Re: Comment on RDFa 1.1 Core: Profiles, term mappings, and URIs as literals (ISSUE-39)

Hi Richard, good morning...

On Aug 7, 2010, at 23:39 , Richard Cyganiak wrote:

> Hi Ivan,
> 
> On 6 Aug 2010, at 06:40, Ivan Herman wrote:
>> The issue I have is that a statement like
>> 
>> <http://xmlns.com/foaf/0.1/name> rdfa:term "name" .
>> 
>> is a statement on a resource with a specific URI. And I see two issues with that, modeling wise:
>> 
>> - the goal of the @profile file is _not_ to make statements on resources but to make statements on strings, ie, the way RDFa processors should manipulate strings that are then converted into URI-s
> 
> From a spec writer's or implementer's point of view, you are right: Term mappings are about establishing mappings between different strings. But I think from a user's point of view, the notion that rdfa:term says something about a *resource* (the property or class) is quite intuitive. It establishes a short name for the class or property. So I don't think there is potential for user confusion here.
> 

Not only. It may establish a relationship for any URI or, shall we say, for a URI for a full vocabulary and a string used for prefix. In this sense, it can also be used for any URI, not only for a property or a class (though, I admit, that will be the more frequent use). But if, using your terminology, I say

<http://www.ex.org> rdfa:prefix "pref" .

Then I can say

<span about="pref:"> ...

But I take your point: I may be too, shall we say, pedantic?

> Also, I don't think that the goal of @profile is to make “statements” in the knowledge representation sense. 

Which is of course a long standing discussion... One could say that, according to the current RDF Semantics document, all RDF triples are statements in the knowledge representation sense... But we may want to keep away from philosophy here.

> The goal is to establish prefix mappings and term mappings. RDF triples are simply the data structure for writing down the mappings. For this purpose it does not matter between what kind of brackets or quotes you stick the string -- the information needed for RDFa processing is there in the data structure.
> 
>> - while you may make the statement above in your @profile file, I may say in mine
>> 
>> <http://xmlns.com/foaf/0.1/name> rdfa:term "blabla" .
>> 
>> These are both RDF statements in RDFa files somewhere so, eg, Sindice is perfectly licensed to collect both. Ie, Sindice will suddenly produce a proliferation of statements on the
>> 
>> <http://xmlns.com/foaf/0.1/name>
>> 
>> resource which are definitely not intended and are meaningless for foaf...
> 
> I don't understand why you consider this a problem. For processing any given HTML+RDFa document, only the @profile links contained in this document are used in processing. This is quite explicitly stated in the current draft. Other statements that may or may not exist in other documents are irrelevant.
> 
> I also don't see what Sindice has to do with this. Yes, Sindice can and will collect any and all RDF statements out there, and thus it has a collection full of incorrect and contradictory claims. However it never merges them into a single RDF graph. It always remembers the URL of the document (graph name) where it found any statement. Thus, nothing of the intention or meaning is lost.


Ok for Sindice, I believe you. But not all search engines are built this way, are they? I remember Andreas Harth contacting me some years ago: I had some erroneous statement in my foaf file and it seems to had some ugly consequences on his engine. I admit I do not know the details.

Anyway. As I said above, I may have been a bit too cautious or pedantic on this issue. Could you do a favour for me? DERI has a major concentration of RDF heads around; could you see with other people there if your proposed changes are indeed acceptable for the pedantic web community as a whole? If so, I am happy to accept it and propose the changes to the WG. As an RDFa implementer it would certainly simplify my code...

Thanks

Ivan




> 
>> (As an aside, if we had literals as subjects that we could use those instead of the URI Resource, but that opens a whole lot of other issues:-)
> 
> (Even if we had literals as subjects, I'd still say that a URI should be used here!)
> 
> Best,
> Richard
> 
> 
> 
>> 
>> On Aug 5, 2010, at 12:53 , Richard Cyganiak wrote:
>> 
>>> Hello RDFa WG members,
>>> 
>>> This is a comment on Profiles in the latest RDFa WD [1]. The draft provides a mechanism for establishing term mappings using RDF triples of the following general shape:
>>> 
>>>  ?x rdfa:term "name" .
>>>  ?x rdfa:uri "http://xmlns.com/foaf/0.1/name" .
>>> 
>>> I ask that the mechanism be changed to the following form:
>>> 
>>>  <http://xmlns.com/foaf/0.1/name> rdfa:term "name" .
>>> 
>>> This change should be made for the following reasons:
>>> 
>>> 1. Conciseness. If one triple is enough to say it, there should be one triple.
>>> 
>>> 2. Putting URIs into RDF literals is almost always an anti-pattern.
>>> 
>>> 3. The principle of least surprise.
>>> 
>>> 4. Using a URI simplifies the creation of self-contained profiles that contain a set of term mappings along with labels for the classes and properties, mappings to other vocabularies, presentation hints etc.
>>> 
>>> 5. Using a URI simplifies the extension of existing RDF Schema documents (e.g., the RDF version of the FOAF spec) to RDFa profiles.
>>> 
>>> Best,
>>> Richard
>>> 
>>> [1] http://www.w3.org/TR/2010/WD-rdfa-core-20100803/
>>> 
>>> 
>>> 
>> 
>> 
>> ----
>> Ivan Herman, W3C Semantic Web Activity Lead
>> Home: http://www.w3.org/People/Ivan/
>> mobile: +31-641044153
>> PGP Key: http://www.ivan-herman.net/pgpkey.html
>> FOAF: http://www.ivan-herman.net/foaf.rdf
>> 
>> 
>> 
>> 
>> 


----
Ivan Herman, W3C Semantic Web Activity Lead
Home: http://www.w3.org/People/Ivan/
mobile: +31-641044153
PGP Key: http://www.ivan-herman.net/pgpkey.html
FOAF: http://www.ivan-herman.net/foaf.rdf