[SunRescue] stuff

[David Cantrell]

I subscribe to the digest, so I'll reply to several messages at once here:

Mike Meredith <hmv at meredithm.fsnet.co.uk> wrote:

> On Tuesday 03 April 2001 23:38, you wrote:
> > [taken to geeks list, please reply there if at all]
> >
> > "Chris Byrne" <chris at chrisbyrne.com> wrote:
> > > Honestly I do wish we were metric if only for the fact that it
> > > would make life easier when dealing with the rest of the world. But
> > > 350 some years of tradition dont get changed overnight.
> >
> > I personally wish that .UK *wasn't* gradually switching to metric. 
> 
> I suspect you're slightly older than me.

I'm [thinks] 27.

>                                           The advantages of switching to 
> metric (what everyone else in the world uses apart from the US) 
> outweighs any possible advantage that Imperial measurement uses.

So I'm told.  Not convinced though.

> People my age and younger have difficulty in dealing with Imperial 
> measurement because when it hasn't been taught it comes across as 
> bizarre.

I was always taught in metric, but I still use Imperial measurements for
just about everything in everyday life.

------------------------------------------------------------------------

"Chris Byrne" <chris at chrisbyrne.com> wrote:

> I think my main problem with perl is the bloat. It has grown and extended
> far beyond what it's original goals were. Normally this isn't a bad thing,
> but that growth has been very convoluted and , from my perspective, poorly
> managed.

The core language is still pretty reasonable I think.  I am less happy
about the gazillion modules which are now distributed as part of the perl
core.  I would also *like* to see various bits taken out of the core and
made into loadable modules.  eg, formats, networking, all the Unix-specifics
like username stuff, and their ilk.  This may happen in perl 6, I dunno.

> That said, it still doesn't condemn the language as a whole. It just
> encourages uses for which it is entirely unsuited, such as writing an entire
> multi-thousand line application with another entire multi-thousand line GUI.
> These could generally be much more efficiiently programmed in C or Java.

<nod>  Used well, perl can be suitable for such tasks, but it takes a
really good perl programmer to do it well.  I'd agree that Java is better
for many people doing that sort of work.  I myself tend to write front-ends
in Java and back-ends in perl.

> One good thing it has done is gotten a lot of people into coding. Of course
> the flip side of that is, as several people have mentioned, the tendency for
> those same people to want to do everything in perl.

And to do it badly cos such people tend to a) only use one language so
don't learn useful techniques like the lispy things you can do with map,
grep etc, and b) they tend to not work in disciplined teams.

> Perl is and always will be at it's heart a regular experession based, string
> oriented scripting language, and any attempt to use it for something else is
> probably inefficient at best, and nightmarish at worst.

There I have to disagree.  True, a program (not a script) written in perl
can never be as fast or memory-efficient as its functional equivalent
written in C.  However, programmer time is often of more concern than
execution time, and so perl wins in those situations.  Of course, you
still need to apply the same level of discipline to your development
process as you do in other languages, which is where plenty of perl shops
fail spectacularly.

------------------------------------------------------------------------

Steve Pacenka <sp17 at cornell.edu> wrote:

> For me at least, Perl holds up pretty well by most of these metrics.  CPAN is 
> an outstanding example of code reuse.

I'm not sure.  At its best, CPAN is fantastic, but it is poorly organised
and has fuck-all quality control.  There's some *real* dross in there and
without downloading, installing and wasting time *using* it it's hard to
tell what's hot and what's not.  In addition, there are no checks whether
a new CPAN entry replicates existing functionality elsewhere.

> It is possible to write unmaintainable code in any language.  (Obfuscated C 
> contest, anyone?)  Blame the coders.

Hear hear!

-- 
David Cantrell | root at alphacomplex.org | http://www.cantrell.org.uk/david/

    This is a signature.  There are many like it but this one is mine.

** I read encrypted mail first, so encrypt if your message is important **