Unfortunately, I was unable to attend the conference on the Saturday, with speakers such as Jack Straw, Doreen Lawrence and Tony Benn – although Jason of Cosmodaddy did, and liveblogged it. But I did go along to the AGM on Friday night instead of enjoying the pub like normal. I Tweeted some of it but here’s a fuller recap of events and my thoughts.

I’m sad to say that like last year’s AGM, a lot of the proceedings were shambolic. Louise Christian, the Chair of Liberty, may be an excellent human rights lawyer but in my view she made a terrible chair of the AGM. Votes to approve the auditors and a nomination to the appeals board were conducted without a call for votes against or abstentions – a numerous show of hands for the motion was taken as unanimity. The results of the elections to Liberty’s council were announced simply by announcing the names of the winners, with no record of the count. When challenged on this, Louise said it was to avoid ‘embarrassment’ to those who got a low number of votes (can you imagine if we did the same for MPs?). There were numerous protests from the floor on the way the AGM was conducted, but time and time again the chair attempted to close the discussion and ‘move on’, even though dissenters still had hands raised and wanted to speak. One member spoke of how it resembled ‘railroading’ and I’m inclined to agree.

Had this been in any other democratically run organisation, I’m sure Liberty’s leadership would have condemned it as opaque and anti-democratic. Given the recent furore surrounding MP’s expenses and lack of transparency, a human rights organisation such as Liberty must have an exemplar level of transparency and proper conduct throughout – not just the minimum, but the very highest standard must be adhered to. And this was sadly lacking from the chair. On one motion (on presumed consent for organ donation) after the vote had been taken, she voiced her personal approval that the motion had been turned down – in my view, conduct not befitting the chair, who should remain neutral both during and after such debates. The fact that the AGM was squeezed into two and a half hours (last year, it was held on a Saturday so we had most of a day to deliberate) so that delegates could enjoy drinks at the end, meant that the chair continually sought to expedite the meeting so we could meet the 9pm deadline. It gave it a rushed feel, as if we were just going through the motions.

Disappointing. But enough of that – there was still plenty of interesting debate from the floor and ordinary membership. Seven motions were presented in total and there was plenty of debate about them – particularly two issues of controversy. One was to what degree people should be denied employment or union membership on their political beliefs or associations they join, and the other was on whether Liberty should endorse ‘presumed consent’ for organ donation.

The first is a tricky issue to pick apart – BNP members are forbidden from becoming police or civil servants, and I’d imagine most well-thinking people would support it. But then again, the same line of reasoning was used by anti-union employers discriminating against left-wing union workers. Some good points were raised in the meeting – someone pointed out how it’s easy to defend trade unionists but much harder to stomach defending fascists, even if the rights they enjoy are universal. The original motion did allow for people to be denied employment in “exceptional” circumstances. I would argue that being an instrument of the state’s power means you have to be bound by tighter rules than a normal employer would; with the police’s commitment to racial equality enshrined in law, and how being a racist means you cannot possibly abide by it, I’d see that as an exceptional circumstance enough to deny a BNP member a job as a police officer. But it has to be done on a case-by-case basis, and the burden of proof has to be on the employer, not the employee.

The motion in the end didn’t pass – an amended version by a trade union delegate did; although I agreed with the sentiment, I thought it was wider-ranging and had much less bite than the original, moving the focus onto unions’ right to bar members because of their beliefs rather than questioning the actions of employers. But the debate was interesting, informed and above all passionate without getting ill-tempered, a much more sober and level affair than the usual emotive bullshit in politics.

The other motion of interest was that of whether presumed consent can be taken for organ donation. There are all kinds of conflicting rights here – the right to life and the right to dignity for the recipients, coupled with the putative donor’s right to one’s own body and religious beliefs. Even with a guarantee of opt-out if you did not want to, and opt-out for the relatives of the deceased wasn’t enough. There are some interesting philosophical dilemmas here – as rights are described as human rights, when you die and cease to be a human, what rights does your corpse inherit? Isn’t ‘presumed consent’ a contradiction in terms? As donated organs are a gift, does turning them into a compulsion destroy the gift value? Should we even care if it means more people will live?

