Beefy Boxes and Bandwidth Generously Provided by pair Networks
Think about Loose Coupling

Re^2: The future of Perl?

by Your Mother (Archbishop)
on Dec 13, 2014 at 21:04 UTC ( #1110279=note: print w/replies, xml ) Need Help??

in reply to Re: The future of Perl?
in thread The future of Perl?

that damage bringed into community by perl6 in its current state is huge and continuing.

This is nonsense and Iím tired of seeing it here without the slightest supporting logic or evidence.

Perl5 was zooming up to a dead end 15 years ago; there was a ton of infighting, apathy, unwillingness to look outside, big gaps between releases, and attitude of if you wonít do it my way Iím taking my commit bit and going home. Perl6 shook that up, broke things loose, and got a lot of new persons interested in Perl5 and moving it forward. The Perl5 renaissance that began c 2004 might have never happened without the Perl6 efforts. Moose, MOP, Moo, Mouse, and the advances that have arguably made Perl5 a superior OO language to Python and Ruby and Java only came in because of Perl6 discussions and dreams.

Replies are listed 'Best First'.
Re^3: The future of Perl?
by BrowserUk (Pope) on Dec 13, 2014 at 21:27 UTC
    Moose, MOP, Moo, Mouse, and the advances that have arguably made Perl5 a superior OO language to Python and Ruby and Java

    I'd like to see that argument!

    I'm not against it; I'd just like to see:

    1. the benefits of Moose/Mop/et al. laid out;
    2. a comparison with the OO in those other languages.

    My perception of Moose is that it is an expensive getter/setter generator. My perception of MOP is that is provides for introspection -- that has few if any legitimate uses.

    I'd like to be persuaded that I'm wrong.

    With the rise and rise of 'Social' network sites: 'Computers are making people easier to use everyday'
    Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error.
    "Science is about questioning the status quo. Questioning authority".
    In the absence of evidence, opinion is indistinguishable from prejudice.

      Up front: I am not the right one for the job; I am largely recycling things Iíve read and heard. My experience domain in OO in other languages is sparse. So, Iíll rely on an expert!

      I'd say the Moose OO model is far more powerful and useful than Python's. Roles, method modifiers, lazy builders, delegation and coercion are all really good for reducing repetitive code and improving OO design. With enough work most of that can be emulated in Python, but Moose gives it to you straight out of the box. ĖRe: Python OO v Moose

      Perl supports a vast ground of different techniques and does not hinder from crossbreeding or intermixing. Types + coercion for example, only hinted in the quote, can be bondage oriented or loose and accommodating or anything in between. You can write Perl that does what Java does in the way it does it (though private methods and such are less fun, you know they can be done in at least a couple ways; anonymous/lexical code and inside-out). You canít do the reverse. Every personal sweet spot in between is possible in Perl. And though some languagesóas I understand it, Iím *not* a xlang OO exert in the slightestómake some of the techniques easier and remove some of the bad behavior (like methods bumping into functions)Ö still Perl can do it.

      Now to recycle another thing someone else said recently: The main problem with Perl isnít Perl, itís that while the CPAN is enormous, Perl at large is not grounded enough in complete, out of the box, applications. WordPress or Wikimedia would have been easy to write in Perl for example but the ethos of the Perl community, in my estimation, finds it more fun to work on gears than complete machinesÖ Iíve thought a lot about this and Iíve never come to any conclusions; why, is it better, should, could it change? PHP only exists and flourished, in my view, not because Perl couldnít do what it ended up being but because the Perl hackers of the day wouldnít.

      I ducked most of the real question, I know. It would be a good bit of work to address for someone who is an expert. It would be a three month research project for me. :P

        What would be the Moose version of this (a queue that maintains a fixed size by evicting the oldest items) ?
        class FixedSizeQueue def initialize(size) @max_size = size @items = [] end def pop @items.shift end def push(item) @items.push item if @items.size > @max_size pop end end end
        With usage
        2:16% irb -r ./fsq.rb irb(main):001:0> q = => #<FixedSizeQueue:0x9317c6c @max_size=3, @items=[]> irb(main):002:0> q.push 2 => nil irb(main):003:0> q.push 3 => nil irb(main):004:0> q.push 5 => nil irb(main):005:0> q => #<FixedSizeQueue:0x9317c6c @max_size=3, @items=[2, 3, 5]> irb(main):006:0> q.push 7 => 2 irb(main):007:0> q => #<FixedSizeQueue:0x9317c6c @max_size=3, @items=[3, 5, 7]> irb(main):009:0> FixedSizeQueue.instance_methods(false) => [:pop, :push]

        I heard this argument that in process of Perl6 development perl5 got better OO, 'mop', Moose , etc etc etc

        I am not bying this, and here is why:

        Not a secret that most of these ideas was already in the air, and invented some 20 or 30 years ago...
        and not a secret that LISP had better and superior OO system (CLOS) which was then stolen into the perl6...

        In my vision, the people that were busy with choosing OO system for perl6 could spent their efforts and develop perl5 mop OO system directly from LISP to perl5, if they wish to.

        What happened with Perl6 - is that at y2000 there was a number of inspiration based on success of perl5 and on fitire perl5++ which is to be perl6...
        what happened then - after 10 years - the project stale, and then existence of the name 'perl6' became more of a problem, and damage to the community started happening.

        You could accept that or not, but this is what happened.
        Not recognizing this problem is another problem, which makes resolution even harder.

        I was also inspired by perl6, and very much like its syntax and power....
        but I do not see any usable implementation of perl6...

Log In?

What's my password?
Create A New User
Node Status?
node history
Node Type: note [id://1110279]
and the web crawler heard nothing...

How do I use this? | Other CB clients
Other Users?
Others taking refuge in the Monastery: (5)
As of 2021-02-28 23:02 GMT
Find Nodes?
    Voting Booth?

    No recent polls found