Hi, Bill, yes, many developers have success stories similar to that.

Back in the 1970's, IBM started recruiting young software talent in a manner that "went against the grain" at the time. Instead of looking for top students in math and science, they started looking for the top students in the liberal arts.

This concept also resonates with the great software author Allen Holub, who felt a very strong connection between music/history (especially history) and software development. What is software development? Reading and writing and identifying patterns. What is history? Reason and writing and often identifying patterns.

(Now, I'm referring the application development and not scientific development. For the latter, you obviously need some training in math/sciences. But truthfully, having worked in this industry since 1987, I rarely need to use anything more than set/subset/union theory and basic math. One time I had to calculate slope manually.

Of course, there are plenty of exceptions. I've known good programmers who had difficulty typing a well-formulated paragraph in an email and had no academic inclinations at all. So being able to write a summary of WW II doesn't mean someone will be a good developer. :)


I have never found software development to be a pure science or a pure art. Software development is a craft. Maybe it leans a little more towards a science. There are red rules you always follow and blue rules you use as guidelines (stored procedures and star-schema data marts for analytic apps, haha). There are people who can build things on questionable architecture, but mask it well enough that no one notices. Of course, that's more likely to occur on single man projects, and this industry has had plenty of those. There is a danger of being self-taught, if you don't have a mentor...the danger is not developing bad practices, but rather learning to be clever enough to mask bad practices.

I don't like the idea of establishing liability and things along those lines. I'm not saying it doesn't happen, it certainly does. But I'm not sure we're prepared to change the formula for our "craft" to invite that kind of evaluation, at least not formally. This is a complicated topic, and Mike Y. I have to say, for all the times I agree with you on things, on this one I tend to see this differently. Not saying I disagree - I just have a different view of it. I've known good or at least competent developers who built crap - partly because they were in bad situations (or allowed themselves to fall into bad situations). I just tend to think this is a very complicated topic - and I'd rather just let the market weed out those who truly shouldn't be building things for people.