Here’s a suggested difference between geeks and policy wonks that might go some distance towards making the two groups get on with one another better. It is my contention that the two groups simply have radically different registers and types of interaction. Policy wonks – like all politically oriented people – are encouraged to think in terms of combative point-making. The most respected and well-thought through acts of Parliament being those that have been fought over the most. The most convincing politicians are the ones who have solid positions that they can stick with and defend. Political life is a combative life, with positions being tested and retested before they’re taken out into the world. In terms of doing things you want to know that the thing you’re going to do is the right thing before you get too far down the line, particularly when the consequences of getting things wrong are so potentially enormous.
The life of the creative geek community is very different. The atmosphere of an event like ETCon is not one of absolutist positions (or at least it is on occasion but it’s mostly frowned upon), but of gradual accretion, iteration and development. Particularly (but not exclusively) in those realms where development requires time but not a lot of capital investment, ideas are thrown out into the world to see if they’ll stand or fall. Those that succeed are iterated upon. Those that fail are either abandoned or taken further by other groups who will try to solve the errors and mistakes that surround them. In terms of making things, each new idea is expected to be flawed and clumsy and full of holes and everyone knows it and works from that point onwards. It’s the model of the technologist community as competitive craftspeople, and it operates on the assumption that whether something will be successful or unsuccessful / useful or useless is something that must be left up to how people interact with it and its take-up with a community. You make it the best you can, in the way you think is right, and let the world decide if you got it right…
I think this is the distinction that explains why there are so many disagreements between the groups. One group looks for immediate application where there may be only potential. One group sees possibility where there is no immediate practical benefit. And in talking to each group, you have to use a different register. There’s no point talking RDF to policy wonks, because they’ll see no application until you can show them something made with RDF that they consider actually politically useful. And there’s no point telling technologists that their creations are politically naive, because they’ll consider them works in progress, building from a position of naivety towards – in time – something legitimately useful and ground-breaking.
It’s a difficult job – understanding which register to use in which circumstance – but it’s an important one for those people who have to straddle disciplines. Because one way or another they’re going to have to work with geeks or wonks who will by necessity have a very different mind-set. Being aware of the distinction will not only create the possibility of legitimate discussion (and minimise the possibility of large cross-disciplinary enmities) but also inspire actual creativity to emerge between the disciplines…
14 replies on “On a difference between wonks and geeks…”
Can you provide a little context for this? I’m not at all convinced that there is a yawning divide between policy wonks and techies. The examples I can think of – Lessig, the W3C – certainly don’t have a problem with the uncertainty of technology. And the model you identify geeks as being comfortable with is how I see the legislative/policy development process. Entrenched positions don’t get one very far there – compromise and experience are key (well, that’s idealized, but anyhow)…
Maybe I’m missing something here?
Is policy-making not also iterative? Consider the initial steps, they being the identification of a problem by an individual or small group of individuals.
Then several possible solutions are kicked about amongst individuals, with increasing momentum telling until a majority-agreed solution evolves into a motion, then a green/white paper or bill.
Published to competing ideologies and opinions, a bill will be subject to several revisions before eventually being passed to the upper chamber, which may also propose revisions or seek to block the proposal entirely before it has a chance of becoming legislation.
In the months and years after the solution passes into legislation, policymakers will keep a close eye on its progress and look to tweak it as is fit.
Broad ideologies may be firmly entrenched, but surely democracy can be argued to be an evolving, iterative doctrine?
By contrast, it could be argued that the geek sphere is particularly attuned to point-scoring. Witness the hackers and crackers keen to be first, crack best, bring down the biggest website or develop the neatest tool before the rest. Despite ideological differences, isn’t politics more consensual than that?
