[rdfweb-dev] responding to a new wave of anti-RDF sentiment

Danny Ayers danny.ayers at gmail.com
Mon Aug 8 15:11:57 UTC 2005


Hi Jo,

I'd have responded sooner, but missed the post until now - Gmail
filtering is too hard for me ;-)

The negative press issue just shot back up my own list after hearing
the Adam Bosworth MySQL presentation [1] (essentially "RSS is the
future data model for the Web"). He dismisses SW ideas, despite his
own suggestions being pretty close to a watered-down (full of holes)
version of the same thing. The annoyance being his dismissal is based
pretty much entirely on misconceptions (e.g. one big ontology). Very
frustrating.

Anyhow I agree 100% with your points, some random comments:

'why-rdf' - yep, that's probably the best angle for tutorial-like
material. I remember a while back a thread on xml-dev ending with a
line like "ok, I totally understand RDF technology. I just don't see
why I should use it in preference to...", which leads to -

'why not xml?' - I'm not sure how often this can be tackled head-on,
but is probably always a good starting point for RDF-exposure bearing
in mind that the SW stack includes XML and generally syntax is
secondary to good data. And XSLT.

cartoons - yes please!  I reckon anything visually appealing works
wonders. The speech bubble conversation thing someone did ages ago was
pretty good (can't find link, sorry) . Also may take a bit of hunting
down but the Semaview one-page RDF intros (PDFs) were pretty nice.
Then of course the little FOAFsters are a masterpiece.

OPML - bit of a bete noir for me, I can't really comment rationally on
it any more so I just say "fool's gold" whenever it's mentioned (just
as accurate yet a little less ad homininal than "snake oil", don'tya
think?). Hopefully natural selection will get rid of the stuff sooner
than the dinosaurs. Microformats can actually have value, but in
contrast OPML...oh don't get me started...

But going back to 'why-rdf', put in a the context of a tutorial,
alongside the benefits it would be nice to somewhere see corresponding
implementations that demo each point. It's not common, but more than
once (following lengthy discussion) I've seen acknowledgement from an
xxML developer that the ideas of RDF have merit, and that, oh ri-ight,
it isn't ever so complicated after all. Yet despite their education
(heh) their attitudes to their development haven't changed. In
contrast when the same sort of people see something like Flickr or
del.icio.us in action, they're eager to hop on board whatever the
underlying technology.

Definitely a +1 to easy PHP kit being desirable, and I think the
recent port of Redland to MS Win is probably a big plus. In both cases
I'd suggest the packaging is pretty important.

For online demos, I reckon benefits really need to be in your face.
I've had loads of fun in the past few months with SPARQL through
Redland, it just oozes coolness. But compare and contrast the query
form [2] which is the only really visible part of what I've been
working on for those months, with the Longwell view of my blog data
[3], which took about an hour to set up (ok, plus several hours
getting decent data out of WordPress).

See also: "You can't tell people anything" [4].

Cheers,
Danny.

[1] http://www.itconversations.com/shows/detail571.html
[2] http://pragmatron.org/query.html
[3] http://dannyayers.com:8080/longwell
[4] http://www.fudco.com/habitat/archives/000014.html


-- 

http://dannyayers.com



More information about the foaf-dev mailing list