Your skills are great, but...

I have had some conversations recently about the psychology and sociology of Computer Science and peripheral fields. Much of this conversation has centered around the "End-to-End Research Vision" for "making the world (of communications) a better place." A friend of mine, who does research in networking protocols, elaborated on the identities of the report's authors. Apparently these guys are well known for inventing various protocols, algorithms for network congestion control, and other networking bits and pieces. Apparently they perceive these contributions to be of Nobel proportions. My friend admits that all of these protocols and algorithms are essentially arbitrary, derived from little or no underlying theory, and evaluated mostly through simulations and emperical data.

This might shed some light on the psychology of Computer Science practitioners. In many or most instances, they produce arbitrary instances of algorithms which they might derive out of the clear blue sky. When things work out well, the designers/programmers are said to have "skills." But these are not ordinary scientific skills. They are magic guru skills. In the mind of the Great Inventor, there can be only one inventor of this or that world-shaking algorithm. The one who devised it did so through an innate talent, and it could not have been produced by someone else.

In the case of the End-to-End group, it seems that "world-shaking" is equivalent to "deployed ubiquitously in manufactured systems." This makes their claim to greatness somewhat specious, since commercial deployment depends mostly on "right-place, right-time," not on some inherent perfection of the design. In the case of networking, ubiquity also arises from the propagation of a few narrow standards into millions of devices. Their claim to fame is at least in part due to the luck of the draw. Theirs were not monumental technical innovations; they were business successes.

I pulled this random sample off of the web-o-Minkowski-space today, in which the author explicitly suggests that "programming skills" are innate and perhaps genetic. The real goal of the article is to explain a supposed finding that "90% of people think they are of above average intelligence." I think the author's reasoning involves some really wild underlying assumptions ranging from genetics to epistemology:

People will polarize towards the jobs they are good at. If you are born with amazing computer skills, you are likely to get a computer-related job. If you have an amazing ability to play football, you may well end up playing a lot of football.

Now, most people aren't going to want to spend the rest of thier lives worrying about thier inadequacy in certain areas. Thus they will tell themselves that thier skills are important. After all, they (the skills) got them (the person) where they are today. They may disregard other skills, rather like PC gamers who don't own consoles will rarely say that yes, a console is overall a superior platform.

Anyway, here's my point: People often judge other people's intelligence in the same category at thier own, and vice-versa.

For example, I think my skills (programming) are better than some other guy's (memorising the winners of the last 25 years' Boulton Wanderers games), because I've been just fine up until now without knowing what he knows. But I've used *my* skills a lot.

I am confidently above average in the programming league, because most people can't program. I'm right at the bottom of the Accounting Skills league. I class myself as above-average, because I'm an above-average programmer.

In conclusion: Different people are value different things. By my own standards, I am above average. By yours, I could be really dumb.

The triumph of reason!