Why Functional Programming Matters

"Why Functional Programming Matters" (1984) by JohnHughes. Available at http://www.cse.chalmers.se/~rjmh/Papers/whyfp.html

The examples in the paper are in MirandaLanguage, which uses LazyEvaluation like HaskellLanguage.
Rats, more "academicy" examples. I was hoping for some biz-app examples. Researchers need to "get out" more. --top


Not really what you're looking for, I know. I think you've got to get into academia and scratch that business-app-empirical-research itch of yours on your own, because no one is going to scratch it for you.

There is, of course, a strong inertial resistance to functional programming in business. However, it's unlikely to be due to FunctionalProgramming itself, as business is inertially resistant to any change.

I've noticed that a significant number of my final year undergraduate IT, Computing, and Games Programming students have expressed -- for the first time in the seven years that I've been teaching full-time -- a genuine eagerness to learn functional programming, along with a certain subtle derision of ObjectOriented programming. As these students will eventually become IT decision-makers, we will no doubt see a move toward functional programming in business.

...Unless, of course, the industry stagnates dramatically or flies in some completely new and unanticipated direction.

This sounds like a YouJustDontGetIt claim. See HowToSellGoldenHammers. And, I am not against FP in general, but merely see it as one of many tools/techniques. -t

You've read something into my writing that was not intended. I'm left wondering how.

It's related to "business is resistant to change". The implication is that they ignore something that is "good" just because they don't want to change. Perhaps you meant it a different way and I switched to defensive mode prematurely (a C2-induced habit).

I meant it a different way. It was intended purely as an observation.

In general I don't see business resistant to change. Often they are the opposite: they hop on some buzzword in fear of being left behind competitors before the buzzword has been sufficiently road-tested. Many businesses live in constant fear of being one-up'd by competitors, and this propagates to IT policy. Perhaps this is an AmericanCulturalAssumption being that any "stable" or commodity business tends to drift overseas where the labor is cheaper. Once an activity becomes a commodity, it tends to be offshored.

Businesses are not reluctant to try new things. Many businesses, for example, are currently trying CloudComputing in pilot projects. How many of those businesses will significantly change how they work because of it? I suspect the "cloud-based app" (or two) will simply join the suite of maintained applications, some of which still date back thirty years to the day the long-gone mainframe system (or its equivalent) was last shut off. Or, in some cases, the day the still-present mainframe system was turned on.

But I am glad that "multi-paradigm" is getting more attention of late, including FP. It's refreshing after the "everything should be an object" push of the past decade or so. The hard part is agreeing what part of an app belongs in which paradigm.

I agree.
CategoryFunctionalProgramming CategoryPaper

View edit of April 3, 2011 or FindPage with title or text search