[erlang-questions] "Erlang as a First Language" -- crazy? or just stupid?
Sun Dec 20 10:35:53 CET 2009
Returning to your original question, teaching Erlang as a first language is
neither stupid nor crazy. It's just not a sound decision.
Forget the syntax, since it just scares away those used to C-like languages.
When I got used to it, I realized that Erlang syntax is very consistent.
However, when teaching a first programming language, syntax is not the main
issue. A friendly environment is. My first programming language was C on
Windows. Not the easiest language to learn, but thank to the IDE, the
comprehensive library, examining a faulty program or checking out the
documentation was a breeze.
Therefore, please do not evaluate just the syntax of a language. Evaluate
the environment instead. Put yourself in a newbie's shoes and ask yourself
how would the environment help you when things will not work and when you'll
have to learn, install and use a new library?
2009/12/20 Michael Turner <leap@REDACTED>
> That's what I'm wondering. I'm holding out for a possible Answer #3:
> "Sure, at least if they are smart enough."
> I might have "smart enough" covered. I'm working on a project in the
> overlap between cognitive science and linguistics. These people aren't
> necessarily math-heavy (on the linguistics side, anyway), but they can
> tolerate odd notations, abstruse jargon and fine conceptual distinctions
> that would evoke only dread, if not nausea, in ordinary folk.
> I've been evaluating Erlang for this project in the only way I think
> possible: after having identified what's important and what's not in a
> language/system for this project, and seeing that Erlang roughly matches
> up, I started writing modeling code.
> Here's where the matchup is clearly bad, from a lifecycle point of view:
> Erlang is still a small minority language. Sure, it's possible that a
> mini-gold-rush is starting for it, and if so, that's good for Erlang,
> mid- to long-term. But in my experience, the sudden appearance of money
> can make the programmer supply situation worse in research circles:
> hackers in academia start getting head-hunted ferociously. What I'm
> working on is likely to remain a lab creature for years, if it's viable
> at all. Somebody besides me has got to be able to support it,
> eventually if not sooner. So I'm thinking maybe the more adept users
> might be good candidates for the role of programmers as well.
> But you see the problem: it might mean that Erlang is effectively their
> first language. Maybe they had a little exposure in high school or an
> elementary programming course in college. At most. And maybe that
> experience even turned them off a little.
> What are the first-language learnability issues with Erlang? I'm poorly
> qualified to think about this. Although I wouldn't say I learn
> programming languages quickly, I don't have major blocks either. Over
> the last 40 years, I've probably written something more significant
> than "Hello, world" in more languages than I have fingers. When I run
> across a term like "atom", I think, "probably lisp-like", and I'm
> off to the races. I never really hacked Prolog, but having done it a
> little helped a lot with with Erlang. Even Pascal helped: that nagging
> little terminator-vs-separator distinction with semicolons and whatnot
> had migrated up into my ulnar tendons since college (so *that's* why
> they hurt sometimes!), but made it back down to my fingers pretty
> quickly when called. So taking up Erlang in my mid-fifties is only
> making me think, "Why didn't I start with it before my hair went
> totally grey?"
> Maybe this isn't the best list for asking this question, because
> probably most of you are like me anyway, in spirit if not in age. But I
> can't think where else to ask.
> -michael turner
> erlang-questions mailing list. See http://www.erlang.org/faq.html
> erlang-questions (at) erlang.org
More information about the erlang-questions