comp.lang.ada
 help / color / mirror / Atom feed
From: "Richard  Riehle" <adaworks@earthlink.net>
Subject: Re: Why I like Ada
Date: Sun, 30 May 2004 18:06:18 GMT
Date: 2004-05-30T18:06:18+00:00	[thread overview]
Message-ID: <uopuc.17041$be.15411@newsread2.news.pas.earthlink.net> (raw)
In-Reply-To: pan.2004.05.26.19.11.04.359100@email.ro


"David Starner" <dvdeug@email.ro> wrote in message
news:pan.2004.05.26.19.11.04.359100@email.ro...

... in response to a conversation about debugging C versus Ada,

> Why do you think it's a function of programming language? It may be at
> a different level, but you can replace divisions with bit-twiddling,
> structured code with hyperoptimized gotos, or manually inline code in
> Ada just like C.

I don't know how much Ada, C, or C++ you have written, but I am going to
assume you have written quite a lot of each.

Back in the 1908's, when I first started coding Ada, I found myself
constantly
checking up on myself by resorting to the debugger, pretty much the same
thing
I did with my C code.   In time, I realized that I needed the debugger less
and
less with Ada.   This does not mean it is never necessary to use the
debugger.
In fact, I still like to use the debugger as an informational tool (what
kind of
code is being generated) rather than an error-correction tool.

With C++, I find myself going into debug mode much more than with Ada.
This is especially true when I am trying to figure out what is going on with
a misbehaving C++ program written by someone else.  Perhaps this is
because I am more comfortable with Ada than with C++.   Even so, when
I do need to debug Ada code, it is much easier, for me, than debugging C++
code.

So, for me, language does matter.

I am more and more becoming concerned with the notion of risk in software
engineering.   The more I compare the relative risks of programming
languages,
the more I realize that Ada, for me, is not as risk-laden as most of the
alternatives.
That is, it seems easier to create risk-free code in Ada than in C or C++,
while
achieving the same level of performance and flexibility.

Of course, your mileage may vary.

Richard Riehle





  reply	other threads:[~2004-05-30 18:06 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-05-22  0:39 Why I like Ada Fionn mac Cuimhaill
2004-05-22 14:11 ` Per Dalgas Jakobsen
2004-05-22 15:23   ` Preben Randhol
2004-05-22 16:27     ` Björn Persson
2004-05-25 21:29       ` Robert I. Eachus
2004-05-26  7:44         ` Peter Amey
2004-05-26  8:18           ` Ludovic Brenta
2004-05-26 16:15             ` Martin Krischik
2004-05-26 19:33               ` David Starner
2004-05-30 18:06                 ` Richard  Riehle [this message]
2004-05-30 18:17                   ` Randy Brukardt
2004-05-30 19:09                     ` Adrian Knoth
2004-05-31 13:27                     ` Björn Persson
     [not found]                   ` <inpkb0d57uiaf6970hk0ctj09orni4piea@4ax.com>
2004-05-31  2:07                     ` Richard  Riehle
2004-05-26 15:30           ` Wes Groleau
2004-06-01  3:57           ` Robert I. Eachus
2004-06-02  1:48             ` Jeffrey Carter
2004-06-02  5:13               ` Robert I. Eachus
2004-05-22 22:01   ` Wes Groleau
  -- strict thread matches above, loose matches on Subject: below --
1994-12-02  9:44 Richard A. O'Keefe
replies disabled

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox