RE: "information resource"

> -----Original Message-----
> From: ext Stuart Williams [mailto:skw@hp.com]
> Sent: 11 October, 2004 16:43
> To: Stickler Patrick (Nokia-TP-MSW/Tampere); "public-
> webarch-comments"@w3.org
> Subject: Re: "information resource"
> 
> 
> Patrick,
> 
> Further to our recent exchanges and my earlier proposal to 
> replace the 
> term "Information Resource" with the term "Web Resource", at 
> our F2F tin 
> Basel this week, the TAG spend a considerable time discussion the 
> definition of "Information Resources". We developed new definitional 
> text which the TAG reached concensus on. This text now appears at the 
> beginning of Section 2.2 [1] and is quoted below.


In short, the new text does not address my concerns. 

Specific comments and a proposed alternate text offered below. 


> <quote>
> 2.2. URI/Resource Relationships
> 
> By design a URI identifies one resource. We do not limit the scope of 
> what might be a resource. The term "resource" is used in a 
> general sense 
> for whatever might be identified by a URI. It is conventional on the 
> hypertext web to describe web pages, images, product 
> catalogs, etc. 

Which should be expected, because the interchange of information is
arguably the key purpose of the hypertext web. Fine.

But "the hypertext web" != "the web". So there's a bit of (possibly
unintended) spin there, I think.


> as 
> “resources”. The distinguishing characteristic of these resources is 
> that all of their essential characteristics can be conveyed in a 
> message. We identify this set as “information resources”.

Hmmm... I believe I can convey "all of the essential characteristics" 
of a particular model of mobile phone in a message. Likewise, for the
planet Venus, the concept of slightly watery fudge pudding, etc.

Also, given a particular "document", e.g. describing the city
of Paris, I would not really consider the content of the document
itself to fully subsume all of its "essential characteristics". 
Rather, I would consider "characteristics" to (also) include metadata 
about  the document, such as the creator, last modified date, topical 
categorizations, etc. which may not be included in every, or any,
representation of that resource.

This text does nothing to distinguish between the actual city
of Paris and a web page about the city of Paris. Both can be
denoted by URIs and have their "essential characteristics conveyed
in a message".

Why not simply state that an "information resource" *is*
information -- i.e. a body of information??? Why talk about
characteristics, essence, etc. which can be interpreted in 
various ways which do not fully support the distinction you
are trying to make? 

(see my proposed alternate text below)

> This document is an example of an information resource. It 
> consists of 
> words and punctuation symbols and graphics and other 
> artifacts that can 
> be encoded, with varying degrees of fidelity, into a sequence 
> of bits. 
>
> There is nothing about the essential information content of this 
> document that cannot in principle be transfered in a representation.

How does "essential characteristics" intersect with "essential
information content"? And can there be content of an information
resource that is not "essential"?

> However, our use of the term resource is intentionally more 
> broad. Other 
> things, such as cars and dogs (and, if you’ve printed this 
> document on 
> physical sheets of paper, the artifact that you are holding in your 
> hand), are resources too. They are not information resources, 
> however, 
> because their essence is not information. 

I like the above text, from "However, our use of...". It reflects
the primary distinction between information resources and other
resources -- in that information resource *are* information.

If you would simply, above, define an information resource as a
resource that constitutes a body of information, then this
text would fit very nicely with that simple, precise definition.

(see my proposed alternate text below)

> Although it is possible to 
> describe a great many things about a car or a dog in a 
> sequence of bits, 
> the sum of those things 

Sum of those *bits*? Or sum of all resources (cars, dogs, etc.)?

> will invariably be an approximation of the 
> essential character of the resource.

I think you're using "character" in a very narrow manner that
most folks won't immediately grasp, or necessarily agree with.

The "character" of a particular wine is not likely to be understood
as the chemical makeup of that wine or the details of how that wine is
produced, but in the sensory affect that wine has on ones senses
of taste and smell (and perhaps also the alcohol content and
subsequent affect).

Likewise, the "essense" of a particular wine is likely to be
understood in a similar fashion to its character or characteristics.

And one can certainly think it reasonable to convey the characteristics
of a wine, even the essense of a wine, in a message.


> We define the term “information resource” because we observe 
> that it is 
> useful in discussions of web technology and may be useful in 
> constructing specifications for facilities built for use on the web.
> </quote>

Fine. No problem with that text. But you have to nail down
the definition of information resource a bit more precisely.

And, BTW, sticking with including the definition of "information
resource" does not preclude the definition of a "web resource",
*both* of which will be "useful in discussions of web technology",
etc.

(see my proposed alternate text below)

> This text moves the definition of "Information Resource" 
> in-line at the 
> beginning of section 2.2 rather than by forward reference to 
> the former 
> section 3.1.
> 
> Please can you let us know if we have addressed your comment to your 
> satisfaction (if possible before our next telcon which will be 18th 
> October).

No you have not, because the new definition "information resource" is
as ambiguous as the previous, and I can still see how one may arrive at
the conclusion that a dog is an information resource, because the
"essential characteristics" of that dog can be conveyed in a message.

Since "conveying characteristics in a message" seems analogous to
"represent the state of the resource", this new definition can be
construed as being the same, in essence ;-) as the previous: that
any resource for which one can convey its essential characteristics
in a message -- i.e provide a representation of its state -- is 
an information resource; thus any web accessible resource is an
information resource.

Sorry. But I do not find (all of) the new text acceptable. There
is some useful new text, as noted above, but the critical ambiguity
(and implicit suggestions about the nature of web accessible
resources) remains.

Here is how I would write this section:

<quote>
2.2. URI/Resource Relationships

By design a URI identifies one resource. We do not limit the scope of 
what might be a resource. The term "resource" is used in a general sense 
for whatever might be identified by a URI. 

While our use of the term resource is intentionally broad, there are, 
however, two particular classes of resources for which it is useful to 
draw an explicit distinction, and thus we define the two following terms 
to refer to these particular classes of resource:

"web resource"

   A "web resource" is a resource which has one or more web accessible
   representations. 

   Colloquially, web resources are said to be "on-the-web". Membership
   of a given resource in the class of web resources may change over time,
   as access to at least one representation of a given resource may change. 
   Furthermore, given mechanisms for security and access control, the set 
   of web resources may vary between different users or contexts. What may
   be web accessible for one user may not be for another. Dispite this
   variability, it is still a useful distinction to make.

"information resource"

   An "information resource" is a resource which constitutes a body
   of information.

   It is conventional on the hypertext web to describe web pages, images, 
   product catalogs, documents, etc. as “resources”. The distinguishing 
   characteristic of these particular kinds of resources is that they all 
   constitute bodies of information, and thus belong to the class of 
   "information resources". Information resources are central to many
   of the most successful and widespread applications of the web.
   
Note that not all web resources are information resources, nor are 
all information resources also web resources. There may be resources
which constitute bodies of information and which are named by URIs
but which are not web accessible. Likewise, there may be resources
which are named by URIs and are web accessible, but do not constitute
bodies of information.

This document is an example of an information resource. It consists of 
words and punctuation symbols and graphics and other artifacts that can 
be encoded, with varying degrees of fidelity, into a sequence of bits. 
All of the information content of this document can, in principle, be 
transfered in a representation. And since a representation of this document 
is also accessible via the web, it is also an example of a web resource.

The city of Paris, France is an example of a resource which is not an 
information resource. While information about the city of Paris may be 
conveyed in one or more representations, the city of Paris does not itself 
constitute a body of information, and therefore any web accessible 
representation of the city of Paris will inevitably constitute only an 
approximation or description of that resource, rather than conveying 
information content inherent in the resource itself. And if the city of 
Paris is denoted by the URI <http://example.com/cities/paris> and a 
representation of the city of Paris is accessible via that URI, then the 
city of Paris would be an example of a web resource, even though it is 
not an example of an information resource.

