I don’t understand why Perl 5 and 6 are still tied together as if they are the same language. There’s no compatibility between the two. You can’t write a library that works in both versions, so you lose CPAN and the thousands of packages that make Perl so powerful. I don’t think there’s even any <i>reliable</i> way to translate Perl 5 source code.<p>This is a vastly larger change than python 2->3, and people are still using python 2.x. In short, there’s no sensible upgrade path, Perl 5 & 6 are very much their own language.<p>If Perl 6 wanted backwards compatibility, they needed to have thought about it a long time ago. Trying to hack in some Perl 5 style functions into it now seems futile...
I cut my teeth on Perl 5 in the early 2000s, and I've been curious about Perl 6 for a long time.<p>Last year I sat down with Perl 6 for long enough to form opinions on the language's merits, which are many. I'm told many of my criticisms have been addressed in the five months since the article was published, but perhaps some of you will enjoy reading the original review. Cheers!<p><a href="https://www.evanmiller.org/a-review-of-perl-6.html" rel="nofollow">https://www.evanmiller.org/a-review-of-perl-6.html</a>
Perl Pessimism: <a href="https://trends.google.com/trends/explore?date=all&q=perl" rel="nofollow">https://trends.google.com/trends/explore?date=all&q=perl</a><p>I have happy memories of Perl, and it started me on the path that led to a nice career of Python awesomeness. Thanks, Perl! That said, I can't imagine a plausible scenario where I'd ever consider using it again. I almost certainly wouldn't use it professionally because it's not exactly going to light up a resume. Yes, there are still Perl shops. But no, there aren't very many of them, and I'd be highly suspicious of an engineering department still writing new code in it today without a very good reason.
I spent a huge amount of time in my 20's writing Perl 4 & 5 scripts. Even though I no longer use Perl, I still feel nostalgic about it.<p>That said, I can't find any reason to pick up Perl 6 and start learning it. With so many other useful and interesting languages out there: Python, Elixir, Elm, Rust, etc... what appeal should Perl 6 have to the masses these days?<p>Is anyone jumping into Perl 6 like they did with Ruby a decade ago and saying, "Wow, programming just got fun again"?<p>I'd love to see a Perl 6 breakthrough, but honestly it feels like those attempts to keep BeOS going... too little, too late. Its time has passed.
Perl 6 appeals to me enough to make it my first new language to study, after a 20-year hiatus from any meaningful programming. This could be taken as an anti-endorsement. But for someone with no legacy skills, and no legacy codebase to draw on, I can afford to be arbitrary. What sold me is the cheat of storing floating point numbers as rationals composed of integers. Also when I tried python I missed the curly braces.
Another year, another chance for all involved to learn the important rationality skill of recognizing sunk costs and opportunity costs, and give up when giving up is the best option.<p><a href="http://lesswrong.com/lw/gx/just_lose_hope_already/" rel="nofollow">http://lesswrong.com/lw/gx/just_lose_hope_already/</a>
I hope that the Perl 6 core devs take notice of this and other threads from the past days. Somehow I think that the efforts have not been directed enough at things that perhaps are not that exciting for them but that matters a lot for potential users.<p>Examples:<p><pre><code> - Better migration/transition from Perl 5
- Performance and stability
- Killer features and explaining these thoroughly
- Best in class tooling, editor support, super easy deployment
</code></pre>
I am sure there are efforts, but perhaps not enough to meet the demands?<p>I say this of course realizing that resources in an open source project are limited. But something has to happen to accelerate the usage.<p>Keep on writing more articles etc to spread information outside the P6 core dev echo chamber.
I used perl when dinosaurs still roamed the land. Can somebody explain why it took so long for Perl 6? I read first p6/parrot vm textbook as a schoolboy more than a decade ago.
Would people feel differently (lots of talk in here along the lines of "why bother at this point?") if Perl 6 had a completely different name?
My first programming job was perl back in '97 for the LAMP stack. There was a lot of neat things that I did with it back then... and clever too. I remember writing a neat Tie::Array that was backed by a database... and then undoing it because it was too clever and not something that was maintainable.<p>The last time I touched perl professionally was in '09 during the perl winter. The web stack was moving in other directions - especially that of Java (I'm a corporate back end line of business keep the gears of the company turning type). I write boring code - which is good. Boring code doesn't surprise people and when I pull it up from 10 years ago, it still works. I try to avoid clever code now - it takes too long to get back into the mindset that I had when I wrote it. I'll admit to one bit of clever code with annotations on an enum that I've written in the past half decade... and that is in place to force what would be runtime errors to be compile time errors instead.<p>I looked at perl 6... there's even a bug report with my name on it ( <a href="https://github.com/rakudo/rakudo/commit/8d04bec" rel="nofollow">https://github.com/rakudo/rakudo/commit/8d04bec</a> ). I really want to like the language - it makes me think about some things differently... and here's the "but". But it all feels too clever. Code needs to be reasonable - something that I can come back to later and pick up again and continue on with minimal switching of the mind. Unicode operators, Junctions (as neat as they are), redefining operators... its letting me and <i>encouraging</i> me to write clever code.<p>I pull up the docs and see things from <a href="https://docs.perl6.org/language/operators" rel="nofollow">https://docs.perl6.org/language/operators</a> that make me wonder if its gone too far. There's more than one way to do it... but there are more of them than Java 8's list ( <a href="https://docs.oracle.com/javase/specs/jls/se8/html/jls-15.html" rel="nofollow">https://docs.oracle.com/javase/specs/jls/se8/html/jls-15.htm...</a> ). I feel that there's more than one way is "more than one approach" not "more than one way to type '+'".<p>The code just doesn't feel reasonable anymore.<p>For what its worth, after not touching perl professional since '09, this past week I was called on to fix a LAMP stack. There's the CGI with a nice 'use CGI.pm' (the last change time on the file was from 2011). I was able to open it up, read it, debug it, and fix it in under an hour. It was straight forward, sensible, and reasonable code. I shutter to think what it will be like in another decade if someone comes by and asks someone to debug some 10 year old perl6 code and they open it up to find unicode scattered through it and trying to remember what =~= or ∘ does.<p>For my scripting needs now... I've been a Java coder for the past decade. When I want a script I'm more likely to fire up groovy. It runs, its stable, and it doesn't try to make me write clever code.<p>Sorry perl... you're just not my go to language for thought to code anymore.. and thinking in perl6 just isn't something that I'll find myself easily writing once or being able to figure out when coming back to later.