RE: New Editors Draft of the httpRange-14 Finding

Hi Rhys,

Yes, I intentionally avoided proposing any definition of information
resource in that review, because I didn't want that thread to turn into
a debate about the proper definition.  

The main problem with the current definition[1] is that it only
addresses static IRs.  The current definition says that "their essential
characteristics can be conveyed in a message"[1].  
But consider a web page that reports the current weather in Oaxaca.  The
essence of any *particular* weather report can be transmitted in a
message, but it is not possible to capture, in a message, the essence of
the current weather report in general.

A second problem is that the definition is unclear.  People are left
guessing whether a particular resource qualifies as an IR or not.

Here are a few suggested definitions of information resource (IR):

 1. a source of representations
 1a. a network source of representations
 1b. a Web source of representations

 2. a function from time and requests to representations 

 3. a network source/sink of representations/requests

All of these assume that there is a suitable WebArch definintion of
"representation".

All in all, I think either #1 or #2 would be best for the WebArch
document -- probably #1 (or one of its variants, #1a or #1b) because it
is simpler and less intimidating, though #2 is more precise.  Maybe the
best choice would be #1 as the basic definition, and #2 in accompanying
explanation.  One negative about #1 is the fact that it would not
obviously cover an IR like a web page for submitting anonymous crime
tips, in which the main purpose of such an IR is to *consume*
information -- not produce representations.  But that is a corner case
and I think accompanying prose could explain that such a page is still
an IR.

To explain why I also included #3 in this list, consider the question:
Why is it important for the WebArch to distinguish between IRs and
non-IRs?  (I recognize that there are some who believe that such a
distinction is not necessary, but for the purpose of this discussion
let's assume that it is.)  Why doesn't the WebArch distinguish between
mammal resources and non-mammal resources?  I think the reason is that
the WebArch talks about the Web, requests and representations, and thus
it is relevant to distinguish between resources that are "on the Web" --
that can accept requests and produce representations -- and other
resources.  In other words, I think a key characteristic of an IR is the
fact that it is (potentially) network-accessible.  That's what really
makes it relevant to Web architecture.


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.
 

> -----Original Message-----
> From: www-tag-request@w3.org [mailto:www-tag-request@w3.org] 
> On Behalf Of Rhys Lewis
> Sent: Thursday, August 30, 2007 5:36 AM
> To: www-tag@w3.org
> Subject: RE: New Editors Draft of the httpRange-14 Finding
> 
> Hello David,
>  
> I'm working through the comments we've received on this draft.
>  
> One of your comments relates to the definition of Information 
> Resource in AWWW [1]. You've made it clear that you disagree 
> with the existing definition, but didn't explain why in your 
> mail. I appreciate that this is probably a long held view for 
> you and I apologise for not being aware of the history.
>  
> It would help me greatly if you could point to a definition 
> that you feel is appropriate or a description of what you 
> feel is incorrect in the current definition in AWWW.
>  
> Best wishes, and thanks again for taking the time to provide comments
>  
> Rhys 
>  
> [1] http://www.w3.org/TR/webarch/#def-information-resource
>

Received on Thursday, 30 August 2007 17:51:50 UTC