Aggregation, the semantic web, and The Daily Me
It was many years ago, long even before the advent of Web 2.0, that I remember first hearing Internet pundits telling us that soon we would be able to subscribe to The Daily Me, a virtual newspaper delivered to our computer screen and customized exactly to our individual requirements. Don’t like sports? There’s no sports page. Are you a fan of the team MODO Hockey in the Swedish Premier League? Here’s the latest scores and highlights from Örnsköldsvik, in a sidebar right on your front page. Are you interested in art forgery, Antarctic exploration, the future of the Canadian potash industry? No matter how specialized your interests, The Daily Me would find the breaking news on your topics and put them right there front and centre for you to read in the order you want to read them.
Didn’t quite work out that way. Instead, we got RSS feeds at first, which I could never get enthusiastic about. The first browser I had with built-in RSS aggregation I think was Apple Computer’s Safari, which came preconfigured with (among others) a feed for the BBC: the first time I clicked on the RSS menu, it told me I had something like 10,000 unread BBC news stories. I was sufficiently intimidated I never opened that particular menu again.
The only hope I can hold out that The Daily Me will ever become reality is if the semantic web ever gets some traction. As far as I can tell, the aggregators I’ve seen so far just replace the experience of taking your sippy cup from one fire hose to another, with the experience of directing all the fire hoses on you simultaneously: equally unsatisfying, if perhaps more efficient. What I would like to see in an aggregator is something that would open each channel, filter whatever is coming through that I’m interested in and discard the rest. For bonus marks, it would then combine related items into threads, sort of the way that Google News does. But for this to be possible, the incoming content would have to be tagged in a semantically meaningful way. Not necessary meaningful to me — I don’t need to know how to read MARC records to use an OPAC — but meaningful to the aggregator.
The semantic web as envisioned by Tim Berners-Lee and his followers may still be many years away; it may end up being the flying car of the information age (that is, a futurist’s prediction that proves impractical in the real world). But even a partial implementation standards for machine-to-machine exchange of semantic markup would go a long way to making a better aggregator, not just better personal aggregators for web-skeptics like me, but for the kinds of professional-grade aggregators that information organizations might employ as the keystone of the web portals of the future.