[oclug] Yet Another Language Discussion (was Re: Python )

[Rod Giffin]

David F. Skoll said:
> On Fri, 20 Jun 2003, Rod Giffin wrote:
>
>> Programmers tend to use a small set of tools, not always choosing the
>> right ones for the job, but ones they are more familliar with.
>
> Well, at some point, you get overwhelmed because of limited time.  I'm
> very familiar with shell scripting, Perl, PHP, Tcl, C, C++, Lisp, awk,
> and sed.  I'm not familiar at all with Java or Python.  I figure "one of
> these days", I'll look at them, but I have limited time and very little
> motivation to learn either Java or Python, because I can't see how
> they'll help me get my work done.

I can understand that, but it seems you have a rather complete tool kit at
your disposal in the first place.  Unless you start working in web
services environments, there is likely going to be very little use for
Java in your skill set - and even then, there are a few alternatives.

Besides, programmers who have a well rounded set of tools at their
disposal can generally apply the same skills to a new language, and become
expert programmers in a new language in relatively little time.  It's not
like there isn't a huge overlap in skills involved.  You would probably
find your ramp up speed for either Java or Python would be measured in
days at the outside.  That's because you have a good skill set to start
with.

This is not what I've found to be "average" in the industry.  However, I
should point out that I have found that programmers that are on this list
tend to be a little more "worldly" with better skill sets than average.

>> You don't build usually build an interactive enterprise
>> class web service around Perl.  You can, and some people have, but
>> there are other tools that are better for the job.
>
> "Better" is subjective.  Although languages are like tools, the analogy
> breaks down somewhat.  The fit between a language and a task is not as
> clear-cut as the fit between a Robertson or Phillips screwdriver and a
> screw.

That's true... that's why I used a bit of an exaggerated tool comparison
of lag bolt and sledge hammer - it was the first thing that came to mind
really.

> There's usually no clear problem feature that dictates language
> A over language B.

Not when language A and B are abstract.  If language A and B could solve a
problem equally well, it doesn't matter which is chosen.  That is not
always the case though.  But I recognize there is also quite a lot of
overlap in the capabilities of certain... maybe even most languages.

Rod.