Front page | perl.perl5.porters |
Postings from July 2007
Re: Beyond 5.10
From: Brandon Black
July 1, 2007 05:09
Re: Beyond 5.10
Message ID: email@example.com
On 7/1/07, Johan Vromans <firstname.lastname@example.org> wrote:
> demerphq <email@example.com> writes:
> > I'm with you here. XML libraries are just too big a ball-o-wax to
> > bundle in the core. Way too many portability issues.
> If we, the experts, find it hard, what do you expect from our
> customers? Are you really surprised that many people and organisations
> do no longer take Perl into account when deciding on new software
> I think we're missing opportunities here.
I think you're focusing a bit much on what you would like to see in
the core versus what the whole world really needs in the core.
Obviously, I could come up with a list of a number of modules I
consider important in my domains of Perl programming, but a lot of
other Perl programmers would see them as useless core bloat. There
are a lot of Perl programmers (and companies, and sub-communities)
that would find XML in the core to be completely superfluous and
useless. It's just really not a great example upon which to rest your
Back on the more general track, the way to give people easy access to
lots of add-on modules without being "experts" is really the job of
either "core+" distributions that have been discussed here (you could
make one yourself), or alternatively, the resposibility of the people
who package up Perl + Modules for the various operating systems Perl
runs on (like, the ActiveState guys, or the various people who package
up .rpms, .debs, etc for the major linux distros, etc). That's how
you make it easy for a true end-user to install "fooapp" and have it
easily depend on lots of non-core Perl libraries (and I think the
distribution guys do a reasonable job on this already).
If the non-experts you're referring to are Perl programmers
themselves, I would hope anyone trying to write code against the API
of a module would be able to CPAN that module themselves.