Looking for my Cognitive Surplus

You’ll have come across the stories, talks and interviews about Clay Shirky’s new book: Cognitive Surplus.

I think that all of us get – and recognise – the basic idea. Most of us spend/waste so very much time watching television. That’s typically pretty passive. However, an increasing number of people are doing something different.

We’re online, but not surfing. We’re making. Making videos and blog posts and discussing photos and creating reviews and all sorts of mad stuff. Here’s the man himself, explaining it all:

more »

Making is… Making?

My estwhile colleague, the excellent David Gauntlett, has posted a new video about the work towards his next book Making is Connecting:

more »

Seeking Answers

Google Answers has been closed while Yahoo! Answers goes from strength to strength. The key difference between the two is that Google’s service paid vetted ‘experts’ to produce results, while Yahoo allows anyone to pitch in. The whole thing leaves a lot of questions.

I’m not sure whether the stats prove an uncomplicated victory for social search and crowdsourced problem-solving, for a start. I’ve really no idea which service produces better answers, being one issue. It probably depends on the question. ‘What’s a good Italian restaurant in Cardiff?’ will work well with the Yahoo! model because it has a wider reach. On the other hand, you might not want to trust folk wisdom for a solution to matters that require a specialised knowledge.

It does show that a free-for-all, give-and-take knowledge source is very addictive and, presumably, helpful enough. Involving people like Stephen Hawking and Oprah Winfrey bought Yahoo! a vital share of attention Google never bothered with. Also, as Brady Forrest points out, Yahoo!’s model could scale organically, while Google’s required the recruitment and vetting of answerers, a time-consuming and distracting business.

Is this victory analagous to what will happen in the battle between the Wikipedia and the Britannica? It seems very similar on face value. Not entirely, though, since their business models are different: Wikipedia survives on charitable donations and drubbing the opposition when it comes to traffic is not nearly as helpful as it has been to Yahoo!

[I interviewed Steven Taylor, RVP of Yahoo! UK here, back in August and he talked a little about the Answers service]

Wisdom and Intelligence

One of the cornerstones of most definitions of Web 2.0 is the idea of the Wisdom of Crowds. In Tim O’Reilly’s seminal essay on the subject, he talks about the blogosphere being an example of this:

If it were merely an amplifier, blogging would be uninteresting. But like Wikipedia, blogging harnesses collective intelligence as a kind of filter. What James Suriowecki calls “the wisdom of crowds” comes into play, and much as PageRank produces better results than analysis of any individual document, the collective attention of the blogosphere selects for value.

Other examples which are sometimes cited include digg, Yahoo! Answers, Wikipedia and del.icio.us. People come together to solve problems and their combined effort produces better results than an individual editor or news team could manage.

However, we’re actually smudging together two contrasting decision-making mechanisms here. Henry Jenkins points out in a post related to game design that there’s a significant difference between Pierre Levy‘s idea of Collective Intelligence and James Surowiecki’s topic, The Wisdom of Crowds.

The Wisdom of Crowds emerges when data from a number of sources is aggregated. The people contributing need to be acting autonomously according to the best of their ability and in competition with others. The famous example is guessing the weight of the prize bull: the average of people’s guesses turns out to be the correct answer.

Collective Intelligence, on the other hand, emerges through deliberation, where people share, alter and evaluate other’s contributions to arrive at common ground.

As Jenkins notes, Wikipedia is much closer to this second model, Collective Intelligence, than the Wisdom of Crowds approach that finds the mathematical mean of all the suggested ‘answers’. The same would be true of Yahoo! Answers and del.icio.us, and indeed of most Web 2.0 applications that revolve around a community approach.

The Wisdom of Crowds model does in some ways apply, however, to things like the digg front page**, flickr interestingness and Google PageRank, which are algorithmically determined based on the combined anonymous and competitive input of many people.

It isn’t really a question of one of these models being better than the other, Jenkins concludes. It’s more that we’re not going to get very far unless we realise that they are two different things:

Both “collective intelligence” and “the wisdom of crowds” offer productive models for game design but we will get nowhere if we confuse the two. They represent very different accounts for knowledge production in the digital age and they will result in very different design choices.

I’d contend that the approach chosen by an application designer very much depends on the nature of the problem that is being addressed. Both could be correct depending on the situation, and probably one approach would be more sensible than the other for any given application. Completely anonymous postings to Wikipedia with no editing hierarchy whatsoever probably wouldn’t be such a great plan, though it would bring it closer to the wisdom of crowds model. On the other hand, the collective intelligence method of measured deliberation and discussion about which stories to put on the front page of digg or which sites should appear at the top of Google searches probably wouldn’t work out too well either.

[**Actually, digg is interesting in this regard. The submission of stories is not anonymous, nor is the voting. This has led to lots of accusations of bloc voting, allegations of a self-reinforcing elite of top diggers, and adjustments to the promotion algorithm to try to prevent this. It is a strange amalgam of social community and wise-crowds news aggregator. The owners (and presumably enough of the users) want it that way. If the owners didn't want the social community aspect, and the problems that has created, they'd remove all mention of user names and make voting anonymous. It's my belief that the gaming aspect to digg is entirely intentional and part of what appears to make it so addictive to its fans.]

