Subject: Re: making 'make replace' safer
To: None <firstname.lastname@example.org>
From: Martijn van Buul <email@example.com>
Date: 07/16/2006 21:36:06
It occurred to me that Raymond Meyer wrote in gmane.os.netbsd.devel.packages:
> I've never used ruby, so I don't how much extra bloat it carries. Why use
> another language when something could be done in C?
Following that logic, why not write it in assembly?
> Unix is written in C so that's what everybody needs to use.
This is quite nonsense. Every language has its own specific strengths and
weaknesses. Assembly is fast and efficient, but not cross-platform and
rather tedious to write applications in. C is just fine for low-level
routines and straightforward calculations, but it's not very object
oriented, making it rather unelegant sometimes, and making good software
design bothersome. C++ is just fine for a lot of purposes (I really don't
grok your antipathy for C++, it doesn't add "bloat") but the lack of a
gargbage collector will annoy some, and its close relation to C makes it
look like a hack every now and then. Pascal has its own merits. The list
goes on and on.
A *good* programmer chooses the correct language for the problem at hand.
If you want to write a log file parser perl might be your friend, if you
want to make a spiffy webpage php beats any CGI-script hands down, a
GUI of some kind might benefit from C++ or (excusez-moi le mot) managed C,
etc. And I haven't even touched Java, Scheme, lua, ruby and python.
Insisting on only C is shortsighted, blunt, and not very professional.
Not to mention selfish and disrespectful.