Author: moleitau

Minority report

moleitauLeave a comment

In the Feburary 2003 Journal of Personality and Social Psychology is a little suntin-suntin’ which might be worth looking over:

““The Minority Slowness Effect: Subtle Inhibitions in the Expression of Views Not Shared by Others”
Five studies revealed that people who hold the minority opinion express that opinion less quickly than people who hold the majority opinion. The difference in speed in the expression of the minority and majority opinions grew as the difference in the size of the minority and majority grew. Also, those with the minority view were particularly slow when they assumed the majority to be large, whereas the opposite was true for those with the majority view. The minority slowness effect was not found to be linked to attitude strength, nor was it influenced by anticipated public disclosure of the attitude.”

Slowness in systems is something I’ve been trying to think about for a while, and recent reflection on not-so-smartmobs has reminded me of this. Thing is, nearly everything webby I’ve ever worked on has tried to be as quick, fast, easy and responsive as possible.

The ethnography we had done showed that the processes we are trying to support with our system can typically be ongoing for 2-5 years I.R.L.; and stuff like Robert Axelrod’s “The evolution of cooperation” points to the role of slowness and turn-based systems in reaching concensus-based change [like waiting 4 years before being able to vote for a government… heheh]

Trying to think of networked online systems that are ‘slow’, and so far all I can think of are distributed computing things like Seti@home, or Phil’s Pepysdiary.com. The latter is not so much ‘slow’, but long, if you see what I mean.

Any thoughts?

» Journal of Personality and Social Psychology, Feb :”The Minority Slowness Effect: Subtle Inhibitions in the Expression of Views Not Shared by Others”: John N. Bassili [thanks Fiona for this…]

Maps and politics

moleitauLeave a comment

This is a nice short read on maps and territory:

“The map is simplified to make it legible. In so doing, the author imbues it with his own vision of the world and his own priorities.

Maps are subject to all kinds of manipulation, from the crudest to the most subtle. They are eminently political objects, and governments rightly consider them an effective propaganda tool.”

Compare with Rashmi‘s:

“it is incorrect to think that Recommender Systems cannot have an agenda, or less of an agenda than categorization. Recommender Systems are explicitly designed to encourage people to buy. Often, they are the technique that helps the telemarketer suggest another product to you in a late evening phone call. In contrast browse, or search systems are much more self-directed. Recommender System algorithms are fine tuned for marketing and sales purposes not for helping you discover information. “

We’ve got to make our navigation, search, taxonomy, user-interface – everything as ‘impartial’ as possible, whilst still making it buzz and fizz enough to get people involved and active within the system. We’ve just started our detail design phase, so these thoughts will be at the front of our minds.

» Le Monde Diplomatique: A political look at territory [via Demos Greenhouse

Three Degrees

moleitauLeave a comment

I tried to install the Beta of MSFT’s new IM-on-steroids app, ThreeDegrees last night. I fell at the first hurdle of seeing the swathe of patches and upgrades to WinXP one had to go and download first.

Reknowned software explorer, BetaNaut and fridge-magnet Yoz Grahame got a little further, but not much further. He’s written a funny and informative report back from the frontiers of insidious-installation here:

“I don’t know which of the scenarios I’m imagining is worse: The one where a crazed developer with MS Paint gets that past QA, or the one where the design team achieves group consensus to prove they’re the gang that’s down with the kids.

…the kids have to be down with installing a metric arseload of supporting extras before they can get jiggy with the winking action. This includes MSN Messenger 5.0 and the MS Black Ops P2P Infiltrator. I had a brief bout of swearing when MSNIM 5 started up because it was clearly ignoring my preference to hide the never-used info tabs on the left. Investigation showed I was wrong; it hadn’t so much ignored my preference as removed the option entirely. Clearly, being able to view Expedia travel deals in a 100-pixel-wide buddy list is too important a feature to ever be turned off. “

The Flash demo on the threedegrees website hasn’t really convinced me it’s worth going through all the pain Yoz is reporting. I can’t see any obvious new functionality or ‘delight’ that the thing could deliver, but then again I’m about ten years older than the oldest person they’re trying to target with this thing. Which is depressing as hell in itself!