Me? I was in a minority that did endorse presumed consent; growing up with my mother working as a nurse for kidney patients on dialysis showed me how horrible a life without working vital organs could be, and their right to life and dignity comes should be given priority over our sensitivities about corpses; I feel that by having an opt-out system for organ donation we reduce the demand for black market organs and help strengthen the rights of the living forced to donate. But I was strongly moved by some of the arguments made (particularly on the rights to the body in reference to slavery, and torture) and it was a difficult decision to stick by and made me question myself.

Being given crises of conscience may seem an odd reason to be a proud member of Liberty, but I am. And even though all of us have common cause and beliefs there was still healthy levels of debate and dissent within the membership – as well there should be for any organisation in a democracy. Even if the chair of the meeting didn’t seem to concerned about due process or dissent, the membership did, and I was proud to be among them.

So go out and vote. Haven’t got your polling card with you, or it got lost in the post? Doesn’t matter – as long as you’re on the electoral roll you can just turn up at the polling station and vote by giving your name & address. Don’t know where the polling station is? Your local council’s website should have a list – check out Directgov’s directory of councils to find yours. Not sure who to vote for? Nosemonkey has an excellent roundup of resources to make sure your vote is informed, including Votematch and EUProfiler (I thought the former was better, for what it’s worth), the list of individual party manifestos and the manifestos of the party groupings within the European Parliament.

That’s it. No excuses now. Go out and vote. And get your mum to go and vote as well while you’re at it. Thanks.

Plenty of the sessions of Thinking Digital weren’t really that digital, to be honest. But there was a distinct theme running through several of them – the irrational, emotional and sensual sides of ourselves, and the struggle to rationalise and co-opt them.

First up – Michael Shermer, the editor of Skeptic magazine. While his politics are a little wonky (including a slightly bizarre rant about the world financial system), the majority of his talk was an interesting discussion of what makes people believe in the paranormal, in conspiracies, aliens, the Illuminati, Holocaust denial and all the rest. It’s sorely tempting to dismiss them as mere cranks, but Shermer did an admirably level-headed job of using what we know about the brain and evolutionary explanation.

What he calls “patternicity” (but I think is more usually called “apophenia“, one of my favourite words), the tendency to patterns in random noise is one tenet, which he explained is a side-effect of our ability to spot threats and predators. Then there’s “agenticity” (which sort-of ties in with anthropocentrism, the cognitive bias that humans (or some other species) must be the cause of observed phenomena, a side-effect of the development of empathy and a theory of mind. The two combine to make it easy for a belief that random events are actually orchestrated by powerful unseen entities. It’s a compelling explanation, although it still covers perhaps too easy and irrelevant a target – only a tiny minority of people holding irrational viewpoints are utter wingnuts. The more dangerous kind of irrational person, the mainstream kind like climate-change deniers, armchair general liberal interventionists, subprime derivatives traders and religious fundamentalists are influenced by more than cognitive kinks such as these – external factors such as money, power and greed – although cognitive biases of course play a part.

From trying to explain the irrational mental consequences of our evolution to exploiting them for fun and profit. Mostly fun, but also a lot of profit (we’re talking billions of dollars here). Yes, it’s the perfume trade. Chandler Burr, the New York Times’ perfume critic (yes, such a post does exist) gave us an entertaining and illuminating talk about the world of scent, a subject I know bog-all about (so forgive me if I get his words wrong here).

Our sense of smell (detecting molecules from the ambient environment) is perhaps our oldest, and yet at the same time while we’ve seen every colour we’ll ever see in our lives, we will never be able to do the same for smell. Artificial molecules will bring us new smells and experiences, and the course of the perfume industry’s history was shifted irreversibly when they were introduced. Chandler gave an illuminating talk into how it was both complex science – experimenting with different aromas and combining – and yet also art, likening movements in the industry to movements such as Impressionism and Abstract Art in the world of painting.

