On 6/17/13 8:34 AM, Luca Matteis wrote:
Come on! If you're building something that works like the Web but isn't using HTTP, then it's *not* the Web. It's something else that has similar dynamics to the Web (like, I dunno, a gazillion of other things?).

You see, that's the fundamental misconception. We have "Web" as a colloquialism for "World Wide Web" and in that loose use the misconception that "Web" or "World Wide Web" are monikers for a system that's HTTP specific.

Simple question: why do many browsers handle the following schemes:

1. ftp
2. mailto
3. tel.

Why do mobile user agents implicitly cater for multiple URI schemes i.e., URI scheme handlers are native to the application development environment? Custom URI resolvers are natural to the architecture of a Web.

As you can see, there is a common theme here i.e., generic terms such as Web, Linked Data etc.. are now being shoehorned into very specific boxes that ultimately contradict the very architecture and philosophy of what's become the World Wide Web.

HTTP (based on increasing ubiquity) is a cost-effective route to World Wide Web participation and exploitation. That doesn't mean its the only option, far from it!

URIs are the kernel. The seed of ingenuity that delivered the World Wide Web on global scale.

Kingsley


On Mon, Jun 17, 2013 at 2:26 PM, Kingsley Idehen <[email protected] <mailto:[email protected]>> wrote:

    On 6/17/13 8:17 AM, Luca Matteis wrote:

    On Mon, Jun 17, 2013 at 2:06 PM, Kingsley Idehen
    <[email protected] <mailto:[email protected]>> wrote:

        The Web isn't about being draconian or tightly coupled to
        anything.


    But the Web *IS* tightly coupled to HTTP! Why can't Linked Data
    then be tightly coupled to RDF?

    The Web isn't tightly coupled to HTTP.

    HTTP is an effective route to a global Web.

    The magic is in the URI, the ability to provide abstraction that
    enables the loose coupling of data access protocols and data
    representation formats.

    FWIW -- when we started releasing Linked Data (at the start of
    this journey) we did so using resolvable URIs for a variety of
    schemes, not just HTTP. Even today, in the context of Web-scale
    verifiable identity, we produce Linked Data solutions that don't
    mandate HTTP scheme URIs while actually exploiting the kind of
    entity relationship fidelity that RDF delivers.

    The beauty of the World Wide Web is that it is actually loosely
    coupled at its architectural core. HTTP is a productive short-cut
    to the Web due its increasing ubiquity.

--
    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




Attachment: smime.p7s
Description: S/MIME Cryptographic Signature

Reply via email to