I’ve already moaned, several times, about the difficulty in explaining what the semantic web is, let alone what it offers EU affairs. Now, via @jangles via Lee Hopkins (from my home town, Adelaide, Australia), a terrific video by Kate Ray:
I hit pause at about 3 minutes in to start this post, so now I’m going back to watch it. As I do, I’ll pull out some quotes and add some comments. But watch it first. Trust me – it’s worth the quarter hour. I’ll go back in and add the names to the quotes later.
Problem: so much stuff online
“If I was going to start a news business tomorrow, I would design [one] designed to produce not one new bit of news … instead, to aggregate news for individuals in ways that matter to them“
That’s from Clay Shirky (see Creating Trust by Example), and the aggregation he’s describing is exactly what I was talking about when describing how a network of national-level curators could help build the European online public space. What Clay doesn’t answer is A. Rentibisch’s recent question about the business model of such a news business.
Now the quantity of online information is only getting bigger, Google won’t scale to it, and doesn’t actually answer people’s questions anyway, so we’re left with handling our information as best we can:
“Now I gotta label my [Gmail] labels … We are always going to be filtering the filters that filter our filters … that filter our filters. ”
Enter the semantic web (& its critics)
The video then goes on to explain what the semantic web offers:
“extra information to help you with the meaning of the information”
… before pointing out the obvious problem:
“Does the world make sense [ontology], or do we make sense of the world [tagcloud]?”
… and moving on to the schisms developing in the field, between the formal, ontology-based approach and the ‘scruffy, sloppy’ end, where
“a little structure goes a long way if you combine it with a human being”.
And there you have why a search engine is not, by itself, going to build the European online public space. Whether you call them curators or newsmasters, you need people, because “the only group capable of making sense of everything is everyone”.
Moreover, scruffy and bottom-up are more feasible, as well as being in line with the way things develop online.
Obvious benefits for Europe
But that’s not to say that some top-down projects – designed to publish ontology-linked opendata for a specific purpose, and develop the apps to read and use it – should not happen.
“Once we have good [semantic] content out there we can build systems that can reason on it and solve problems, answer complicated questions, make amazing discoveries and linkages between things …”
When I first heard Tim Berners-Lee talk about this I was working in DG INFSO, and so looking at how online communications could help build the European Research Area.
It seemed blindingly obvious that the EC’s and Member States’ research programmes should publish their programmes, projects and research results in semantic web format – there are literally too many benefits to enumerate here, so look at my first post on the subject.
One can imagine similar benefits for any other Commission DG – from helping businesses expand across the Single Market to spotting social trends and tracking environmental pollution.
I simply cannot think of an EU policy or programme that would not be massively helped by having national data flow more easily over Europe’s national, cultural and linguistic hurdles.
And that, of course, is what the semantic web can do.
Priming the pump
The critics base their issues on the feasibility of rolling out the semantic web the way Berners-Lee sees it given that it has to be added on top of the amazingly decentralised, bottom-up Web we already have. They’re probably right – it’s a classic tale of the chicken and the egg.
But chicken and egg problems can be overcome by pump-priming.
Moreover, reaping the above benefits for the EU is mainly a matter of acting on public data, owned and managed by a handful of national and EU authorities. That’s feasible – semantic web formats are open standards, after all, so publishing public data using them isn’t a question of herding cats.
And who know? Imagine if the EU and the national authorities did put their data out there in semantic form. Now imagine that entrepreneurs used it to show what semantic web applications can do. Maybe that would kick-start something. Europe’s done it before (got a GSM?).
Even if it doesn’t, the benefits to specific EU policies and programmes would be immense. But it’s in the nature of the web to surprise us, and it’ll be the applications one cannot imagine which will really change things:
“[the technologists building and arguing over it] are not going to be able to imagine what things people will be able to do with it. I can’t. If we build everything I imagine, we’ve failed” – Tim Berners-Lee, inventor of the World Wide Web.