Interspersed in this talk was a series of smell tests of individual scents – we all got given individual strips to sniff – that are combined together into a full perfume. Chandler was good at explaining how each scent can be regarded as a note, and how you hear a note depends on the context it’s put in and the accompanying notes and tones – one chemical which is a component of artificial strawberry can equally be used in a different combination to form steak. The final sample we smelled was the final product the components were used in, and it was amazing to be able to pick apart the different scents thanks to the talk we had. The perfume in question was Silver Mark by Jack Black (no not that one) which was awesome – spicy and woody and probably not to everyone’s taste.

A different spin on tech for fun and profit was the final talk by Caleb Chung, toy inventor. While this sounds like the most awesome job in the world it seemed to be initially a tale of continued rejection and perseverance, of being forced to keep budgets low and ideas neutered. Chung toiled for years before making a big break with the Furby – which in fact he was almost apologetic for; it seemed he didn’t wager how much of a global phenomenon the simple, cute but above all their interactivity – interacting with their environment in a way that might seem crude now but was apparently enchanting then.

Having sold 50 billion (or so) Furbys, Caleb became a made man, an experience he likened to a death in the family, as it left him without purpose for the first time in his life. On the plus side it gave him free reign to do what he liked, and so after much work he came up with Pleo – an animatronic robot dinosaur with an advanced AI that reacts and learns from its surroundings. He brought one out to play, and I was astonished. I knew it was just a device of gears and rubber, but it really did accurately simulate a baby animal which could walk, cry for ‘food’ (its favourite was banknotes, ironic given the huge R&D costs), climb and react to human contact with (pseudo-)emotional behaviour.

Perhaps what made this even more like something out of Bladerunner was the fact that it’s not just humans that react empathetically to it. One Pleo owner shot this video (fast forward to 1:40) while taking it to an aquarium – one of the attendants there said they had never seen behaviour like that from the dolphins before. It’s a rare occasion that I’ve been both gobsmacked and smiling, but this video was one of them:

Sadly, Pleo’s manufacturers went bankrupt earlier this year, perhaps hamstrung by the device’s high cost. It’s a device perhaps a little too advanced for its time, too expensive for mass adoption like the Furby, which is a pity as they were finding out new and interesting uses for it beyond the toy world – such as being the ideal pet for those with dementia or mental difficulties who need companionship yet also unable to take care of a real animal. Perhaps when the cost of the parts and manufacturing come down a bit, we’ll see a second generation of Pleos being mass-adopted, and with it an interesting set of questions on the ethics of artificial pets (do they have rights like animals) and what it means to be empathetic.

These examples of using science and technology to understand – and where possible – provoke human irrationality and emotion – were perhaps the most eye-opening and enjoyable aspect of Thinking Digital for me. Partly because it’s a field I’ve dabbled in but can’t claim to even be a skilled amateur at, so tricky is it to grasp. Even comparative failures like Pleo involve a level of understanding of human psychology way above my own, and as for being able to critique and create perfume – forget it. But it’s good to know one’s deficiencies, it’s the best spur to learning we have.

Coming up – more geekery, the future of media, and something that provoked my fear of growing old. In the meantime there’s also a post about a subject I do know a fair bit on the work blog as well.

The sheer amount of data that CERN produces is in the hundreds of petabytes, so a special distributed computing network is needed to help process it. But that’s not just the preserve of CERN – Simone Brunozzi of Amazon was here to talk about cloud computing, or distributed computing services for all. The parallel he drew up was of electricity – factories used to have their own in-house power generation, but eventually moved to a national grid; cloud computing does the same for processor power. Simone talked a lot of good stuff – of creating applications that are robust, scalable and on demand. And scalable is difficult – it’s not just a case of throwing more processors at it, it takes a lot of clever management and architecture around it. In an age where we’re going for mobile and lightweight devices, and universal broadband is becoming a reality, then I can see the justification of cloud computing, but it’s still not going to be a household name – its future seems more b2b and quietly in the background. However, I worry as and when the first major cloud security compromise happens (which it will, security is not easy either and I thought Simone was a bit dismissive), and the privacy implications of who’s able to look at your data when it’s uploaded to a cloud application.

Curtis Wong of Microsoft Research showed off their geeky toy, WorldWide Telescope. What struck me was how great it was having free content (everything produced by NASA is public domain) and with something that allowed people to create their own content; the video of a six-year-old kid talking you through his journey through the stars made me think – damn, I would have loved this as a kid. So much better than just a poster of the solar system on your bedroom wall. It’s not the only such software out there – there’s Celestia for example, but the community & user-generated content aspects make WWT a more fun prospect. That said, I have my quibbles with WWT – the web version’s in Silverlight and no Mac desktop version – come on Microsoft, you must know that’s such a clich�… :) There’s a TED video if you want to see more. Speaking of which, more TED-like stuff in the next post…

In a similar vein, there was also an interesting talk by Jim TerKeurst, director of the Institute of Digital Innovation, something I must confess has never really been on my horizon. Jim showcased some of the IDI’s fellows, who have worked on diverse range in the arts and technology – both art and artefacts. As a things geek, I was initially more impressed with the artefacts, such as chandeliers made from recycled plastic, the silver nanotechnology or the chairs that change according to your mood & clothing – annoyingly the IDI’s site doesn’t seem to showcase much, which is a pity. That said, later on one of the performers supported by the IDI, The Sancho Plan, gave us a great of their combination of live percussion controlling weird and wonderful computer animation, which I really liked – check out one of their videos:

The morning sessions weren’t just about targeting social problems or supporting the arts and creativity, but also about cold hard business. Well, with a cuddly side. Alex Hunter of Virgin group (though a committed Diet Coke drinker) talked about how he’s reshaping the Virgin website for a Web 2.0 & social media outlook. It was for the most part a well-presented Cluetrain Manifesto but still had some interesting lessons; Alex regards Digg’s blog as the best corporate blog – not just because it’s written by the guys at the top, but because it’s a multiplicity of voices and they respond to their fans. Geeks with fans, who would think it? But then, Digg know the audience they’re blogging for, and it’s harder for non-tech brands, so be careful of using them as an example.

Still, Alex was evangelistic about embracing social media in the business word, and made it clear it works for brands big and small (citing Qype and Zappos as examples). We also got some insights in the Virgin process – they have Virgin Eye a beautiful visualisation of mentions of their brands on the web (from over 5,000 sources) and other “labs”-style projects from Virgin at Explore Virgin. They have a new website, more of a community platform out which they’ve spent a year and a half listening, researching and creating, which is an impressive level of care and attention (although in a world where online fads come & go in days, risks being stale on the day it launches).

From another business point of view, Harry Drnec talked about his experiences as MD of Red Bull. His philosophy was from the emotional end of the spectrum rather than the practical – find your consumer, touch them, thrill them. Marketing wank? Possibly. But there’s no denying how attached people are to Red Bull as a brand, despite the ridiculous price it sells at (Red Bull made it a policy of not cutting price to increase sales, preferring the premium cachet). Now he’s trying to do the same for computers – make them rely on as little skill on the user’s part as possible. A noble goal, but I hope they don’t confuse simplicity and intuitiveness; by making things too simple to use we risk destroying their power and potential. Intuitiveness is what counts.

Right, enough business. Next post – hardcore geekery and genuine leftfield afternoon weirdness.

When someone replies to you on Twitter in public, they use the @ sign like IRC. When looking at others’ replies in your stream, you had the choice of either seeing every single @ reply that people you followed made, or just those to other people you followed. 2% chose the former option, 98% chose the latter. Yesterday Twitter changed it to the latter only.

Cue backlash from the 2%. Cue backlash against the backlash (from me, amongst others). So Twitter freaked out, backed off and fessed up – discriminating between those you follow and those you don’t was proving to be not scalable. So the solution was to reverse the decision, and make @s to everyone visible – unless they were done through the “reply” button. So you now have to rely on how others use the system rather than have control over it – the worst possible solution.

Intermission: Nobody gives a shit about this. Why am I writing about this? WHY?

A half-arsed solution if ever there was one, and one that caused me annoyance. I flippantly Tweeted my annoyance – which I still stand by – and this got a mention in the Guardian’s Tech blog (thanks guys). But this is one of those things that needs more than 140 characters to elaborate on,so here goes…

Twitter’s approach was a classic fail in consulting users. The @ was a community-created asset and Twitter messed with it for no apparent good reason. Cure was worse than problem, and then they were forced into an icky compromise that suits no-one. The solution? They could fessed up it was causing problems, and announced a change well in advance. To help users prepare for it, they could extend the API to allow clients like Twhirl & Tweetdeck know the user IDs of who you follow and who you don’t. Then the clients can make the discrimination between followed and not followed, instead of the server, and the choice can be exercised at that end. Scalable, user-chosen, none of the problems encountered above.

Right, that’s it. Of all the things I’m meant to be writing about, I didn’t expect to write at length about this. Better stuff to come, promise.

If you’re thinking about commenting, don’t – I’ve wasted enough of the planet’s time as it is, please don’t add to it.

The talk was slightly disappointing in that it was a bit too general and strayed into unrelated aspects security debate as well. For example, talking about Richard Reid the shoebomber and how if he had been successful, we would not be taking our shoes off at airports now as nobody would have known how he did it, is a lovely anecdote but didn’t do much to inform the debate. Ditto complaining about school web filtering or the state of the tabloid press.

Still, there were plenty of good points and interesting notes. Charlie kicked things off when he pointed out how privacy is a relatively recent social construct; in medieval times peasants would live communally and the nobility would be constantly under the eyes of their servants; only through the industrial age and the institution of a middle class, segregated in their own houses, has privacy become a norm to expect.

Spin forward a couple of centuries later, and we are no longer merely industrial, but sufficiently technically advanced that we produce data about everything, which spills out of our houses and lives. As a result the normative privacy we expect is an ever-more elusive ideal, now teetering on the edge of a paradox. One the one hand, people publish vast amounts of information willingly, whether it be on blogs, Twitter, Flickr or whatever. On the other, we clamour for the government to stop snooping on us, gathering massive databases or performing mass-mining operations.

Are the two mutually incompatible – are we just hypocrites? Not quite. As our information-generative capacities have evolved, privacy has evolved into a part of a more sophisticated framework; privacy is no longer a simple norm, but a beneficial side-effect of a much greater good – the ability to control what information about is is dispersed. With this control, we choose what we can and cannot publish (for example, you’ll read my views on tech and politics here, but only very rarely will I disclose information about my family).

But Cory very astutely pointed out that this choice is rarely informed. We are very bad at valuing the impact of information at the time we release it, compared to how we might value it in future; a teenager posting pictures of him drinking and smoking weed onto Bebo might not matter to him now, but when he’s applying for a job or running for office in a few decades’ time, they may come back to haunt him. You can add into that the mix the fact that taboos and social conventions change over time. Charlie provided the example of parents would take photos of their young children playing in the nude – regarded as innocent in the 1970s, but an activity now tainted with anti-paedophile hysteria of the present day.

So, we’re producing shitloads of personal data, the control over which we are unable to judge accurately. That’s part one of the problem. Part two is that that data may not be accurate or can be misleading in the first place; if your browser history is full of pages about HIV, is it because you are a sufferer or just researching on behalf of a friend, or helping your child with their biology homework? Without wider context, it’s easy to draw the wrong conclusion.

