Re: Discussion with Ian and Henri about HTML5+RDFa (part 2/2)

Hi Manu,

first of all, thanks for forwarding the discussion to the list!

Just my initial and personal reaction on Henri's point of CURIE. I can
see the validity of his cut-and-paste issue. My concern, however, is
that I would not like to see the current RDFa design break. Indeed,
whether we like it or not, some of the URIs that do come up in practice
are ugly and error prone to write; if one is forced to write full URI-s
everywhere again and again, than this is less than user-friendly. (I
have burned my finger myself with the usage of uri-s in @about, for
example.) CURIE-s are an effective answer to that issue, too.

Some technical thoughts that might help in moving forward

- I am not sure what the thoughts of the RDFa task force were on the
matter but I always regarded the @prefix solution as being conceptually
equivalent to @xmlns in the sense that @prefix can be used on any
element (thereby affecting the subtree). Ie, if an RDFa code sniplet is
to be cut and paste (the CC case is a typical example) than by putting
the @prefix on the enclosing <span>, for example, the issue can be
handled. _I know_ this does not answer all concerns, but it does handle
many possible cases.

- Just as we have safe curies, we could have a special syntax for full
URI-s if one is concerned about that. Something like

<... property="<URI>" ...>

much like URIs are denoted in a number of RDF related serialization,
like turtle or various OWL syntaxes (and using CURIE-s otherwise). It
would require very little adjustments for existing parsers/distillers,
and would be an add-on to current syntax. To avoid confusion, I would
also accept the '<...>' syntax in the values of @about, though it is not
really necessary there.

Just my 2 cents...

Ivan

Manu Sporny wrote:
> Discussion with Henri Sivonen
> -----------------------------
> 
> Henri had two major disagreements with RDFa as it stands right now.
> 
> CURIEs
> ------
> 
> Henri stated that the use of xmlns: will break the DOM Consistency
> Principle between XHTML and HTML. I let him know about @prefix, which
> would address the DOM Consistency Principle issue, but would still not
> be good enough in his mind. He is most concerned with HTML authors
> cutting and pasting snippets of RDFa that they believe to contain
> triples, but when pasted, fail to generate any because the prefix
> mappings were declared at the top of a page.
> 
> He was also concerned that authors would get frustrated when their
> cut/paste RDFa did not produce the same triples that they saw on website
> X... or worse, they wouldn't check it and would see no benefit by
> embedding RDFa in their website.
> 
> To address this issue, he (and Dan Brickley) believe that allowing
> @typeof, @property, @datatype and @rel/@rev to accept URIs would address
> this issue with the people in WHATWG that don't like CURIEs. They
> believe CURIEs are too fragile as implemented and it would be better to
> specify all URIs fully instead of providing a prefixing mechanism.
> 
> Non-expert usage and RDFa Education
> -----------------------------------
> 
> Education about RDFa would also be an issue with the majority of web
> authors who don't care about web semantics and just want to get their
> page operational.
> 
> He was concerned that authors who once used rel="license" are now being
> asked to embed more complex metadata (such as Creative Commons licenses)
> without truly grasping the subtleties of doing so (attaching them to the
> wrong subject). This is an issue because bad semantic markup doesn't
> generate the same sort of jarring page display issue that bad HTML
> markup does.
> 
> The problem is with less-than-guru web authors who don't necessarily
> care about web semantics and thus generate bad semantic data out of
> ignorance. The common mis-use of @rev was cited as one possible outcome
> - mis-used so badly that it is commonly not trusted by search engines.
> 
> Vocabulary Scalability
> ----------------------
> 
> He also did not believe in follow-your-nose to be a very useful concept
> and noted that even if we continue down the @prefix and
> Microformats-like RDFa markup routes, with special tokens/reserved words
> specified in a separate file, it could cause a scalability issue when
> vocabularies find themselves quite popular. The example of the W3C
> serving up many, many gigabytes of the same HTML4.01 DTD every day was
> cited as an example of what happens when your "vocabulary" becomes popular.
> 
> He was concerned that by requiring parsers to load references from a
> remote file that one would either put the burden on web authors to stash
> those files on their web servers or put the burden on vocabulary authors
> to ensure that their vocabulary document can be transmitted millions of
> times a month.
> 
> He cited that using something of the form of a pseudo-namespaced
> "foaf-foo" where each token was specified in a spec somewhere, but there
> was no way to follow your nose to it, or validate against it, would
> solve the "failure-due-to-popularity" issue.
> 
> Henri was certainly sympathetic to embedding semantics in HTML for
> everyone that needed the functionality (not just the 80% that
> Microformats addresses) in HTML. He believes that removing CURIEs would
> go a long way towards addressing his concern with the way RDFa is
> currently implemented.
> 
> -- manu
> 

-- 

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