Re: Putting Perl Back on Top in the Fields of Scientific and Financial Computing

[spx2]

Also knowing Perl well requires more time than knowing Python well.
You need time to study Perl's quircks and variables, to give only a few examples:

- perlvar
- perlop

Also, the book Modern Perl, if everybody agrees that it's such a nice book and all, should be integrated in perldoc or in Perl's official documentation, because only in this way people can have easy access to it (It should be distributed with the perldoc package in Ubuntu for example).
PDL is pretty slow, also, compared to other available software for doing what it doesn't provide the speed, ease of use of other pieces of software.
Also, a weird thing is that PDL depends on OpenGL, so on a machine without X you can't actually do anything with PDL(or you can but it takes you some additional $amount_of_time).That is not normal since machines that only crunch numbers needn't have X on them.
Also, how portable is PDL actually ? Haven't tried it on windows.
So Python is used for financial and scientific computing because its syntax is easier to learn.
Perl is something that was an improvisation and it's starting to get more organized. But nobody knows that it's organized because they don't have the time to read all the blogs in the Perl "ecosystem" and go on IRC and ask a ton of questions on various channels. So if they include the "Modern Perl" modules and docs in the documentation that comes with every Perl distro then maybe the situation will change.

But in bioinformatics I think Perl is used a lot, I see recurring questions from students in bioinformatics here on perlmonks and some books are out on that and the BioPerl package is pretty vast, lots of people worked on it and I'm pretty sure it's development was well-funded by some big companies in the domain.

[chm]

PDL is pretty slow, also, compared to other available software for doing what it doesn't provide the speed, ease of use of other pieces of software.

I've been a PDL developer for a few years now and am the current release manager. This is the first time I've heard any reports that "PDL is pretty slow". I would appreciate references and benchmarks for this.

Also, a weird thing is that PDL depends on OpenGL, so on a machine without X you can't actually do anything with PDL (or you can but it takes you some additional $amount_of_time). That is not normal since machines that only crunch numbers needn't have X on them.

OpenGL is not X and X is not OpenGL. We're moving the baseline graphics capability to one based on OpenGL because it is the best common denominator for 2D and 3D display across all the major perl platforms.

Also, how portable is PDL actually? Haven't tried it on windows.

See the CPAN Testers matrix for the PDL-2.4.7 stable release at http://matrix.cpantesters.org/?dist=PDL+2.4.7. For current and accurate information on PDL, I refer you to our web site at http://pdl.perl.org

[JavaFan]

Also, the book Modern Perl, if everybody agrees

Muhahahaha. *Everyone agrees*? This is the Perl community. Even the relative small number of people on p5p (the people that usually do the actual work on the documentation) cannot agree on anything.

Perl is something that was an improvisation and it's starting to get more organized.

Perl organized?

But nobody knows that it's organized because they don't have the time to read all the blogs in the Perl "ecosystem" and go on IRC and ask a ton of questions on various channels.

Ah, it's organized because it has too many blogs and channels for a sane person to follow!

I would say that Perl isn't organized. Larry Wall != p5p != TPF != YEF != OSCON != ... "Modern Perl" is just someones view of how one can program. A view that has some followers. A view that also many people disagree with.

Perl is all about "there's more than one way of doing it". And that's not limited to having more than one way of programming a basic thing. It extends to the entire "cloud" around it.

So if they include the "Modern Perl" modules and docs in the documentation that comes with every Perl distro then maybe the situation will change.

Unlikely. If only that most people don't even read the documentation.

[spx2]

<quote>Unlikely. If only that most people don't even read the documentation. </quote>

that's because the documentation is bulky sometimes and for a beginner it's hard to quickly find what they're looking for, IOW the documentation is a reference.

I would say, put that famous book in the docs and see what happens. I would expect a positive outcome. Am I right chromatic ?

[chromatic]

I would say, put that famous book in the docs and see what happens. I would expect a positive outcome. Am I right chromatic ?

The core documentation has enough problems without another 70,000 words added to it.

I wrote the book and deliberately skipped some features, argued against using others, and ignored the entire default OO system in favor of Moose (before explaining the default object system as something you might have to maintain). It's one way to learn to write Perl 5 effectively. I hope it's useful for people, but it's certainly not the only way.

(Also I'd hate to have to debate stylistic choices on p5p before updating it for a new version.)

[JavaFan]

that's because the documentation is bulky sometimes

Nah. What I mean is, people don't read documentation. Period. Not of their VCR, and not about software either. Many wouldn't even know how easy or hard it is to use the Perl documentation because they never tried.

Besides, if the problem is bulky documentation, adding an entire book isn't reducing that problem.

I would say, put that famous book in the docs and see what happens.

I'd say, submit a patch for that to p5p and see what happens.

Am I right chromatic

Chromatic doesn't seem the most objective person to ask about it.