It gets really bad in part three – what happens when governments and corporations start using these vast pools of data, possibly mistakenly disclosed, possibly inaccurate, to make judgements about us; would a health insurer, getting hold of your browser history in the example, force you to undergo an HIV test or increase your premiums?

In this sense, privacy, in the sense of safeguarding information about us that is true, but which is not widely in the public domain and we’d rather not let people know about, is not the real problem. The problem is about what institutions do with all information about us, true or not, publicly available or not, embarrassing or not. And that’s a bigger question than privacy – we get back to control again.

As Charlie put at the end: “the relationship between privacy, security and the state is broken.” The social contract we once relied on for privacy is being pulled and reshaped into a wider one about information control. But while personal control of all our own data is an ideal goal (that us geeks love to profess), it’s really one fraught with complications – if we ourselves can’t make the right judgements about deciding what to do with our personal information, what hope do institutions have?

Doubt is the key to reworking out this social contract – a healthy dose of uncertainty and a warning about context whenever we deal with personal information. In some ways this means accommodating conflicting ideas – a “Digital Britain” made more reliable and efficient through IT excellence is underpinned by an assumption of good-quality data, but we must also entertain the possibility it is wrong or inappropriate and build mechanisms around it so we don’t make mistaken judgements, and that mistakes can be easily corrected or reversed. Oddly, the closest equivalent to this I can think right now is how any sane person should read Wikipedia – treat what you read as plausible, but never be willing to accept it as truth on blind faith alone; the community behind it does its best to keep it accurate, while full well knowing mistakes will always be there.

Never perfect, but honest with yourself about your fallibility. That might just bring some sanity to the situation right now. But good luck recommending that to anyone in a position of power.

Footnote: Better and more coherent posts on the Doctorow/Stross talk can be found from Richard King and Chris Swan.

For starters, we already have a mechanism for kicking Prime Ministers out of power. They’re called elections. True, they only come along every 4-5 years, but then democracies can only function with some continuity in administration. If you don’t want to wait till next year to cast your vote, then fucking tough, and count your blessings you live in a country where the citizenry can vote leaders out of power.

Second, you’re missing the wood for the trees. Gordon Brown’s premiership has been nothing short of a disaster, but it’s been very much a team effort. It isn’t just the man at the top, it’s the whole bloody team around him. Darling, Mandelson, Smith, Blears, Geoff fucking Hoon. Not to mention the old guard of Blair, Clarke, Blunkett, etc. setting it all up. The whole lot of them are responsible for the miserable idea-starved philosophy of New Labour, with its craving for power, its thrall for made-up wealth and its authoritarian managerialism in every department.

Third, what comes next? If Brown resigns, then Harman or Miliband or possibly Straw or some other New Labour apparatchik takes his place. A PM resigning does not mean a change in the ruling party, or even an election, and you don’t have to go that far back in history (Wilson, Thatcher, Blair) for examples. If you’re not happy with the current government then you really should be making a case for what the next one should be bringing (and fuck knows, the current Opposition need every bit of help in that department).

Finally, you’re a fuckwit because you’re a petition-signing fuckwit. Petitions have some very good uses – raising awareness of a previously unknown issue, or for co-ordinating action for the disadvantaged and disparate, or to help lobby for changing unjust laws. But these don’t account for many of petitions on the Number 10 site. Many of them are just ways for the inarticulate and incoherent to vent frustration at something they dislike, and petitioning the Prime Minister to resign is the stellar example. It has no hope of changing anything. It advocates nothing positive whatsoever. All it is is another way of childishly wailing “I don’t like you” to someone who doesn’t give a fuck, just so you can congratulate yourself with the delusion that you are sticking it to The Man. Well done, you must be so proud. Fuckwit.

Footnote: That was a bit of a rant, and I’m not normally this grumpy so here’s something to cheer you up. Have a looksie at the rejected petitions, some of which are terrible and some pretty good, such as this one which Tom heartily recommended.