Hey /g/ why is Perl 5.10 the best programming language ever?

Were you a dotcom webdev in the late mid/late 90's? That would be your only excuse.

>>58446858
>perl
kek, C# .NET is better because only Apple knows how to make the best programming languages.

>>58446881
If it wasn't for Perl, Linux wouldn't be what it is today.

>>58446858
Perl 5 is obselete with the release of perl 6.

>>58447128
Perl 6 is shit. Only fucking posers code with Perl 6.

>>58446905
So it could've been actually cool ?

>>58446858
didn't all the normies stay in Perl and everyone else went to Ruby or Python?

>>58447998
Yeah, that's why Ruby and Python aren't as good as Perl 5.

I have a lot of warm memories associated with Perl, but even I am willing to admit is extremely far from perfect.

>>58447128
Perl 6 will never be finished. It's a meme.

Also, no decent neural network development library for Perl, and none seen on the horizon. Python has Theano and Tensorflow, along with multiple frontends. Perl is a great language with rich history, but it's close to death now. I don't warite anything in Perl anymore except for run-once programs (and Perl's great for those).

>>58446858
Know this... when you go to the pharmacy in north america and pay for your prescription with insurance... Perl 5 is there in the background. It will be for a long time.

>>58446858
>I built a massive software infrastructure in Perl
This. I've personally been using Perl for nearly 15 years and never found a problem I couldn't solve using it

>>58448093
>Also, no decent neural network development library for Perl
https://metacpan.org/pod/AI::FANN for standard Feed forwards networks for Tensor Flow, I've used Inline to write an interface to one of the languages that are supported, not perfect but works well enough to be usable

>>58449904
>FANN
>no GPU support

You know where to put it, anon.

>>58446858
Why did they make Perl 6 uncompatible with Perl 5?

>>58449946
They planned (still plan?) to make it compatible. Original idea was if you put v6; at the start of your program, it's written in perl6; otherwise, it's written in perl5 - and perl6 interpreter should be able to execute it properly.

>>58449924
That's not a Perl problem though

>>58449992
>Also, no decent neural network development library for Perl

What's a good book to learn Perl?

>>58450249
The Camel Book

>>58448093
who gives a shit, though?
Neural networks are a meme for most but some very trite problems where statistics actually work.

In fact, for those problems - say guitar downtuning effects without artifacts - it often isn't tried.

>>58450669
>for most but some very trite problems where statistics actually work.
You can't say that after Google's Tensorflow beat #1 Shogi player 10 years ahead of when everyone thought computers would.
You can't say that after waifu2x was released.
Neural networks might have been a meme some time ago. But not anymore.

>>58450249
chromatic's modern perl and Dominus' higher order perl

>>58450688
>You can't say that after Google's Tensorflow beat #1 Shogi player 10 years ahead of when everyone thought computers would.
I can, though. It's not an interesting problem except for two big autismal companies 1-upping each other with board games.
In fact, the speech synthesis and the sound-from-visuals generation were way more interesting, if they were ready for launch.

>You can't say that after waifu2x was released.
I can damn well say that, because that's one of the trite applications I spoke about before.

Deep faggotry is one of those things that get really old really fast once you grok the basics, just like compiler contruction. And just like compiler construction, there is a limit on what the advance of that technology can archive, in particular when related to CV applications.

>>58451130
>It's not an interesting problem
It's an unprecedented real world result impossible with any other tech. It doesn't matter if you're interested in it or not.

>there is a limit on what the advance of that technology can aRchive
And the limit moves further and further away every year.

>>58451149
>And the limit moves further and further away every year.
Nah, it's pretty clear what can be archived with it.

>>58451239
That makes no sense. No one expected the outcome of the Shogi match.

>>58451271
I did.

>>58451312
I do not believe you.

>>58446858

Ruby guy here.

I have much respect for Perl, it does some things different from Ruby, but writing Perl is fun. Also bonus points for not using one of the normie stacks (MEAN, LAMP, Golang).

What are good frameworks for Perl? I've heard Catalyst is OK, are there others?

File: L2taZCY.png (113KB, 500x592px) Image search: [Google]

113KB, 500x592px

>>58451354
proof me wrong, [visible]feggit[/visible]

>>58451271
>>58451312
I pretty much expected it too - they wouldn't have done it if they weren't super confident they'd succeed, or at least come close to succeeding (so they could at least hype up how advanced they were). Only the media, and people who don't understand how large companies work wouldn't have expected this.

>>58451418
You expected it after they announced it. What counts is expecting it before they did. So before they announced the match, you'd think, oh, neural networks have greatly advanced lately - I think we're good enough to make a Shogi AI, even though the same AI algorithm that worked for Chess is not computationally feasible for Shogi for at least 10 more years.

>>58451414
I don't have to. I just don't believe you.

>>58451529
>I've lost the argument, so I'll desperately move the goal posts and hope that no one points out what a faggot I sound like

>>58451529
>I don't have to. I just don't believe you.
Alright.