Well it’s quite possible, of course, that I’m talking about a very particular kind of geek and a very particular kind of wonk. I think there may be an element of truth to that, but I also think it goes slightly deeper than that. Think about the acts of legislation that you’re talking about and all your iterations moves are – as it were – before launch. Although the law will be revised after the fact – and any changes will necessarily be entered into with that understood beforehand – legislation must, on the whole, have been bashed around and unpicked and sown together by those experts behind the scenes almost indefinitely before it will become law. The ‘release’ of a law that people are agreed beforehand to be bad or clumsy would seem counter-productive. Among a large group of organisations and individuals creating software and websites online, it is absolutely normal.
I’d love there to be a some kind of concept of a political beta law that’s in effect and usable but might break down under a variety of circumstances, and which requires feedback from the general public, but unfortunately – in its purest form – that’s not particularly practical. At the moment there is no conceptual test-bed into which multiple competing laws can be thrown to see which ones will inspire further work on those laws or become recognised as useful in their own right.
Move this iterate-first model into the think-tank kind of environment and you’re working exclusively from research, precedent, extrapolation and model-making. Policy-making is the paper prototyping in which the paper-prototype becomes the thing itself.
Now of course, if you work in a commercial organisation or have your own company, exactly the same restrictions apply (albeit to a lesser extent). You can create a project, but you really should attempt to iterate in-house if you want to keep a certain competitive advantage, you should attempt to do research and work through all of the issues involved before launch, you have to make sure that your product is at a usable form. But this is a function of business, not of geekhood. The geek model, alternatively, has culturally been one of attempts to get things right to an extent in private before launch, but also to allow or even encourage immediate feedback and subsequent rapid iteration in the wild, with the result that there’s proliferation of alternative versions of intriguing models followed by live competitiveness, and the gradual success (fast or slow) of one or two models over the many that existed beforehand. This could be a function of our place in the history of computing and communications, of course.
I could also talk about the difference between arguing a point and demonstrating a point as a cultural difference, but maybe another time.
Very interesting… its worth thinking about DEMOS and their supposed ‘open source’ model of policy development, which does (in theory) release beta model policies, if not beta model laws.
It’s also worth considering the role of the US in the UK policy scene. With its 51 legislative bodies (including the Capitol), the US is where UK policy-makers look to see what works and what doesn’t. Our welfare policy, for instance, is born out of watching various States setting up policies and seeing which ones “stick”, to use geek parlance (for the record, Wisconsin stuck).
Now what about rebuilding the intelligence services around collaborative geek principles…
Maybe you hang around with nice geeks too much?
I know many who are highly combative and some who consider social niceties like concensus forming and listening to others as something to be optimised out.
It could be argued that the geek model’s propensity to release products which work “to an extent [but] encourage immediate feedback” would be irresponsible in the public sphere.
It works well enough in the open source community, but Microsoft, for example, is now on the ropes for releasing buggy software and fixing it later.
All kinds of policies are subject to pilots (perhaps analogous to your “political beta”?), and these, too, serve the basis for concrete policies and best practice.
But such heavy reliance on developing policy in the wild as you go along, as with software rewrites, might not wash in, say, the health service.
Also, I would’ve thought immediate (and prior) feedback would have been built in to the democratic system as a pre-requisite. The problem is, most people seem to have forgotten that.
Interesting.
To James: I don’t necessarily mean that they have to do consensus forming, but that they understand that you build something and stick it into the wild that SOMEONE will take the idea if it’s any good and take it further. Whether you like it or not. I mean Dave Winer is not a particularly play-nice-with-others kind of man, but he understood that there was no way that he could put RSS out into the open in such a way that it would either not be (1) ignored (2) not adopted but fiddled around with or (3) adopted AND fiddled around with. The equivalent would be creating a law that was just roundly broken and wouldn’t work at all and might even be damaging if it was widely adopted (but all laws are by their nature widely adopted) and then waiting for subsequent versions to fix those problems until such a point that everyone decided they’d go along with the law and agree to it.
