Re: httpRange-14 Change Proposal [editorial tweaks]

On 3/29/12 10:02 AM, Danny Ayers wrote:
> On 29 March 2012 15:44, Kingsley Idehen<kidehen@openlinksw.com>  wrote:
>> On 3/29/12 9:05 AM, Danny Ayers wrote:
>> What about inferring, with clarity, that a resource and its representation
>> are *potentially* ambiguous without additional clarity provide by relations
>> of the kind delivered by :describedby.
>>
>> We are dealing resource representation resolution and fidelity. If you come
>> at the resource from a certain world view (e.g., information space) you have
>> a resource with fine-grained representation.  For instance, and HTML
>> resource is structured enough for browsers to render Web Pages.
> Yep, that's how I read it too.
>
> As far as a browser is concerned there isn't really a distinction
> between "representation"/"authoritative representation" as I put it,
> "nominal representation"/"representation" as Jeni has put it or
> "HTML"/"HTML+metadata" as would be on the wire in this case.
>
>> Come at it from a different world view (e.g., data space) and you have a
>> blob since the content doesn't conform to the required structural fidelity
>> of an entity-attribute-value graph, and even worse, actual pathways to said
>> graph become skewed (re. follow-your-nose navigation via name->address
>> indirection ) if hyperlink (e.g., HTTP URI) based name and address choices
>> introduce ambiguity.
> There can still be ambiguity whatever we do, but the describedby (or
> 303-implied) metadata makes it possible to reduce it, no?

Yes, *reduce* is cleaner since we can never totally eradicate ambiguity 
once the system has a modicum of intelligence :-)
>
> Cheers,
> Danny.
>


-- 

Regards,

Kingsley Idehen 
Founder&  CEO
OpenLink Software
Company Web: http://www.openlinksw.com
Personal Weblog: http://www.openlinksw.com/blog/~kidehen
Twitter/Identi.ca handle: @kidehen
Google+ Profile: https://plus.google.com/112399767740508618350/about
LinkedIn Profile: http://www.linkedin.com/in/kidehen