W3C home > Mailing lists > Public > public-rdfa-wg@w3.org > March 2010

Re: A rose by any other name is just as thorny...

From: Mark Birbeck <mark.birbeck@webbackplane.com>
Date: Mon, 29 Mar 2010 13:44:16 +0100
Message-ID: <640dd5061003290544i3a59ac5bv4fe2482d4c2a47b2@mail.gmail.com>
To: Ivan Herman <ivan@w3.org>
Cc: Shane McCarron <shane@aptest.com>, Toby Inkster <tai@g5n.co.uk>, Manu Sporny <msporny@digitalbazaar.com>, RDFa WG <public-rdfa-wg@w3.org>
Hi Ivan/Manu/Toby/Shane,

With respect, I feel we're going round in circles here......

Many of the problems you're trying to address are exactly the reason
we have the solution that we currently have.

For example, it's easy to say that we want the @rel/@rev values to be
mapped in a certain way, but how *exactly* does that happen? The
solution we came up with -- as of course, you know -- was to say that
there are tokens that are processed first, before any other CURIE
processing takes place.

Now, people seem to be saying that they want to move away from that.
Fair enough, I guess...but don't be surprised if the problems that pop
up when you move away from this approach are the very problems that
this technique aimed to solve!

When we discussed this issue we did consider the alternatives,
including making the @rel/@rev values be suffixes with a default
namespace prefix. In fact, I remember being really in favour of this
technique, but it was rejected by the group (rightly), due to a number
of problems:

1. The upper-case/lower-case issue.

2. What does @rel="next" in the scope of a new default prefix mean?
(And if it doesn't mean xhv:next then we're no longer
backwards-compatible.)

3. Junk triples.

I may have missed seeing #2 on the list recently, but #1 and #3 are
definitely back from the dead.

Anyway, forgive me if I'm missing something, but at the moment I'm
having difficulty seeing quite where this whole discussion is headed.
And worse, I'm getting an unnerving feeling that nearly anything is up
for redefinition -- even things that we did for good reason, many
moons ago.

(Of course 'unnerving feelings' are not always based on reality, so
feel free to tell me I needn't worry. ;))

By the way, the argument for using a default profile was so that other
languages could override it.

Regards,

Mark

--
Mark Birbeck, webBackplane

mark.birbeck@webBackplane.com

http://webBackplane.com/mark-birbeck

webBackplane is a trading name of Backplane Ltd. (company number
05972288, registered office: 2nd Floor, 69/85 Tabernacle Street,
London, EC2A 4RR)


On Mon, Mar 29, 2010 at 11:31 AM, Ivan Herman <ivan@w3.org> wrote:
>
> On Mar 28, 2010, at 20:19 , Shane McCarron wrote:
>
>> Actually, I don't like this at all.  I don't want to lose the ability to have mixed case relationships - it would violate the specific use case we are trying to satisfy (make foaf the default vocabulary in a snippet).
>
> Exactly. If I cannot say (Toby's example)
>
> <a vocab= "http://www.w3.org/2000/01/rdf-schema#" rel="seeAlso" ...>
>
> than this is fairly bad. To come back to one of the (granted, very RDF oriented) self-documenting OWL vocabulary definition, ie, where an OWL vocabulary is defined in terms of RDFa to yield both the RDF code and its documentation in one place, a very natural way of doing that would be to use @vocab for the owl terms. But OWL is full of camel case terms. In general, using camel case terms in RDF vocabularies is very widespread.
>
> Ie: restricting to lower case in @rel/@rev is not a good idea...:-(
>
>>
>>
>> Maybe this was said already - I don't remember.  Can't we just say that references to the XHTML vocabulary are mapped to lower case and any other vocabularies are not?  I know it is sort of a hack, but it would mean that we don't need to to any remote retrieval to know if the vocab is case insensitive or not.
>
> It is a hack. It is even an ugly hack:-) But if the case insensitiveness of the XHTML case is the _only_ reason why we would use a default @profile for (X)HTML, we could do something like that. (We still have the issue of junk triples, though...)
>
> But what this should say is that 'if the @vocab value is '....xhtml', then keywords are case insensitive for @rel/@rev. ie, we would not still not list the keywords in the document, the hack would be @vocab value based, so to say, right?
>
> Ivan
>
>>
>> Toby Inkster wrote:
>>> On Sat, 2010-03-27 at 21:29 -0400, Manu Sporny wrote:
>>>
>>>
>>>> We may want to say that rel/rev are always mapped to lower-case since
>>>> they are legacy attributes. Everything else is case-sensitive.
>>>>
>>>>
>>>
>>> Assuming we're just talking about rel/rev values that do not contain a
>>> colon, I could support that. It does break e.g.
>>>
>>> <a vocab=
>>> "http://xmlns.com/foaf/0.1/"
>>>  rel="primaryTopic" ...>
>>>
>>> and
>>>
>>> <a vocab=
>>> "http://www.w3.org/2000/01/rdf-schema#"
>>>  rel="seeAlso" ...>
>>>
>>> though.
>>>
>>>
>>>
>>
>> --
>> Shane P. McCarron                          Phone: +1 763 786-8160 x120
>> Managing Director                            Fax: +1 763 786-8180
>> ApTest Minnesota                            Inet:
>> shane@aptest.com
>>
>>
>>
>
>
> ----
> 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
>
>
>
>
>
>
Received on Monday, 29 March 2010 12:44:54 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 27 April 2012 04:55:06 GMT