Conference Notes

Supernova '05: Jeff Weiner from Yahoo!

The talk that this piece refers to took place at Supernova 2005 on Tuesday 21st of June around 11.30am.

Now I think I can say with relative certainty that the Chris Anderson talk did not go down terribly well with the assembled backchannellers, who were for the most part regular conference-goers desperately looking for intellectual novelty beyond the stuff that they can read up about in weblogs. The whole weblogging culture has – in my opinion – pretty dramatically changed the conference dynamic. Now it’s not good enough for someone to stand up and talk about the same thing that they’ve been thinking about or doing for the last six months. Many of the audience will be more than familiar with the subject already. They’re going to be looking for novelty. I think this affected the reaction to Google at this year’s ETech a fair amount as well. Everyone had already seen Google Maps. Everyone had cooed over it and celebrated it and got all excited about it weeks before the conference itself. And so when it was presented as the big novelty in the auditorium, accusations of shark-jumping or just being dull flew around the place.

As I said in the previous post, just because people have been over-exposed to things or crave novelty doesn’t really mean that the talk is any less good or pertinent. It doesn’t mean that no progress has been made either (only that people are too flighty and unfocused and impatient to wait until the end of a presentation to find out what’s new).

Nonetheless, there is something specially gratifying about being presented with a talk that you didn’t expect and which actually makes you think in different directions. For me, the first talk that really did that at Supernova was by Jeff Weiner from Yahoo!

He presented an exercise in sizing the amount of knowledge in the world. He didn’t take it too seriously – it was just an attempt to articulate some scale. But it goes a bit like this. If you imagine a typical person and all the things they know about, they probably know a little something about an enormous range of things. There’s a lot of information stuck in their heads, one way or another, whether that be about the best places to eat or the best ways to clean a sink or the best bands of the 1990s or particle physics. If you include all the stuff they know about their family and their neighbourhood and their job then you could probably get someone to write a few sides of A4 paper about a few thousand topics. He postulated maybe 10 pages per subject on average with about 5000 subjects per person.

Now, ignoring overlap for the sake of convenience (and probably because we don’t really seem to generally worry about duplication online anyway), if you multiplied those 50,000 pages by the seven billion people in the world, you’d end up with hundreds of trillians of potential documents full of information. And how many do current search engines think are on the internet? Only eight billion. As a proportion of the total knowledge that we might be able to capture out of people’s heads, that’s a trivial number. Significantly less than one percent. This, says Jeff, is no larger than a rounding error…

This is where it got interesting for me – if there’s all that information in the world, then it’s the responsibility of an organisation like Yahoo! to get that information to the people who might want or need it. And to do that, they need to get people to actively participate. And here’s where he puts up the ‘vision statement’ and it’s even nicer than Google’s. It goes like this:

Enable people to find, use, share and expand all human knowledge

That emphasis on the creative aspect of knowledge – the fact that people make it and that an organisation could be there to help them and to make it easier for them to share it and even to employ it – really appealed to me. The only thing that subsequently spoiled it was when he revealed that you could make the acronym F.U.S.E. out of all the core parts of the enterprise. That was a lurch too far for me…

In terms of demo’s – the bit that stuck in my head but seems to be pretty invisible to most of the people around me was the My Web search interface. Fundamentally, this is no more or less than a tagless version of with one major difference – you can search through your bookmarked pages in full rather than just your tags and descriptions of them – and there’s the full expertise of Yahoo’s search engine to make that searching accurate. I wonder if if it’s too small a conceptual difference from the other bookmarking services to really gain conceptual currency, but it’s certainly more useful than many of them. It may get lost under folksonomies…

A couple of other minor notes from the backchannel around this particular talk – there was general derision for the F.U.S.E. acronym of course, but also considerable irritation about the use of an older woman (normally mother) as the stereotypical naive or relatively naive user. Moreover, there was a bit of a query about the tone of the piece which seemed really aggressive and a certain nervousness expressed for our much-loved colleagues at Flickr when Weiner mentioned their acquisition. Some form of collective intake of breath and tensing happened in that moment. I’ll be interested to know what it means…

As ever, there are more comprehensive notes about this session available, should you be interested: supernova_weiner.txt.

One reply on “Supernova '05: Jeff Weiner from Yahoo!”

considerable irritation about the use of an older woman (normally mother) as the stereotypical naive or relatively naive user

Writing about Eric Raymond’s invocation of ‘Aunt Tillie’ (hereinafter A.T.), John Gruber pointed out back here that – apart from the obvious sexism & ageism – this way of thinking shows a general contempt for users:

“If thereís a glib, nutshell synopsis for why Linux desktop software tends to suck, itís this: Raymond and his ilk have no respect for anyone but
themselves. They have no respect for the fact that UI design is a special talent. […] And, most importantly, they have no respect at all for real
users. The idea that GUI software needs to be designed for ìdumb usersî ó which is Raymondís own term, and an indication of what he really means
when he refers to dear old A.T. ó is completely wrong.

“Great software developers donít design for morons. They design for smart, perceptive people ó people just like themselves. They have profound respect for their users.”

Comments are closed.