develooper Front page | perl.perl5.porters | Postings from June 2006

Re: Its time we set the score straight on Perl 5 and Perl 6 and debunkour own self generated FUD.

Dean Arnold
June 16, 2006 08:45
Re: Its time we set the score straight on Perl 5 and Perl 6 and debunkour own self generated FUD.
Message ID:
Tom Horsley wrote:
> On Fri, 2006-06-16 at 16:18 +0200, demerphq wrote:
>> We need to enhance the visibility of Perl in the
>> corporate sphere. We need to find useful features that the market
>> wants and start delivering them. 
> I happen to love perl for getting quick, one-of, temporary
> jobs done very easily, but if you really want perl to have
> corporate use for real long term projects, you are bound to
> run into people like me who wouldn't touch perl for production
> use with a ten foot pole (sorry :-).

Actually, I feel sorry for you. By relegating Perl to such
a menial role, you're missing a great opportunity for significant
productivity improvement, as well as the ability to leverage
all the great domain-specific solutions available on CPAN.
You may also need to examine why you (or your organization) are
unable to develop production quality Perl, when others - including some
very large ecommerce concerns - have been quite successful in doing
so for many years.

> Perl really is a write-only language, 

What leads you to that conclusion ? Yes, Perl permits line noise
to be compiled and executed...but that doesn't mean you have
to write your Perl that way. By your reasoning, you'll need to throw
out all those systems that use 'C' to code their core, since C
likewise permits very concise, severely obfuscated syntax, and thus
simply can't be trusted for production use.

Perl is like the English language: one can choose to write
Shakespearian sonnets, or public restroom vulgarities. Its a matter
of choice and discipline.

> and you just can't do
> production work write-only - someone has to maintain production
> code, and the language is so fantastically huge, no two programmers
> even appear to be writing in the same language.

Again, where's the discipline ? What of perltidy, naming standards, etc. ?

> I'm not sure what feature you could deliver to address this
> problem, but there's your first challenge :-).

Presumably, all you appear to need is an enhanced perltidy that
can enforce naming standards, and flag any overly
sigilistic bits of code ? Perl Programming lists via nntp and http.
Comments to Ask Bjørn Hansen at | Group listing | About