W3C home > Mailing lists > Public > semantic-web@w3.org > July 2009

Re: Dons flame resistant (3 hours) interface about Linked Data URIs

From: Kingsley Idehen <kidehen@openlinksw.com>
Date: Sun, 12 Jul 2009 14:57:58 +0100
Message-ID: <4A59EBE6.9070906@openlinksw.com>
To: Steve Harris <steve.harris@garlik.com>
CC: semantic-web at W3C <semantic-web@w3c.org>, public-lod@w3.org
Steve Harris wrote:
> On 10 Jul 2009, at 15:36, Kingsley Idehen wrote:
>> Steve Harris wrote:
>>> On 10 Jul 2009, at 14:31, Kingsley Idehen wrote:
>>>> Steve et. al,
>>>>
>>>> If we are going to take the "how the Web was born" theme re. 
>>>> figuring out the path forward, then what's wrong with RDFa? If 
>>>> people sort of know how to write HTML, why not show them how to add 
>>>> rich metadata via RDFa? That said, we have a deeper problem re. 
>>>> Linked Data, and in my opinion it starts not fulling expressing the 
>>>> essence of the matter with clarity. The fundamental issues are
>>>
>>> RDFa doesn't generally solve the Syntax complexity problem.
>> It solves the "groking what your actually doing"problem for those  
>> who  author  HTML  docs.
>
> Perhaps, but I'm not totally convinced. I think the mapping between 
> RDFa and triples is sufficiently complex that it may not help.
>
>>> Though, possibly RDFa documents that are not "nice" HTML (ie. not 
>>> really readable by humans) could be quite hacker-friendly. I've been 
>>> meaning to look into this.
>> RDFa is the best starting point for enhancing Metadata carried by an 
>> HTML document. Once you understand that you are describing something, 
>> and that you do so using Subject, Predicate, Object statements, the 
>> essence of the matter is much much clearer.
>
> If people make the leap between RDFa syntax and triples, yes.
>
>> Once  high level annotation tools for embedding RDFa in HTML are 
>> unleashed, this whole matter will become much clearer to a very broad 
>> spectrum of Web users :-)
>
> Now, that I definitely disagree with. The broad spectrum of web users 
> do not edit HTML, and I would guess that the majority of HTML out 
> there is machine generated. At least in part.
>
> That's not to say that I think such a tool is a bad idea, I don't, but 
> that it wont be any kind of universal panacea.
Steve,

I believe HTML out there today will eventually be replaced by HTML 
carrying Rich Metadata that has been encoded using RDFa or  HTML5's 
microdata feature.

Thanks to RDFa (in particular) I can now hold conversations with SEO 
specialists about improving discoverability of Web pages via a few new 
HTML tag properties. This kind of  conversation simply couldn't happen 
12 months ago.

I don't believe people craft HTML by hand, the biggest impact is going 
to be with the generators of HTML who will be selfishly guided by 
optimal placement in the indexes of Yahoo! and Google, at first. The 
broader implications of the Web of Linked Data will follow from that 
point onwards.

Kingsley
>
> - Steve
>


-- 


Regards,

Kingsley Idehen	      Weblog: http://www.openlinksw.com/blog/~kidehen
President & CEO 
OpenLink Software     Web: http://www.openlinksw.com
Received on Sunday, 12 July 2009 13:58:38 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 1 March 2016 07:42:13 UTC