Tuesday, 8 January 2008

Follow the money

Computer Science Education: Where Are the Software Engineers of Tomorrow?.

Summary: Industry/Academics complaining about how schools do not adequately prepare students to be the software engineers (SWEs) of tomorrow.

My Opinion

Before reading this part, I would mention that I am in full agreement with the above linked communication (except maybe the Ada part: tried it, hated it!)

Who are the SWEs of tomorrow that they are talking about? They are the people who can implement (and therefore understand) web browsers, virtual machines and operating systems. Ok, there are more things here like graphics and programming languages, but I can't list them all :-)

But are universities responsible for this state of affairs? I would submit that universities have become businesses and so they are only interested in providing what customers (students) want. And what do students want? Good grades and a piece of paper with their name on it. When I went to school, there was rarely a person who wasn't interested only in these two things. I recall how much that frustrated some professors who genuinely wanted to enlighten their students.

The authors of the above are complaining to the wrong people. The universities don't care. They just want money. So do the students when they graduate. If you make it hard for them to make money (say, by outsourcing) guess what? They aren't going to join your industry. Now sure, there will always be those who are wired to do computer programming, but if you want to find them, you need to open your wallet otherwise someone else will.


Leslie P. Polzer said...

The question is, who decides what's adequate and what's not.

Is a CS graduate an adequate subject for the industry if he's able to do the standard share of Java and C(++) programming? I bet.
Will he be particularly efficient with those tools and his knowledge? I bet not.

So the basic problem is probably a wrong image of effective software engineers:
request a bunch of code monkeys and you'll get those. Oh, we don't have enough, though, sorry.

In a parable: it's like using O(n^4) algorithms and then complaining that there's not enough computation power.


Sohail Somani said...

In a roundabout way, I suppose I was trying to say the same thing. The blame should be on the industry for wanting cookie cutter programmers that "know the syntax".

I can point out many cases where it would take 10 people who "knew the syntax" twice as long to do something as one super programmer. For some reason, the industry just doesn't see these productivity metrics.