From: ········@bayou.uh.edu
Subject: Re: C++ briar patch (Was: Object IDs are bad)
Date:
Message-ID: <5las4t$h7k$1@Masala.CC.UH.EDU>
Chris Bitmead uid(x22068) (·············@Alcatel.com.au) wrote:
: In article <············@canopus.cc.umanitoba.ca> ·······@silver.cs.umanitoba.ca (Daniel J. Salomon) writes:
: >C++ is not a pretty language, and its is not a safe language, but you
: >have to use it for a couple of advanced projects before you fully
: >appreciate its contribution to language design.
: ROTFL.
Yeah you found it funny too. I still chuckle when I read the above
passage. The one thing that makes one realize what a crock C++ is
IS when one does large projects. C++ looks almost nice if all you are
doing is "hello world" programs (and if you came from a Cobol
background -- one of the few languages with the dubious distinction
of being inferior to C++). Add a few lines however, and then
the nightmare starts.
: Well, the last two big projects I've been on, edit/compile/debug times
: have blown out to about 3 hours, so you get plenty of time to
: contemplate this amazing contribution. It may not be positive however.
Hey, ever hear of a full day to compile? I'm not the one who ends
up doing it (I recompile portions which takes 15 minutes or so, and
that's on a Silicon Graphics Inidgo^2), but I do end up waiting for
these builds to finish. Guess what? They still get errors and
so an entire day is blown so someone can remove a trailing symbol
from some source line.
: Seriously though, the battle I always fight with C++ is practically
: zero prodictivity. It's like wading neck deep in molassas.
I see you think much more highly of C++ than I do.
: It doesn't matter too much about how clever STL is or is not. The only
: tactic that wins against the C++ compiler is minimising the number of
: compiles. That doesn't give me much scope to experiment with clever
: template thingys, even ignoring the fact that more templates will blow
: out compile times even more.
Amen. Forget incremental changes to code to test new ideas or
concepts -- not when compile times are this ridiculously high.
It's a damn shame too 'cause with legacy systems (at least
the ones I work with), documentation is like treasure -- you'll
be lucky if you find it, so experimentation with code is
about the only way you can be sure of understanding anything, and
when that entails waiting for ridiculous periods of time, well
let's just say that peace of mind becomes a scarce resource.
: >In short, there are solid reasons for the success of C++. It is not
: >just a fluke, or some kind of industry conspiracy. Critics should try
: >to include some solid facts in their attacks, instead of just throwing
: >out wild accusations or ridicule.
: Solid fact: If I wasn't using C++, I wouldn't have time to read this
: news group... I'd be doing actual work instead of waiting for
: compiles.
So in a way C++ has added to your liesure time at work :). I know
the feeling, I find myself frequently twiddling my thumbs, reading
newsgroups, or perusing web sites while waiting for compilations.
--
Cya,
Ahmed
In article <············@Masala.CC.UH.EDU> ········@Bayou.UH.EDU (········@bayou.uh.edu) writes:
>: Well, the last two big projects I've been on, edit/compile/debug times
>: have blown out to about 3 hours, so you get plenty of time to
>: contemplate this amazing contribution. It may not be positive however.
>
>Hey, ever hear of a full day to compile? I'm not the one who ends
>up doing it (I recompile portions which takes 15 minutes or so, and
>that's on a Silicon Graphics Inidgo^2), but I do end up waiting for
>these builds to finish. Guess what? They still get errors and
>so an entire day is blown so someone can remove a trailing symbol
>from some source line.
Well when compile times are 3hrs that's basicly 1/2 day. I mean, you
can't fit in a 3rd compile unless it's overnight. My last C++ project
took about 1-2 weeks to compile the whole system. You can imagine the
excitement when it is realised that someone had a bug in something
low-level, and you have to re-build everything from the bottom up.
>Amen. Forget incremental changes to code to test new ideas or
>concepts -- not when compile times are this ridiculously high.
You've got that right. You design something for C++, and you're locked
in right from the first line of code until 2 decades from now.
>It's a damn shame too 'cause with legacy systems (at least
>the ones I work with), documentation is like treasure -- you'll
>be lucky if you find it, so experimentation with code is
>about the only way you can be sure of understanding anything, and
Experimentation? Forget it. The current generation of software will be
completely unmaintainable in 5 years from now.
>when that entails waiting for ridiculous periods of time, well
>let's just say that peace of mind becomes a scarce resource.
>
>
>: >In short, there are solid reasons for the success of C++. It is not
>: >just a fluke, or some kind of industry conspiracy. Critics should try
>: >to include some solid facts in their attacks, instead of just throwing
>: >out wild accusations or ridicule.
>
>: Solid fact: If I wasn't using C++, I wouldn't have time to read this
>: news group... I'd be doing actual work instead of waiting for
>: compiles.
>
>So in a way C++ has added to your liesure time at work :).
Like I said. The really good thing about C++ is that it may achieve
what the economists couldn't - full employment.
>I know
>the feeling, I find myself frequently twiddling my thumbs, reading
>newsgroups, or perusing web sites while waiting for compilations.
Chris Bitmead uid(x22068) <·············@Alcatel.com.au> wrote in article
<······················@Alcatel.com.au>...
> >
> >: Solid fact: If I wasn't using C++, I wouldn't have time to read this
> >: news group... I'd be doing actual work instead of waiting for
> >: compiles.
> >
That all depends. <g> I'm working in SmartLisp (more of a Scheme than a
Lisp<shrug>), but rather than waiting on compiles, I wait for datamining agents
to find answers in multi-gigabyte knowledgebases. If I get curious about how
things are going, I just go over to the machine, hit the escape command and
inspect the running agent's variables before letting it proceed (or perhaps
tweaking it a bit).
The moral of the story: if you choose your problem domain very carefully, you
can get to cybersurf just as much as if you were using C++, but instead of
having to work so hard in between, the _machine_ is actually doing most of the
heavy lifting!
Now that's a language you can really get to like. ;-)
Bill House
--
http://www.dazsi.com
Note: my e-mail address has been altered to
confuse the enemy. The views I express are
mine alone (unless you agree with me).