The terms "information resource" and "web resource" are not essential
to the fundamental architecture of the web, but we define these two terms 
because we observe that they are useful in discussions of web technology 
and web usage, and may be useful in constructing specifications for 
facilities built for use on the web.
</quote>

I expect that you and other members of the TAG may wish to finesse
the above wording, but I think that it captures exactly what can,
and should, be defined in AWWW regarding information and web 
resources, and what a reasonably broad majority of folks should find 
acceptable.

Regards,

Patrick



> Many thanks,
> 
> Stuart Williams
> On behalf of W3C TAG
> -- 
> [1] http://www.w3.org/2001/tag/webarch/#id-resources
> 
> Patrick.Stickler@nokia.com wrote:
> 
> >Thanks Stuart. I find the proposed changes acceptable.
> >
> >Cheers,
> >
> >Patrick
> >
> >
> >  
> >
> >>-----Original Message-----
> >>From: ext Stuart Williams [mailto:skw@hp.com]
> >>Sent: 20 September, 2004 15:52
> >>To: public-webarch-comments@w3.org
> >>Cc: Roy T. Fielding; Stickler Patrick (Nokia-TP-MSW/Tampere); Tim
> >>Berners-Lee
> >>Subject: Re: "information resource"
> >>
> >>
> >>I'd like make the following proposal which I hope will 
> >>address Patrick's 
> >>comment [1] and be acceptable to other parties with an interest in 
> >>Patrick's comment.
> >>
> >>1) Replace all occurences of the noun phrase "information 
> >>resource" with 
> >>the noun phrase "web resource".
> >>
> >>2) Replace the defining sentence for the noun phrase "information 
> >>resource" (section 3.1 1st para, 1st sentence) :
> >>
> >>  "The term Information Resource refers to resources that convey 
> >>information. Any resource that has a representation is an 
> information 
> >>resource."
> >>
> >>with
> >>
> >>  "The term Web Resource is applicable to resources for which web 
> >>acesssible representations are available and/or which may be 
> >>interacted 
> >>with through an exchange of representations."
> >>
> >>3) [Optional]  Consider adding a nearby sentence: 
> "Colloquially, Web 
> >>Resources are said to be "on-the-web"."
> >>
> >>Best regards
> >>
> >>Stuart Williams
> >>--
> >>[1]  
> >>http://lists.w3.org/Archives/Public/public-webarch-comments/20
> >>04JulSep/0047.html
> >>
> >>Roy T. Fielding wrote:
> >>
> >>    
> >>
> >>>I agree with the core of Patrick's argument, excerpted below, with
> >>>the minor exception that some web resources are 
> representation sinks
> >>>that have no difficulty processing information that is 
> sent to them,
> >>>even though they don't have representations of their own.  
> Those are
> >>>significant to the web machinery, but don't participate in the
> >>>information retrieval (hypertext) Web.
> >>>
> >>>Maybe we need to distinguish resources from web resources and from
> >>>hypertext resources?  Or maybe the architecture just 
> >>>      
> >>>
> >>doesn't care, and
> >>    
> >>
> >>>we can go back to describing how it works instead of how it might
> >>>be modeled in an abstract but artificial way.
> >>>
> >>>....Roy
> >>>
> >>>On Sep 9, 2004, at 2:00 AM, <Patrick.Stickler@nokia.com> wrote:
> >>>
> >>>      
> >>>
> >>>>But the resolution of that confusion need not posit any claims or
> >>>>constraints about the inherent nature of the resource itself, only
> >>>>about the accessibility of representations of that resource.
> >>>>
> >>>>I.e.
> >>>>
> >>>>-- 
> >>>>
> >>>>"resource"         Anything that can be referred to, 
> >>>>        
> >>>>
> >>named, described,
> >>    
> >>
> >>>>                   talked about, etc.
> >>>>
> >>>>"web resource"     A resource which has web accessible 
> >>>>        
> >>>>
> >>representations
> >>    
> >>
> >>>>                   (i.e. is significant to the web machinery).
> >>>>                   "web resource" is a subclass of "resource".
> >>>>
> >>>>"representation"   An octet stream (entity) returned by a 
> >>>>        
> >>>>
> >>server which
> >>    
> >>
> >>>>                   reflects the state of a resource. A 
> >>>>        
> >>>>
> >>representation is
> >>    
> >>
> >>>>                   also a resource, which can be denoted 
> >>>>        
> >>>>
> >>by a distinct
> >>    
> >>
> >>>>                   URI. A representation of a 
> >>>>        
> >>>>
> >>representation (resource)
> >>    
> >>
> >>>>                   corresponds to a bit-equal copy of itself.
> >>>>                   "representation" is a subclass of "web 
> >>>>        
> >>>>
> >>resource".
> >>    
> >>
> >>>>-- 
> >>>>
> >>>>IMO, the above three definitions should be sufficient to 
> >>>>        
> >>>>
> >>clarify the
> >>    
> >>
> >>>>confusion between what a resource is and what resources 
> >>>>        
> >>>>
> >>are relevant to
> >>    
> >>
> >>>>the web and why,  and how representations (the "atomic" 
> >>>>        
> >>>>
> >>resources of
> >>    
> >>
> >>>>the web) relate to the broader set of web resources -- 
> >>>>        
> >>>>
> >>many of which
> >>    
> >>
> >>>>correspond to abstract "bodies of information" such as web pages.
> >>>>
> >>>>Nowhere above is it necessary to say anything about the 
> >>>>        
> >>>>
> >>inherent nature
> >>    
> >>
> >>>>of resources or of web resources, or to posit any kind of class of
> >>>>"information resources" in order to describe the behavior and 
> >>>>architecture
> >>>>of web servers and clients (aside from the atomic, binary 
> nature of
> >>>>representations).
> >>>>        
> >>>>
> >>>On Sep 9, 2004, at 3:51 AM, <Patrick.Stickler@nokia.com> wrote:
> >>>
> >>>      
> >>>
> >>>>My explicit proposal would be to replace the words "information 
> >>>>resource"
> >>>>with either "web resource" or "web accessible resource" which IMO
> >>>>would coincide more precisely with the actual definition and not
> >>>>potentially imply any position relating to httpRange-14.
> >>>>
> >>>>        
> >>>>
> >>>...
> >>>
> >>>      
> >>>
> >>>>Anything can be a resource. I am opposed to any constraints by
> >>>>the web architecture on the nature of resources denoted 
> by URIs and
> >>>>for which representations are made web accessible.
> >>>>
> >>>>My view, in a nutshell:
> >>>>
> >>>>Anything can be a resource.
> >>>>Any resource can be denoted by a URI.
> >>>>Any resource can have web accessible representations.
> >>>>A web resource is a resource with web accessible representations.
> >>>>A representation is a resource.
> >>>>A representation can be denoted by a distinct URI.
> >>>>A representation is the atomic primitive of the web.
> >>>>A representation corresponds to a binary data stream.
> >>>>The representation of a representation is a bit-equal copy 
> >>>>        
> >>>>
> >>of itself.
> >>    
> >>
> >>>>The web architecture faciliates interaction with 
> >>>>        
> >>>>
> >>representations of 
> >>    
> >>
> >>>>resources.
> >>>>The semantic web architecture facilitates interaction with 
> >>>>descriptions of resources.
> >>>>The intersection of the web and semantic web architectures are a 
> >>>>shared set of URIs.
> >>>>
> >>>>Patrick
> >>>>        
> >>>>
> >>>
> >>>      
> >>>
> >>    
> >>
> >
> >  
> >
> 
> 
> 

Received on Tuesday, 12 October 2004 09:12:58 UTC