On Tuesday 08 Mar 2011 17:41:19 Louis-Philippe wrote: > Some stand on the idea that learning a new language is worth it these days > if it brings you to new programming paradigms and/or new possibilities. In > that light, I see Bash scripters could be seduced by Perl, but the gang in > the Cool-Factor-Zone/Ruby/Python are mostly inclined to move toward Erlang, > Scala, Clojure, Haskell, Objective-C, Javascript... not Perl, because as > flexible and great as Perl can be it doesn't offer much new advantages to > these peoples.
Well, one thing I like about Perl is the fact that lists are not references: http://use.perl.org/~Shlomi+Fish/journal/36493 It allows for some nice tricks. I also like that in Perl operators are not ambiguous (i.e: we have ".", "+" and "," which are the same operator in Python and other languages). I also prefer Perl's Moose to Ruby's built-in OOP which seems much less powerful and less flexible (though Ruby still has a cleaner syntax). These may not be such earth-shuttering advantages. Naturally, Perl people often say they stay in Perl due to CPAN, which is very comprehensive and which has very good infrastructure. It's hard to know how many people will want to write a program that incorporates many different use- cases, that is more often than not, very possible using CPAN (so-called "mashups" - only not necessarily web-based ones). Some people told me that most programmers hate the "There is more than one way to do it" philosophy of Perl. And naturally, we may need to just create usable CMSes similar to WordPress or Drupal which people can install on their servers, and it also seems I was unhappy with all the web-devel frameworks I've tried (with Catalyst, that is the king of the hill being the complex and idiosyncratic framework that it is.) > I think Perl should learn to get back to enviable position > by looking at the programming language ecosystem in general and ask itself > where it is positionned in this jungle and what is it's unique offer. IMHO > one good cousin to compare to is Lua... there is not much Cool-Factor in > there, but it offer unique features: is really small, embedable, > extendable, fast (and by far the fastest), compilable to bytecode, and > else. Lua is a growing star, not because it has a special personality, > like Perl, but because it has features... Not that personality is bad, > look at Ruby and Python, both have strong personalities, but it's only a > small fraction of makes these killer languages. Perl need to know its > goal and perfect it. Some of you might think the goal is known and done, > but think again, Perl could shine brighter for what it is if it knew > pragmatically what it's destiny is. Right. Perl 6 has a lot to offer because as opposed to Ruby and Python which are an incremental step forward to Perl 5 (and sometimes a few small steps backward), Perl 6 is a huge step forward from all of them. Many of the Perl 6 features have been back-ported to perl 5 , and naturally while Rakudo is quite usable , it's still kinda slow, buggy, and very incomplete. I'm sure people are working on it, so it will be resolved at a point. Larry Wall said that the Perl 6 effort was "The second system effect done right.", and it's possible some second system effect projects in the past have proved to be popular. Well, Perl 6 does not aim to completely replace Perl 5 and perl5 and these languages will continue to evolve and be enhanced, but I still think Perl 6 should have been called differently (though it's too late to change now). Maybe we should ask ourselves which killer feature or features we'd like to backport from Perl 6 and implement it in perl5. Regards, Shlomi Fish P.S: a Python (and formerly Perl 4 and many other languages, down to an early version of Fortran) programmer I've talked with who had to study Objective-C to work on a Mac project said he thinks that Rubyists won't like ObjC because it's much less expressive and powerful than most dynamic languages. Some people have nicknamed Objective C "Objectionable C" or "Subjective C" or whatever. The main selling point of ObjC is the fact that it is the only usable language that can be used to write applications for the iPhone. However, while many companies would like to have an iPhone app just so they'll have an iPhone app that people would like (i.e: a bank or a site, or whatever), most of the developers who develop iPhone apps and wish to make money out of them have to pass through many hoops in the app store, buy a costly Macintosh computer to develop for it, and usually end up selling very few copies to be able to make a living from. In comparison to the Android, whose development environment is based on Eclipse, and which has a portable emulator, and which Google does not regular the App Store, the iPhone application developers are much unhappier, and I suspect it will eventually backfire at Apple. So I think many of the Ruby/Python or even Perl 5 crowd would prefer to learn C or even C++, which have far bigger utility and enlightenment than only ObjC. (Though, ObjC is a small extension over C and similar in its intentions to C++, but it does it very differently.) BTW, http://c-begin.wikia.com/ was recently started as the C, C++ (and possibly also ObjC) equivalent to http://perl-begin.org/ and being a wiki, you are all encouraged to contribute. -- ----------------------------------------------------------------- Shlomi Fish http://www.shlomifish.org/ Funny Anti-Terrorism Story - http://shlom.in/enemy Mastering 'cat' is almost as difficult as herding cats. -- http://www.shlomifish.org/humour/bits/Mastering-Cat/ Please reply to list if it's a mailing list post - http://shlom.in/reply .
