Re: Let's blow some new life into this community group

On 7/4/12 1:49 PM, Melvin Carvalho wrote:
>
>
> On 4 July 2012 19:43, Kingsley Idehen <kidehen@openlinksw.com 
> <mailto:kidehen@openlinksw.com>> wrote:
>
>     On 7/4/12 12:19 PM, Melvin Carvalho wrote:
>>
>>
>>     On 4 July 2012 17:44, Michiel de Jong <michiel@unhosted.org
>>     <mailto:michiel@unhosted.org>> wrote:
>>
>>         On Wed, Jul 4, 2012 at 5:31 PM, Kingsley Idehen
>>         <kidehen@openlinksw.com <mailto:kidehen@openlinksw.com>> wrote:
>>         > On 7/4/12 9:37 AM, Michiel de Jong wrote:
>>         >>>
>>         >>> >integrate with the web.
>>         >>
>>         >> when you say 'integrate with the web', maybe you mean
>>         sparql+rdfbase
>>         >> again, if so, then i probably disagree that we want to
>>         impose that
>>         >> choice.
>>         >
>>         >
>>         > He means, the basic pattern of the Web. Denote (name)
>>         things using URIs. Use
>>         > URIs that resolve to useful information. That's it.
>>         >
>>
>>         ok, then we agree :)
>>
>>     Yes that's right.  And in a certain way, HTTP is one of the focal
>>     points, for the web.  Two nice things with HTTP are 1) it has a
>>     large and growing network effect 2) that it can be made to
>>     interact with other protocols (is kind of what I was saying above).
>>
>
>     Remember, we don't need to be protocol specific. Just use the best
>     protocol for the task at hand. That's the beauty of URI
>     abstraction re. entity denotation (naming) and de-reference
>     (resolution) to useful information, via the Web :-)
>
>
> This is correct.  But a widely used protocol like HTTP benefits from 
> 20 years of tooling and the network effect, which translates into a 
> big advantage, to those that embrace it.

Of course, but in scenarios where its optimal.
>
> Of the newer protocols (eg webfinger to name one of many) , it's 
> anyone's guess which will succeed.  But, I watch on with interest :)

Webfinger uses HTTP for data access. Basically, it uses HTTP for 
something HTTP does well. At the same time, it doesn't use HTTP for 
real-world entity denotation (naming) since that's an area where the 
cost of HTTP (re. end-user intuition) is significant.

Kingsley

>
>
>     -- 
>
>     Regards,
>
>     Kingsley Idehen	
>     Founder & CEO
>     OpenLink Software
>     Company Web:http://www.openlinksw.com
>     Personal Weblog:http://www.openlinksw.com/blog/~kidehen  <http://www.openlinksw.com/blog/%7Ekidehen>
>     Twitter/Identi.ca handle: @kidehen
>     Google+ Profile:https://plus.google.com/112399767740508618350/about
>     LinkedIn Profile:http://www.linkedin.com/in/kidehen
>
>
>
>
>


-- 

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