[Gllug] C/C++ mentor

[Aaron Trevena]

On 29/11/06, Pete Ryland <pdr at pdr.cx> wrote:
> On 29/11/06, Aaron Trevena <aaron.trevena at gmail.com> wrote:
> > You don't need to grok everything the language has to offer at all.
> > I've been doing Perl professionally for both small and large projects,
> > everything from quick cgi scripts and one liners, to CRM, Supply Chain
> > Management and Aviation. No more than twice, since I was a newbie,
> > have I come accross some Perl in my work that was outside of what I
> > already I was familiar with, and I'm no Perl Guru, I've never needed
> > to touch XS/SWIG, internals, B::, are more than 10% of the debugger.
>
> This is possibly true within a given industry and country, but travel
> a little bit and you'll find things done in completely different ways.

I don't know about that. I've worked professionally with coders from
NZ and SA and didn't find they used different idioms, nor have the any
of the coders straight out of India used different subsets of perl
than anybody else with the same skill level.

> Funnily enough, I have used both xs and swig for wrapping C code
> (they're both very easy to use, and in fact swig uses xs) and done
> some OO stuff in perl.  In fact, I think there's a lot of people on
> this list that indeed use OO regularly in perl.

Well, anybody who writes it for money, I would expect, I certainly
wouldn't hire or even sub-contract to somebody who didn't.

> > In fact, it's well known that you only need to know a small subset to
> > get a job done, that's why sysadmins use it, cgi newbies used it back
>
> Yes, you only need a small subset to write something, but I was
> talking about reading others' code.

No difference. Most programming is dealing with other peoples code.
I've only been relatively lucky to work on greenfield development in
my two current contracts, but even one of those I have to spend a lot
of time reading their code, and I'd say that whats far more different
between companies is how they design their db schema.

SQL and DDL have a fairly small syntax, but some people seem reluctant
to use anything outside of their particular understanding, whether
it's using subselects instead of joins, the horrors of de-normalising
or not even normalising, etc.

I'm working with clients who don't have much experience in Perl (and
some of them don't even have much of a background in analysis and
programming), and the biggest problems aren't people using subsets of
perl, but lack of analysis and database design wisdom.

> > C++, Java and C#  all have a LOT more to learn, just compare a
> > helloworld program at the start of any book on the subject, and don't
> > get me started on the horrors in C++ like the STL. *shudder*.
>
> In terms of syntax, they are all much smaller than perl.  Libraries
> are another thing, and that's what references are for.  You certainly
> don't want to be using a reference to understand the *syntax* of a
> program.

I don't think I ever used a reference to understand the syntax of a
Perl program since I was a newb, and I found I had to use references
to grok C++, when I started learning it again, far more than when I
learnt perl.

> And actually, verbosity of the helloworld program is a sure sign of a
> simple syntax.  Take 'bf' for example.

Right - and that's *easier* to grok different peoples code from then?

A.

-- 
http://www.aarontrevena.co.uk
LAMP System Integration, Development and Hosting
-- 
Gllug mailing list  -  Gllug at gllug.org.uk
http://lists.gllug.org.uk/mailman/listinfo/gllug