RE: network endpoints

> From: Jonathan Rees [mailto:jar@creativecommons.org]
>
> On Apr 28, 2008, at 1:13 PM, Booth, David (HP Software -
> Boston) wrote:
> >> From: Jonathan Rees [mailto:jar@creativecommons.org]
> >> [ . . . ]
> >> So I do not consider this a discussion on the definition of
> >> awww:InformationResource. It may be unclear but we are not
> >> at liberty to redefine it - it's published.
> >
> > I think we *need* to redefine it.  The published definition at
> > http://www.w3.org/TR/2004/REC-webarch-20041215/#def-information-
> > resource
> > is unworkable.  It is: (a) known to be flawed; (b) not how the rest
> > of the document implicitly defines it; and (c) not the actual
> > concept on which the Web architecture is based.
>
> I agree that we need a new definition, and with your reasons. I don't
> agree that we should recycle the same name. I'm having this same
> argument with Alan about whether one ought to attempt a fresh
> definition of information resource...
>
> We could have for example:
>         awww:IR-as-explicitly-defined
>         awww:IR-as-implicitly-defined
>         dbooth:IR-on-which-web-architecture-is-based
> or shorter names, or even serial numbers (although those would be
> hard to remember).

Sounds good.

>
> >> I consider what we're doing to be
> >> (in part) an attempt to articulate the intended semantics and
> >> recommended practice around 200 responses.
> >
> > Right, but I don't see how we can do that without a workable
> > definition of "information resource".
>
> We need to find the right IR-like class, and define some name (URI or
> informal) to denote that class. We could recycle a name that exists,
> but for now I think it's less confusing to just use a different name.

Sounds good.  Perhaps we should just be more careful to point to the actual definition we're using.

> >
> > BTW, I just noticed an interesting miscommunication.  I've been
> > using the term awww:InformationResource to refer to the concept of
> > "information resource" that I believe the architecture of the WWW
> > *intended* (basically a function from time and requests to
> > representations) -- *not* how the term is currently defined at
> > http://www.w3.org/TR/2004/REC-webarch-20041215/#def-information-
> > resource
> > since that definition is erroneous.
>
> Indeed. I have been using awww: generally to mean "in the sense of
> the AWWW document, even if we don't really understand or agree with
> it". But I think I see your point. We have a collision. Shall I
> change the name I'm using, or would you like to change yours?

I think it makes more sense for me to change mine.  I'll start using ftrr:IR or ftrr:InformationResource, as defined here:
http://lists.w3.org/Archives/Public/public-awwsw/2008Apr/0046.html

>
> Generally when I use "information resource" unqualified I am doing my
> best to understand what Tim means by it and/or what AWWW intends to
> mean by it, and I don't mind using a term I don't understand for the
> time being because I'm using it as a probe to find out what others
> think is and isn't "good" with respect to web architecture.

Sounds reasonable.

>
> > If we can get back to use cases it would be helpful to me at least,
> > since I'm struggling to figure out what it is that folks are
> > finding difficult to do.  For example, I've offered an n3
> > definition of what can be inferred from a 200 response, but Alan
> > (privately) has complained that the inferred information is
> > tautological, so it seems like he's wanting something different,
> > but I don't know exactly what.  Would it help to look at an example
> > when viewing an IR as a function from time and requests to
> > representations?  We briefly looked at the relationship between
> > these two IRs in W3C's TR space:
> >
> >   http://www.w3.org/TR/rdf-mt/  (generic)
> >   http://www.w3.org/TR/2004/REC-rdf-mt-20040210/ (dated URI)
> >
> > Would it be helpful to others to see those relationships between
> > those, and the rules that are used to infer them, coded in n3?  If
> > not, can someone suggest a use case example they'd like to see
> > analyzed?
>
> It would be helpful to me to see how you would do it and contrast
> that with how others would do it, since I think your definitions of
> IR are materially different from those of some others in this group.

If you take Roy's definition to be essentially "a function from time to representation sets", then the ftrr:IR definition is pretty much equivalent to Roy's definition.  The function in Roy's definition is pretty much a curried version of the function in the ftrr:IR definition.

>
> Here are some of the cases I'm thinking of:
> [ . . . ]

I'll address these in a separate message.


David Booth, Ph.D.
HP Software
+1 617 629 8881 office  |  dbooth@hp.com
http://www.hp.com/go/software

Opinions expressed herein are those of the author and do not represent the official views of HP unless explicitly stated otherwise.

Received on Tuesday, 29 April 2008 04:30:39 UTC