Tuesday, February 23, 2010

The Progress Driven Life

The reason to do what you do could come from any one of many different sources. Some people do what they do because it's what their parents did. Some do it because they think it's right to do it. Some people do what they do because it's what they think they're good at. Some people do what they do because they think they're having the best of times doing it.

I'm in the last two camps, I think I'm good at what I do (and think that it is fun), and get upset when I am asked to do things I'm bad at. This can be frustrating sometimes.

In reference to an argument between myself and a colleague at Broadsword, I'd like to propose that everyone should strive to find out what they're best at and do it, rather than work towards being a lawyer, doctor, dentist, stock broker.

Some time ago I posted (might not even be here) that I felt a little bad sometimes that being a games programmer was a bit like being a drug dealer, in that we make things that actively try to addict people into playing them. Especially with the advent of the MMO, we're seeing people losing their lives to the drug that is games. I felt that what I was doing was somehow wrong.

In the argument, which was started over the question of "what science should we do more of to make the best progress for the human race?", I came to the conclusion that pharmaceutical cosmetic drug research, though not obviously useful in any way other than to beauticians, was still an active science, and therefore should still be on the list of things that humans should be proud of in the name of progress. Accidents happen in random labs and there's no saying that cosmetic chemists won't be the ones to find the next big thing.

Which leads me back to games. One the objections I have to large scale games publishers is their fear of the odd, of the new. This lack of innovation, though there is plenty of work done very efficiently, may stifle new emergent technologies or strategies for progress in games development. I fear that unless these companies engage in a 20% time of active innovation, they're going to remain large market share, but with little growth.

Games development is odd, it's smaller and larger than other software development at the same time. We have a much lower expectation of robustness of code than most of the other industries, as no-one's going to lose a life over badly written code, but our software has to take into account so many more variables than anything else. Where else do you find that the baby science of artificial intelligence is so commonly used and performance pushed to the limit of hardware?

From this furnace of rapid development we find many improvements to technology that otherwise just wouldn't exist. If it wasn't for games, there would be very little of the technology that runs the beautiful Pixar films, there wouldn't be the large TVs that now clutter our living rooms, there wouldn't be Blu-Ray or film download services, and there wouldn't be easy to drive military-tanks or our beautiful camera phones.

Every form of technology feeds into every other form of technology, and the sooner we realise this the sooner we can all feel proud that Mario helped you survive cancer.
Do what you do best, never assume that what you do will not help the human race. Just do what you must, because you can. For the good of us all. Except those who are dead.

Monday, February 22, 2010


I found a very good definition of the amount of control we have over ourselves in the book Happiness Hypothesis. It describes us as being two minds, the rider (who has a whip and a good sense of tactics and strategy), and the elephant (not slow, but hard to change direction, and generally knows what it's doing and follows the path of least resistance without input from the rider).

This idea of rider and mount doesn't just work for people, it also works for any other large scale operation involving a fast moving mind with a slower moving ultimate decision process. Business is definitely one area that it links very well (based on readings of Built to Last and Good to Great), government also, I think.

The importance of value in this view of life is not to be unappreciated. Value is the way of setting the goals of the elephant. If you want to make yourself do something, you can either use up a lot of energy (will power in constant use) thrashing your elephant to move in the direction you want, or you can change the elephants values (one shot use of will power to change values).
In business, you can adjust the direction by adding incentives and fines for certain types of behaviour, and the same for government.

So, if you want to change what you do, you have to change your values. But firstly, you have to find out what your values are. How do you do that? For starters, your rational brain won't really know what your elephant's values are, so you have to do a bit of observing. Watch what you spend time on, because that's a good indicator of values. If you find it easy to commit to something, then it's probably true that your elephant holds that in high regard. Once you have a handle on what your current set of values are, then you can start to think about what you might want to change to make your life a nicer one.

Values can be changed, but you really have to work at understanding the ones you already have, otherwise you might end up trying to add oppositional but orthogonal forces. A reasonable example could be smokers. Some won't give up because they really value their cigarette time, and even if they are rationally aware of all the dangers of smoking and how much better off they will be financially, they can't really give up because they enjoy the chatting with other smokers element of their life. A good way to move into a better quitting position might be to always stand away from the pack of other smokers, in order to reduce the positive effect of smoking. Take out the positive side and the elephant disconnects and can be taught to appreciate the healthier living afterwards.

Another might be television, if you feel you're addicted, then begin by recording TV, and turning off the television until a show you want to watch has finished recording, then skip all the adverts and announcements. This might help reduce the effect of "just one more show" that is prevalent amongst the couch potato generation. Also, plan to do something, anything, some DIY or read a book, just so you have something to do while you're not watching telly. Hopefully, the values of time spent can be changed through distracting your elephant.

Basically, you're trying to change the rules of physics. Any time you change the laws of physics, you change the optimal solution. Your elephant naturally tries to find the path of least resistance. The elephants in businesses are even easier to divert, just use money. Fine businesses for things you don't want them doing. Reward them for things you do.

Okay, so that sounds easy... why hasn't anyone done that before? Well, they have, but the corporations of the U.S. found a loophole. It's the same loophole that anyone that's done genetics to any degree has found. The Selfish Gene points out that many highly complex interactions can come about through the relatively simple process of reproduction with inheritance. Sometimes cheating is actually the best fit in a complex landscape. You have to be careful that you don't create artificial fitness landscape that fits cheaters better than people how are doing the right thing. Sometimes it's possible to find a local maxima that matches the true intention of the law givers, but more often than not, a badly thought out law can offer a terrifying maximal solution, that by all accounts, is judged fair on paper.
Fixing this usually means using less variables, lower the complexity of the fitness landscape. Omit any naturally occurring fitness rewards or fines (such as, transport costs or pr). Keep the laws simple, and massively promote open books so the media can contribute to the morality of the corporations involved.

Government has the hardest job, it's an attempt to change the actions of its people, by changing the values of its people, by changing laws that affect both its people and itself. To do this it has to find value in change, and change the opinion of its members to allow the introduction or change in the laws, which might mean it needs to create a situation with which to convince its audience. Hard work working a slow system from the second or third removed driving seat.

Happiness Hypothesis
by Jonathan Haidt
Built to Last and Good to Great
by James C Collins and Jerry I Porras
The Selfish Gene
by Richard Dawkins