Book review: The Undercover Economist -- by Tim Harford

Book review: The Undercover Economist -- by Tim Harford

This is my speed review for the busy blog reader!

This is a good book to give a simple introduction to economics in everyday action. You don't need to read the whole thing to get the gist as it seems to get a big repetitive (this is my comment on just about every non-fiction book these days).

Also to be honest, I couldn't be entirely comfortable about the argument towards the end of the book that third world sweatshops are good because economics tells us so. This is all based on the theory of comparative advantage. This is all very well, but what happens if one country has no comparative advantage, that there is nothing they can do better than anyone else.

Book review: Maelstrom - by Anne McCaffrey & Elizabeth Scarborough

Maelstrom - by Anne McCaffrey & Elizabeth Scarborough

Don't bother. Just don't bother. It is the Nth book in the Twins of Petaybee series and I have no regrets about not having read any of the earlier ones. Did Anne McCaffrey really have any hand in this dross?

I wondered throughout the book if it was in fact intended as a children's book. It certainly lacked any kind of sophistication in its characterisation, storyline, etc. However, Brisbane City Council had it listed as adult fiction, and Amazon reviewers gave it 5 stars (out of 5). Generally I have found Amazon reviews pretty useful assessments, but in this case they praised the book and I thought it was just awful.

Of course, I have long held that any work of fiction written by Foo & Bar is bound to be a disaster. Now this opinion is based on statistical observation that I rarely like novels written by two (or more) people. Logically I cannot rationalise why novels written by multiple authors should be bad. After all people collaborate to make TV and movies and music, and that seems to work, but put two people in front of a typewriter and it seems to spell disaster.

more on research groups

Following email exchange, it would seem that Ricky and I are in violent agreement on research teams. We are resolved (I think) that:

* lone academics counting angels on the heads of pins are Bad
* researchers voluntarily working collaboratively to solve significant problems is Good
* researchers forced to work collaboratively for reasons of politics etc is Bad

Wikipedia - yeah, it works

The death of Anna Nicole Smith and its reporting on Wikipedia lead to many acts of cyber-vandalism on the page. Some people seem to see this as proof that a everyone-free-to-edit environment like Wikipedia is doomed to failure.

On the contrary, it is incidents like this that prove that Wikipedia works. It is resiliant to shocks. Yeah, a few people got about 15 seconds of fame while their tasteless remarks sat on the wikipedia page, but by the time they had emailed or instant-messaged their mates to say "look at what I did", those remarks would have been removed and responsible Wikipedia users would have been taking steps to get the page locked down. If you want to see the vandalism, it's perpetuated by the sites commenting on the vandalism rather than on Wikipedia itself (ironic, really).

Like all human activity, we come up with rules and processes to control undesirable outcomes, like driving on the same side of the road (well, on a country-by-country basis at least), putting the bad dudes in prison, and locking down Wikipedia pages that are subject to vandalism. If new undesirable outcomes start happening, we change our rules and processes. Wikipedia over the course of its existance has evolved to cope with these incidents.

Vandalism on Wikipedia certainly happens and indeed is nearly impossible to prevent, but it recovers quickly from it. While Wikipedia's recent change patrol (those self-appointed bastions of "what's right on Wikipedia") can be a bunch of over-officious zealots a lot of the time, still they do perform the important social function of ensuring that vandalism is quickly removed (having an underlying verson control system makes it easy to roll back changes -- hurray for IT).

Wikipedia is sustainable precisely because enough people get enough benefit out of it to think they should put something back into it to keep it thriving. People contribute in different ways, whether it be by providing new content or being an editor of existing content or being in the recent change patrol or even donating cash. It's when people stop making their various contributions that the vandals come in, destroy it all and it loses all credibility. Much like urban neighbourhoods and graffitti and why "zero tolerance" policies actually work.

in praise of research collaboration

Ricky argues against research collaboration. Well, he doesn't exactly say that but read it for yourselves.

My counter-argument to his claim that it is mostly individuals who make the breakthroughs is that we tend to remember and idolise those individuals (the cult of the celebrity, as today's headlines on Anna Nicole Smith demonstrate), whereas it is not practical to enumerate the vast armies of people who collaborated on other advances. Who invented the CD-ROM or space shuttle or GPS or ...? Was there an individual? Was there a collaboration? What about the computer? Turing? Babbage? That's a heck of a long way from the laptop on your desk today.

Ricky mentions DNA. Sure, people think of Crick and Watson as the individuals who cracked DNA. Yet there were many other people in that story, including the controversy over their use of Rosalind Franklin's data. Even the wikipedia story mentions many other people who contributed. Now Ricky will argue that the informal collaborations of academics produced the desired results. Yes, but would not a formal collaboration of the correct range of skills in the one lab with a common goal have found it sooner?

The danger of the sole academic left to their own devices is the "angels on the head of a pin" phenomenom where they refine their expertise in a matter of no practical use to anyone. In the case of IT research in which we tend to build things, there is infinite scope to build infinitely many things of no interest to anyone but their builder. I am no fan of the sole academic. I'd rather see teams harnessed towards a specific useful purpose, but with a bit of slack for skunk works for those occasional flashes of individual brilliance.

And finally playing the man not the ball, where has Ricky received his research funding from lately? Being a sole academic or being part of a research team? As Hilaire Belloc says in his poem about Jim who ran away from a nurse and was eaten by a lion, "always keep a hold of nurse, for fear of finding something worse!"