W3C home > Mailing lists > Public > public-media-annotation@w3.org > November 2008

Re: URIs as value

From: Felix Sasaki <fsasaki@w3.org>
Date: Wed, 12 Nov 2008 09:26:54 +0900
Message-ID: <491A22CE.4040003@w3.org>
To: Silvia Pfeiffer <silviapfeiffer1@gmail.com>
CC: Pierre-Antoine Champin <swlists-040405@champin.net>, public-media-annotation@w3.org

Silvia Pfeiffer さんは書きました:
> On Wed, Nov 12, 2008 at 10:01 AM, Felix Sasaki <fsasaki@w3.org> wrote:
>   
>> Saying "we provide an RDF based format and mandate its usage in the API" is
>> like an invitation to browser vendors not to implement the API. I think that
>> would be a mistake. Remember that still we mostly have people from the
>> academic world on board. I don't want to loose them, but want to have an
>> approach which encourages more particiation from also that industry.
>>     
>
> I don't think RDF was suggested as the only format. Just an
> addition/alternative way of specifying the data.

Understand, and sorry for the misunderstanding on my side. The tricky 
part will be the API design, when we have to define return types for 
e.g. getCreator . Of course we could say "text or URI", but what happens 
with the URI? Is it resolved? What happens if the resolution does not 
lead to RDF data but to an image? And so on.

>  Of course, the
> traditional text should continue to be available. I just like the way
> in which RDF specifies a piece of data in a way that it can be
> uniquely identified.
>
> I think we should not close this door but keep it as an alternative
> means of giving the data. A such, we don't even need to specify
> anything but "text" in the data elements - a URI is text, too. Though
> it's worth a thought to consider as data value a text-uri-combination.
>   

Not sure if this is the way to go ... applications of the API need to 
know what do do with the return types. And trying to guess whether
"myProfile"
is an URI pointing to a local copy of my foaf file is really hard.

> BTW: RDF is coming out of the research/standards doors and is starting
> to be used by real applications and industry. It's too obvious a
> source of valuable data that any company will be able to pass it by. I
> for one have been considering for my company to use the semantic web
> in addition to website APIs to mine for information that cannot be
> found in any other way.
>
> I can understand your concerns though - we don't want this to be
> *dependent* on the semantic web. But I don't think there is any harm
> in keeping options open.
>   

Agree. My impression is that we will have this discussion mostly when we 
look into specific methods and discuss useful return types - useful for 
hopefully many applications.

Felix
Received on Wednesday, 12 November 2008 00:27:33 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 12 November 2008 00:27:34 GMT