- From: Dan Brickley <danbri@w3.org>
- Date: Fri, 4 Jun 2004 15:00:11 -0400
- To: John Black <JohnBlack@deltek.com>
- Cc: Bijan Parsia <bparsia@isr.umd.edu>, public-sw-meaning@w3.org
> > > These are shock-jock type of arguments > > > > Huh? > > > Shock jocks make their ideas feel profoundly important by > attaching other highly emotionally charged (shocking) ideas to > them. Our ideas are, doubtless, profoundly pedestrian. But the issue we're dealing with (communication using next-generation Web technology) is a profoundly important one. That importance is sometimes illustrated by connecting it with things that matter to people's every day lives. Like being happy, sad, rich, poor, oppressed, misrepresented, misunderstood. > The examples in Dan's and Jeremy's emails have that feel to > me - anti-homosexual bias, racism, but so does totalitarianism. I chose emotive examples on purpose: it was the point of the exercise, even. All the while that everyone agrees about everything, we can be utterly naive about RDF. All the while we agree to disagree by disputing the facts but completely sharing the vocabularies within which those facts are expressed, we have it pretty easy too. The problems come when the terms used to express claims about the world are disputed terms. It shouldn't be suprising that terms are more contested when they describe areas of disagreement and emotive debate. I could have picked less emotive terms (eg. relating to definitions of employment, unemployment) which are nevertheless politically charged and contested. I could also have picked more emotionally charged topics and terms, eg. notions of alive/living/etc in the context of debates around abortion policy. Or the naive definitions of 'terrorist' we see floating around in DAML+OIL or OWL, as if encoding in a formal language automatically brought with it credibility and intellectual rigour. Or perhaps we could be discussing the definition of "torture" as "an act by which severe pain or suffering, whether physical or mental, is intentionally inflicted on a person"[1], investigating whether organisations which don't buy into a common definition are nevertheless capable of communicating using RDF/OWL technology. One person's torture is another's "rough handling", I'm told. I preferred to stick with the all-time classic topic of Sex. There are vastly more topical and controversial themes, very relevant to our work, that we could have focussed on instead. We may be discussing emotive issues, but it is the corner cases that test the technology; I don't think we can stick to areas where things are controversy-free without missing the point. For areas of discussion where terminology isn't politicised, RDF/OWL pretty much works out of the box. For the rest, we still have some work to do. Considering such cases doesn't make us 'shock jocks', though given the emotional baggage that comes with the territory, it is probably worth reminding folks to be extra careful with their examples. The Web is the world's most important system for publishing and sharing documents and data, and it needs to work for everyone, including parties who are mutually antagonistic. RDF/OWL needs to be deployable in a world where language and terminology is politicised, because that's the kind of world we live in. Incidentally, the RDF design inherits from PICS, W3C's earlier "Platform for Internet Content Selection". The 1996 note on "Using PICS Well"[2] is worth bearing in mind when we think of the social impact of our technical work. The PICS design decentralised the creation of content rating vocabularies in part through a recognition that rating and classification schemes can embody contested, controversial worldviews. RDF's design differs from PICS' in that a single instance document can now draw upon terms defined in several independently created vocabularies. By making this possible, we don't necessarily make it wise. (I argued a variant on this in [3], dunno if slides are any use on their own though). My understanding of what we're trying to do here (public-sw-meaning) is that we're probing at the different social and technical notions of 'shared meaning' that get tangled up when such systems are deployed in the wild. cheers, Dan [1] http://www.amnesty.org.uk/torture/definition.shtml citing the UN Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, http://www.unhchr.ch/html/menu3/b/h_cat39.htm [2] http://www.w3.org/TR/NOTE-PICS-Statement [3] http://www.w3.org/2001/sw/Europe/talks/20040309-danbri/
Received on Friday, 4 June 2004 15:00:11 UTC