To Rob: In terms of software and websites, the Government actually don’t produce that much generally – and when they do it’s in this interplay between areas that they have to operate. But certainly, with regard to LAWS and the like, it would be EMINENTLY irresponsible for a government to put a beta law into play. That’s why the attitude has developed in such a fashion – because there is no simple Darwinian ‘market’ or ‘community’ for laws to achieve dominance within or demonstrate their utility inside.
Tom, I think you make a very good point here, as it happens. Of course, you are dealing with archetypes, but if the archetypcal wonk could learn from the iterative shared-development-of-ideas model that many geeks are comfortable with, then we would end up with better policy. The adversarial nature of much that passes for politics and policy making is one of its principal weaknesses.
Mmmm. I think you’re a bit off-base here. Geeks and policy wonks both work iteratively, but within various sized circles and in different ways. OK, Tory and Labour policy advisers are unlikely to work in conjuction with one another, but then developers from different technical organisations won’t likely work too closely with each other on particular products. The entire geek world is not open source. In fact, since laws are generally based on simple principles and discussed openly in media and politics, they could be considered more collaborative. We all have a say, whether we know anything about politics or not.
In computer terms, laws *are* put into beta; they are trialled, they are tested, they are given small spheres of influence to experiment in. The problem is that these trials – and the laws which are subsequently enacted – have far-reaching consequences. They can’t just be released and then worked on some more, without a lot of work. Law 1.0 and Law 1.1 might well occur, but there’s a lot of damage that could happen in between.
You’re not buying a single product from a political party. You’re buying an entire ethos. It’s not about cherry-picking what works best for you, and building a package from that: it’s about choosing which is better suited to your ideals. That, maybe, is what is wrong with oppositional politics; there’s not enough compromise on success rather than ideology. But then again, when something works, it sticks. That’s why we’ve had the NHS for 50 years, through governments of the right and the left.
But I think it’s wrong to imply that there is too much that politicians could learn from the working methods of technical development. It’s just too cosy.
You think you’ve got problems? Just try and explain the concept of a wiki to a group of lawyers!
Democracy 2.0?
Interesting discussion over at plasticbag.org, where Tom suggests politicymakers should adopt the geek community’s iterative, test-and-improve model for lawmaking: I’d love there to be some kind of concept political beta law. There is no conceptual tes…
Yesterday I had the opportunity to observe a fascinating lessons-learned from a systems integrator-type geek vis-a-vis some major player U.S. policy wonks. A summary of his rather agitated discourse — U.S. policy wonks are constitutionally incapable of deciding what they want from a project and speccing it. He went on to note that cultural differences between the Arab policy wonks, who not only decided and specced but also bought a beta and submitted it to rigorous testing and scaling, and Western indecisiveness might account for the difference. Us back-office folks now have a challenging migration project (the beta is working, but the subsequent product is obsolete). But the early adopter was much more interested in function and results, while the U.S. policy wonks remain stalled on the decision process itself, and the technology developed onwards. Since the current product has been discussed as a major component of a national ID system, including the U.K.’s, does our systems integrator have a valid point?
Democracy 2.0?
Interesting discussion over at plasticbag.org, where Tom suggests policymakers should adopt the geek community’s iterative, test-and-improve model for lawmaking: I’d love there to be some kind of concept political beta law. There is no conceptual test-…
This is old hat. Green and White Papers are ‘ALPHA’ and ‘BETA’ versions of the Law respectively. And most Legstlation is shaped in way that allows for some parts to be retrospectively changed following implementation. This has been happening since the Parliament Act (1911).
The DEMOS proposal, as I understood it, was that it should be OK for one Policy Wonk to take the work of another and adapt it under some form of GNU-type licence without being ashamed to accept the the underlying ideas were generated elsewhere. The trajectory is that [i]the polis[/i] should move to paradigm in which policy is non-combative but is communally developed.
Tom’s approach seems to be governed by his alienation from the political process: i.e. he only hears about laws when they become a media issue and it has a potential to effect him.