Re: Classification of ISSUE-57 change proposals

On 3/30/12 2:23 PM, Noah Mendelsohn wrote:
>
> I was not trying to say anything subtle about representations in my 
> earlier e-mail, though that is the term used in the SDW finding. The 
> point is that HTTP responses should be self-describing.
>
> Noah 
Yes! This is such an important point.

For RDF systems, relations such as :describedby etc.. are useful.

For Linked Data systems that extend RDF to the Web in general, 
:describedby relations are useful but also problematic since they 
introduce relation semantics to users and developers -- that are at best 
-- initially oblivious to their virtues. Thus, forcing it to matter 
triggers bootstrap inertia.

The bootstrap path looks something like this:

[Document Web Users & Developers] --> [Linked Data Web Users & 
Developers] --> [Semantic Web Users & Developers] .

The problem with Linked Data is that it rides a very delicate line 
between the Web as used today and the semantically enhanced aspects 
taking shape right now and well into the future.

The challenge is ultimately about deft management of bootstrap inertia 
or escape velocity. The migration flow and direction is a critical aid 
in this endeavor.

Jonathan is fixing the narrative which we all agree has been broken 
since inception re., HttpRange-14. Thus, I see the conclusion of his 
effort valuable to all, ultimately.

-- 

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