Re: Thoughts on @href/@resource completing hanging @rels

Ben Adida wrote:
> Manu Sporny wrote:
>> As for spurious triples in newly created XHTML+RDFa pages, well, there
>> is no such thing.
> 
> I disagree with this statement quite strongly. People won't write
> XHTML+RDFa from scratch. They'll start with XHTML and sprinkle in RDFa
> progressively.

Then I must strongly disagree with your statement which strongly
disagrees with my previous statement :)

We all agree that moving XHTML around is a gigantic pain and should be
avoided at all reasonable costs... however, I have a large sample set of
people proving that folks are quite agreeable to moving their XHTML
around when the benefits outweigh the costs:

Microformats

Additionally, we had to go through and restructure all of our corporate
and service websites when we decided to add Microformats to our website.
Most of the people I know that have added Microformats to their page
(lots) have had to div-span-itize their sites. Most websites generate
the XHTML using PHP (or ASP, or Ruby, or <insert web language du jour
here>), so, this is not as big of a deal as it may seem.

Sure, a couple of our web developers griped that they shouldn't have to
move stuff around to make Microformats work... but they did move the
XHTML around because the benefits far outweighed the cost of doing so.

Besides, as I stated, we never had <div> and <span> tags around the data
on our pages... we had to change the XHTML structure just to enable
semantics on our pages.

Which, like it or not, the vast majority of people adopting Microformats
or RDFa will have to do.

> They will add a hanging @rel, and if that suddenly forces them to
> restructure large chunks of HTML because they can't include plain
> images
> and links within that chunk, that reflects poorly on RDFa as a whole.
> This has already happened with Creative Commons.

I very strongly disagree... if adding a hanging @rel screws things up...
they'll remove that hanging rel before restructuring their XHTML
(because that is the easiest thing to do).

Is it really that difficult to tell people that if they start seeing
wierd triples generated, that they should move their hanging @rel down
to the thing that should be generating the triple?

Do you really think that people won't have tools in their browser (like
Operator) to check to see if they are generating the proper triples?

> That's typically what happens with RDF/XML, btw. Adding a few properties
> is easy, but then you start striping and it gets *really* messy, and
> many people give up.

Seeing the number of people struggling to mark up very complex use cases
using Microformats tells me that people don't give up that easily.
Rather, the people that give up must not have much to gain by marking up
their data using Microformats OR RDFa.

I think that we shouldn't forget that chaining is an advanced RDFa
concept and thus, has some complex issues that come along with it. Those
complex issues are very easy to address:

Either you
  a) don't use hanging @rels, or
  b) understand that @about/@href/@resource completes hanging @rels.

The paragraph above is quite easy to express and understand - meaning it
can be taught quite easily. If we turn this into an education problem
instead of a language problem, we simplify teaching, developing and
using RDFa. It is a good trade-off, IMHO.

-- manu

-- 
Manu Sporny
President/CEO - Digital Bazaar, Inc.
blog: Intro to the Semantic Web in 6 minutes (video)
http://blog.digitalbazaar.com/2007/12/26/semantic-web-intro

Received on Monday, 21 January 2008 20:46:17 UTC