From Big Cats to Barthes

I’ve just been checking out Wikibooks, a project of the Wikimedia foundation that aims to create free books. Like Wikipedia, anybody can contribute to the books either by adding new material or editing existing books. Those books that are complete or voted ‘good enough’ are also available as PDF documents and even print editions created through Lulu.

A branch of the project is devoted to children’s books, WikiJunior, where you’ll find books about things like the solar system, big cats and the Kings and Queens of England. There are also things like A-level and GCSE textbooks, lots of computer science stuff and hundred of others. The community votes on which new books to develop, though looking at the history of many pages, a lot of the books are the creation of one enthusiast with corrections and additions from others. Wikibooks appears to be a considerably more sedate and good-natured bunch than the wikipedia crowd, with little evidence of the edit-wars, vandalism and obsessive nitpicking that characterises some of the more controversial wikipedia items. Perhaps this is because the project is less well-known, with a smaller community. Perhaps it’s because books are typically big things that require a lot of work and so command some respect.

The aims of the project, like wikipedia, are to democratise and spread knowledge and information. Traditional publishers, say the organisers, fail to recognise merit because their business models rely on creating best-sellers and so they’re risk-averse:

Traditional publishing houses make the bulk of their income from re-issues of classic books, new books by authors with long track records, or celebrities who are famous in their own right. The chances of a truly good new work being published solely on the basis of merit skyrocket when you overturn the traditional business model and tap the wellspring of new talent out there using the ‘net.

With this project we have reached a crossroad between the books of yesterday, and the encyclopedia of everything for tomorrow. Simply by reading this book and telling your friends, you have advanced the cause of free access to information and of democratizing the field of publishing.

There are issues, of course. I read through the PDF version of Big Cats, which is deemed complete, available in print-format and on its way to a second edition. The information it contains appears to be accurate, well-researched and carefully written to suit a young audience. Unfortunately, though, it was a bit odd. There’s lots of half-finished edits, changes in tone and register and the layout is pretty basic. Ultimately, I wouldn’t buy it.

So what does that mean? If one of the most highly developed books available is still not good enough, is the project a failure? This is the sort of charge that’s levelled at Wikipedia: it contains incorrect information, so it’s no good.

That’s not really the right way to look at wiki projects, though. The point of wikis, in my view, is that they are always works in progress. That’s their strength and their weakness. Unlike print editions, new information can be added at any time. When Pluto ceased to be classified as a planet, thousands of books were suddenly out-of-date; Wikipedia was immediately up-to-date.

This philosophy intersects strangely with the idea of books, though. The idea of a book has connotations of completeness, correctness and authority. (Correct in the sense that we don’t expect spelling mistakes, etc.) The idea of an unfinished book is paradoxical – if it’s not complete, then in some senses it’s not yet a book.

What you’re looking at when you read pretty much any wiki project is not something analagous to anything produced on a printing press. It is a palimpsest. The Romans wrote on wax tablets that could be re-used. Medieval monks wrote on vellum, a form of calf leather. If they needed new paper or made a mistake, they could peel off the current layer and write on it again. Modern scholars use ultraviolet and multispectral imaging to try to decipher the history of the page. Wikis lay this process bare. The ghosts of previous versions, previous authors, can be seen in the crookedness of the edits; its history page provides an X-Ray of its genesis. Portents of its future are on the discussion pages: some of these prophecies will come to pass while others will be forgotten.

WikiBooks might thus be viewed as the ultimate in post-modern writing. Derrida and Barthes talked about books having a ‘magic tablet’ quality. That there were other meanings and expressions hiding beneath the surface:

The Palimpsest introduces the idea of erasure as part of a layering process. There can be a fluid relationship between these layers. Texts and erasures are superimposed to bring about other texts or erasures. A new erasure creates text; a new text creates erasure.

The “oddness” of Wikibooks is only apparent in the print and PDF versions. To publish them in these formats runs directly against the nature of its progenitor. Wiki pages are liquid; they exist at this moment in time, and they are always moving through time as edits and changes accrete continually. When those moments are frozen, captured into a snapshot, it’s like taking a still from a film. We know that the future and past of that picture already exists, but we can only guess at it.

(found through Derek Wenmoth‘s fab education blog)

A Win for Wikis

A new report says Wikis are more important than social networks when it comes to business technology buyers. The report, from Knowledge Storm and Universal McCann, is available here – registration required. It’s also a cut-and-paste protected PDF, the devil’s own file format.

But basically, it says that, of 5300 participants:

77% of these buyers have little or no experience of with social networks. The report suggests that these people are still using the web to “get” information and that the “giving back” part of social networking might make them uncomfortable.

On the other hand, 86% of respondents said they were familiar with wikis, and more than 50% are weekly wiki visitors. 52% stated that wikis influenced their purchasing decisions.

Interestingly, the only wiki mentioned in the report as an example was Wikipedia, highlighting that the importance and influence of the online encyclopaedia goes far beyond its apparent status as a neutral reference tool. While it (thankfully) doesn’t carry product reviews, one can only presume that articles covering competing technologies and IT strategies are extremely influential.

Two recent social networks have been launched specifically for techies: Aggreg8 from Microsoft and, closer to home, the new ZDNet.co.uk communities endeavour. Would it have been a better idea to produce wiki-inspired features instead? Well, maybe, except, like most of us, IT buyers are most likely to be passive rather than active users: only 6% said they regularly contributed to wikis.