» Yoz Grahame’s Cheerleader: Three degrees of separation, and rising

Superposition

moleitau

” In the comic or ‘graphic novel’ From Hell by Alan Moore and Eddie Campbell, the creation of the uncanny can be attributed to a perfect interaction of certain thematic elements and of visual procedures, including the position of the camera, the repetition of textual and graphic material and the superposition of imagination and reality.”

»

Postfilter

moleitauLeave a comment

Storing in public a rough note of an ultracrepidatary half-formed thought/position had/held talking with Tony:

Professional craft is a response to scarcity. Where there is no scarcity, there is no need for professional craft. It becomes personal based on subjective drivers. Opinions, points-of-view, and reportage are now non-scarce. In the post-postfilter world, what professional crafts do we look to?

“You can’t handle the truth!!!”

moleitau3 Comments

William over at iSociety posted this last week and it’s been swirling around in my head, but not being that well read or good at philosophical investigation, I’m not getting very far… ‘Ne sutor ultra crepidam’, indeed.

“Recommender systems, reputation systems and blogging networks all demonstrate an American pragmatist approach to truth: most people think x, and the system which consulted them is fair and open, therefore x is true. They suffer from a Pollyanna effect, whereby negative comments play no role; degrees of positive preference determine what’s valid.”

William paints it as an either/or; but our experience is that the play of BBCi search/select-few-human-built taxonomy truth-through-refutation against the googlesphere* / many-anon-humans-decide postivist truth seems to yield a happy solution (for now) – it’s the top-down + bottom-up model we’re shooting for on my current project too.

So here’s the thing: in your opinion, does the googlesphere pollute the taxonomist’s view over time – will it always win? Should we lock all the library scientists in a (luxury) retreat and keep them pure holy fools for our own good??? I really should get round to reading the Surgeon of Crowthorne shouldn’t I.

Or go get some coffee, calm down and shut the hell up.

* Googlesphere [noun]: The post pyra/google blogosphere…

Pearls in the mud.

moleitauLeave a comment

Tom Steinberg at iWire on the brou-ha-ha about blogs, power-laws and Clay‘s part in it all:

“the blogosphere environment actually conspires against the successful evolution of difficult ideas, unless they get programmed into a form of application. This is a flip side to the creativity of the blog world, where the same constraints (i.e. noise and miscommunication) can often lead to serendipity and innovation.”

Not sure I buy everything Tom says there, but it’s a thought-provoker to be sure. Once we’re up-and-running with a decent, active population, we’ve definitely got our own ‘pearls-in-the-mud’ problem on our project. We’ve got a couple of things we think might help.

One is that we will have a strong geographical focus – so that the big national power-law curve of what people are working on becomes many smaller domains. These will have their own zipf curves I’m sure, but more comprehensible and accessible. Themes within these smaller local domains can, and will go TransLocal, which is when things get interesting.

Secondly, these smaller domains will have access to human editors: acting in sherpa-not-censor mode; who can cluster quicker, smarter and cheaper than an algorithm; at least while we are in start-up mode – also giving feedback and encouragement to those building and sailing their ships.

Thirdly – the ‘t’ word. Taxonomy. We have a large, but discrete problem domain, which gives us a large, but discrete taxonomy we can generate. Done well, it will give people structure to build their own ideas around, go translocal and ultimately the ability to improve that structure based on their experience.

We’re trying to get just the right amount of mud for good stuff to happen, but some killer pearl-detection officers and equipment on hand for everyone to enjoy.

Can we have our cake and eat it? The next couple of months will show us.

» iSociety: Scaling Clay Shirky