A Funny Java Flavoured Look at the World

Sunday, July 30, 2006

Can you spot a good programmer from 50 paces?

My parents often struggle to grasp the idea of me being a programmer, they are always asking

"what is it you actually do"

My Mum sort of understands but my Dad cannot understand that the work I do cannot ever be held/seen physically. My parents do often look for the work Java, knowing that I use this witchcraft in my daily job.

So it was with interest that my Mum sent me the article by cutting it out of the paper and then posting it to me, which in my book means that she has failed the programmer test before she has even taken it.

How can I tell if I'll be any good as a programmer?


The article links to a study that claims it can tell who is going to be a good programmer or not by using just three lines of code or as the study says
"programming sheep from non-programming goats"
Using his 3 lines code his results show that 44% of people who can programs, 39% who can't and 8% who didn't answer.

The actual study is 21 pages long and was done by Saeed Dehnadi and Richard Bornat at Middlesex University's school of computing and can be found here

http://www.cs.mdx.ac.uk/research/PhDArea/saeed/paper1.pdf

and although interesting it's not that interesting or to word it better I'm not that interested. So what are the 3 lines of code that can sort it all out.

a = 10;
b = 20;

a = b;

I can't really understand how this can split the people so efficently. The are about 10 different answers so it's not just chance or selection (ish).

Whether it means you are a good programmer on the other hand is a different question because the people he was testing where doing a computer course and he got them to do this test at the start of the course and then saw what results they got. I'm sure there are people who failed this course but could still turn out to be good programmers.

The question of what is a good programmer, now that would be interesting study? Well as long as the answer wasn't some kind of crazy formula.

3 Comments:

  • Check out this talk given by Paul Graham on what makes great hackers.

    http://www.itconversations.com/shows/detail188.html

    By Anonymous Anonymous, at Mon Jul 31, 06:13:00 pm 2006  

  • I don't think anyone knows how to make this determination yet. I mean, as a career professional I've seen so many educated people make poor programmers.

    I grew up programming, and then went and got the formal education on top of it.

    One of the biggest problems I have seen is that people who are able to get things "accomplished" often aren't aware of the larger pictures and slowly create monsters that require huge replacement projects. The other problem is people that try to create too general a solution and end up with something monstrously complex.

    Somehow, I think there is an aspect of art to this... and I'll enjoy this field as long as there is.

    By Anonymous Anonymous, at Tue Aug 01, 12:26:00 am 2006  

  • @grokodile:

    Yeah, that's the whole smart/"gets things done" dichotomy that Joel Spolsky describes here.

    Basically:

    - Smart people who can't get things done create generic, complicated monstrosities
    - effective people who aren't smart do things the dumb, unmaintainable way

    By Blogger Harold Combs, at Tue Aug 01, 01:58:00 pm 2006  

Post a Comment

<< Home