To 6 or not to 6

ruoso has asked for the wisdom of the Perl Monks concerning the following question:

As some may know, my field of work is information systems. I've been working with this for some time now, and from time to time, I develop a new framework to make my work better. Some, unfortunally, got lost in the proprietary world, but notabily one, is free software (Perl Oak).

It's been quite a long time (1 or 2 years) since I started thinking about raising Oak2 from the ground up, I've made some work (available on CVS only), but nothing really relevant.

But this year I got wonderful news. A project that I submitted to FINEP (a brazillian's government reasearch agency) was approved, and I'll have resources for machines and people for 2 years. The project aims to raise Oak2 to version 1.0 in the end of the 24th month (eXtreme Programming concepts used, so we'll release 1.0 with all we can get on this time frame)

Well, considering the 24 months will start counting in the next couple months, I was wondering if it's sane to bet on Perl 6, or if I should stick with Perl 5...

UPDATE: The title was /(66|[^6]{2})/ BTW, for those who doesn't like poetry, the title means "To 6 or not to 6", which clearly is a parody of the famous /(bb|[^b]{2})/, or, "To be or not to be"...

daniel

Comment on To 6 or not to 6

Replies are listed 'Best First'.
Re: To 6 or not to 6 by GrandFather (Saint) on Jan 27, 2006 at 18:11 UTC
It may be that even were Perl 6 available right now it would still be wise to stick with Perl 5 for a major project involving a team the presumably hasn't much Perl 6 experience. DWIM is Perl's answer to Gödel	[reply]
Re: To 6 or not to 6 by duff (Parson) on Jan 27, 2006 at 18:57 UTC
It is definitely too soon to start betting on perl 6 since we don't have the actual perl6 compiler yet. Sure there's pugs and if you can live with the reference implementation running on top of ghc and can live with random changes as the language (and implementation) evolve, then maybe you would want to bet on perl6. Personally, I'm going to wait until there's a real compiler (I figure 2-5 years before we have a compiler that's stable enough for production use) duff	[reply]
Re: To 6 or not to 6 by dragonchild (Archbishop) on Jan 27, 2006 at 18:34 UTC
The Perl6 project is actually two projects - there's the design of the Perl6 language and the design of the VM, called Parrot. The cool thing about this is that Perl5 will run on Parrot through a project called Ponie. This means that you don't have to bet on Perl5 or Perl6 because both will be supported on Parrot. My criteria for good software: Does it work? Can someone else come in, make a change, and be reasonably certain no bugs were introduced?	[reply]
Re^2: To 6 or not to 6 by traveler (Parson) on Jan 27, 2006 at 19:14 UTC
Note that searching for ponie on CPAN returns as its first hit Acme::Pony which is totally unrelated :-)	[reply]
Re^3: To 6 or not to 6 by nothingmuch (Priest) on Jan 30, 2006 at 09:06 UTC
Personally I prefer ACME::POE:::Knee. -nuffin zz zZ Z Z #!perl	[reply]
Re: To 6 or not to 6 by samtregar (Abbot) on Jan 27, 2006 at 18:41 UTC
When the clock starts ticking download this link to get a stable version of Perl: http://www.perl.com/CPAN/src/stable.tar.gz I suppose there's some chance it will be Perl 6.x.x, but I doubt it! More likely you'll get the latest 5.8.x which is working great for me at present. -sam	[reply]
Re: To 6 or not to 6 by CountZero (Bishop) on Jan 27, 2006 at 21:24 UTC
Definitely stay with Perl5 or you may find yourself having to wait until the last day of your 24 months before you can start working on your program. Not even eXtreme Programming will save you then! Of course there a number of Perl6 concepts already implemented as modules in Perl5: search for Perl6 on CPAN. CountZero "If you have four groups working on a compiler, you'll get a 4-pass compiler." - Conway's Law	[reply]
Re: To 6 or not to 6 by nothingmuch (Priest) on Jan 30, 2006 at 09:12 UTC
You should stick with perl 5. You can use perl 5 today much more easily than e.g. pugs, which is definately not a stable platform (albeit lots of fun). I do suggest you look at, and familiarize yourselves with Perl 6, the pugs project, the synopses, etc, and design something that is easily portable to Perl 6. This is mostly a design issue - Perl 6 has roles, grammars, better support for higher order functions, etc. This means that while some things look equally ugly in Perl 5, one of them might be beautiful and easy to read in Perl 6 - so this could affect your decisions. Much more important, IMHO, is community involvement and interoperability. If you want this Oak project to work out well, please make sure that: You have a publically accessible respository, where people are encouraged to contribute (especially tests and documentation) You have an IRC channel and/or mailing list so that people can meet, get help, and group think You reuse familiar modules from the CPAN, especially for small mundane tasks, or contribute/replace modules that aren't fitting. Keep things generic! Keep things very simple - many CPAN modules are hard to use because of an installation headache. Since Perl is so portable this can actually a pretty tough issue to deal with. This also applies to your interface, but of course you already knew that =) Invest some funded hours in the tasks that you are not likely to get volunteers to do, but that are equally important - documentation, high quality tests, etc. These bits of advice are mostly from watching the success and failure of the Catalyst project. -nuffin zz zZ Z Z #!perl	[reply]
Re: To 6 or not to 6 by Elian (Parson) on Jan 30, 2006 at 19:13 UTC
Writing code in a language which doesn't fully exist yet seems rather ill-advised, the same way that writing to a library that's not functional or basing a project on software or hardware that doesn't yet exist is ill-advised. Doesn't matter if it's perl 6, Duke Nukem Forever, or whatever. (The same advice, several years ago, would've been applicable to betting on Apache 2 or Mozilla or Boston's Big Dig) This isn't to say that perl 6 won't exist. It may, it may not, and when it does you may or may not care. Betting a project on another project that is significantly incomplete and whose completion date has a significant amount of uncertainty is just unwise. This isn't to say you shouldn't use some of perl 6's features, not at all. The perl 6 modules on CPAN all provide interesting and useful functionality, and if you think that perl 6 may be the way you want to go in the end then it'd be sensible to use them where you can, to make transition easier.	[reply]
Re: To 6 or not to 6 by adrianh (Chancellor) on Jan 29, 2006 at 19:28 UTC
Well, considering the 24 months will start counting in the next couple months, I was wondering if it's sane to bet on Perl 6, or if I should stick with Perl 5... Of course there are other languages too :-)	[reply]