you like it.
> >
> > Jonathan
> >
> >>
> >> http://sw.neurocommons.org/2008/inforesource.png
> >>
> >> which I will not take the time to prettify now (I don't know why
> >> the background is gray)
> >>
>
>
>
Received on Wednesday, 16 April 2008 16:59:17 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 July 2008 07:55:27 GMT