wpeckham wrote:

> I like what Raiph has brought to this

Thanks for the thumbs up. Fwiw I was disappointed with the way I expressed my exasperation with sundialsvcs and apologize for that.

----

NevemTeve wrote:

> Maybe we could see Perl6 as a new solution, a solution for a problem we haven't found yet.

Perl 6 was indeed designed to solve a wide range of problems that haven't yet really bitten deep but will in the future. An almost arbitrary pick of three is:

* The need to transition to having default, or at least easily used, string types that model C=G (Character=Grapheme). Devs are only just beginning to realize that what users call a "character" is what Unicode calls a "grapheme". The consequences are profound, and that's an understatement/euphemism. And Unicode 9 turns the screw on forcing even Western devs to deal with this due to Unicode's evolution of the nexus between emojis and graphemes and the rapidly ballooning popularity of using emojis even in Western text. Afaik Perl 6 was the first language in the world to tackle C=G head on and I note that, as of this month, of the 135 languages that have solutions to the rosettacode string length task (display how long a string is), only 3 do C=G-by-default -- Perl 6, Elixir and Swift -- and almost all the others don't even bother to *attempt* it. When so few languages are able to give a correct answer to how long a string is you know something's seriously and deeply wrong.

* Easy concurrency and parallelism. I'm not going to get into arguing whether Perl 6's (lack of a) type theoretic (I mean that in the formal mathematical sense) approach to this is a strength or weakness. I'm just going to claim it's becoming ever more important that newbies, both newbie sysadmins and newbie devs, can easily produce, consume, and compose concurrent and parallel computations in one liners and beyond.

* The online Perl culture. Quite a few online Perl places were pretty toxic by 2000. Perl 5 folk have done a lot of work to try to improve things but Perl 6 places got a fresh start followed by 16 years worth of consistently emphasizing the need for a fun and friendly environment, from Audrey's -Ofun meme to Larry's scare-curmudgeon Camelia. As the two sub-communities slowly reunite (which is happening even if most folk aren't aware of it) they're creating a nice new blend of having-fun and getting-it-done. We're seeing comments from old Perl 5 hands like yesterday's "I'm having so much fun with this. Some of the silliest little things just tickle me. Having to make so many new twisty brain paths... This really is wonderfully stitched together ... Honestly, I wasn't expecting it ... It's strange how you can get so much more built-in stuff, yet still not be constrained. ... And even with that freedom, there seems to be more inherent order". (http://irclog.perlgeek.de/perl6/2016-09-30#i_13317869)

----

sygOO wrote:

> for inspiration (satisfaction ??) I have looked into LUA and dart and golang (amongst others) in the intervening years

Oh sure. To quote my last comment on http://blogs.perl.org/users/steven_h...roduct-on.html I'd say the P6 perspective is akin to "please go learn them if you have the time and inclination" because we want what is learned to feed back in to the evolution of Perl.

> Ultimately, the community will decide.

Indeed. I think Perl is going to stick around for decades with Perl 6 playing an increasingly important part because of my faith in the correspondence between anarchism (of the beautiful non-violent variety) and open source. We shall see.

1 members found this post helpful.

Thanks, raiph.

Those rosettacode examples[0] don't show very well why grapheme length is important. A more clear example would be one where the character length and grapheme length are different. Behold...
Just you wait, I'm sure the masses will be beating down Perl6's door so they can get the length of their zalgo strings!

I also just want to note that the Perl6 compiler kindly let me know that I should use "~" instead of "." for string concatenation. That's a nice touch.

[0] https://www.rosettacode.org/wiki/String_length

It seems rather strange to me that a programming language would be seen as "fun" when my present job consists of riding herd on somewhat less than a half-million lines of the stuff that was written over the course of the past twelve years.

"Stuff" that does over $3 million of on-line revenue every month . . .

"Stuff" that is partly responsible for the jobs of more than 150 other people.

The concept of business risk does not seem to enter into the minds of language designers these days. They wrote (sometimes, quite by accident) a tool that became the linchpin of widely-deployed commercial applications, but today they don't seem to perceive a duty to what has been done using what they made. In PHP: "we decided that we didn't like the MySQL interface or the 'old' regular-expression functions, so we took them all out and didn't give you a plug-compatible replacement." In Perl: "we wrote an entirely new language, unrelated to the old one and decided nonetheless to christen it, 'Perl' Version 6. But the implementation is so kewel ..."

When people say that "there is considerable business risk in using open-source software that is designed by committee," this(!) is what they are talking about. In the past year, all three of the mainstream open-source implementation languages (Perl, PHP, Python) have wandered far from their roots in search of some abstracted programming-language Nirvana.

2 members found this post helpful.

> Thanks, raiph.

This has been a stressful thread for me and my responses have been more like books than comments. I really appreciate this encouragement.

> Those rosettacode examples[0] don't show very well why grapheme length is important. A more clear example would be one where the character length and grapheme length are different.

Thanks for providing the example.

The thing I most struggle with is that even *mentioning* the word "graphemes" is probably way more confusing than just drawing attention to the fact that calling codepoints "characters", and the count of codepoints "character length", is monumentally misleading. Some may say it's just a matter of words but I think that abusing the word "character", in the way that the Unicode spec and that rosettacode page arguably do, is, I think, a big part of the reason why almost no western devs have yet figured out that they're hurtling headlong toward literally the ultimate Unicode character issue.

> Just you wait, I'm sure the masses will be beating down Perl6's door so they can get the length of their zalgo strings!



Just to be clear, I don't think anyone much cares about dealing with zalgo strings. But I'm guessing about half the planet cares about text in their native language. A good deal of chinese, indian, etc. text is, effectively, full of zalgo strings...

> I also just want to note that the Perl6 compiler kindly let me know that I should use "~" instead of "." for string concatenation. That's a nice touch.

The Perl 6 standard for error messages is that anything that's LTA is worthy of improvement. (LTA = Less Than Awesome)

While raiphs' exortations have at least nudged me into going back and having a look at P6, the elephant in the room is the concerns raised in post #33.

sygOO wrote:

> the elephant in the room is the concerns raised in post #33.

Really? I don't understand why. But I'll go ahead and respond to it.

> It seems rather strange to me that a programming language would be seen as "fun"

Many proprietary businesses ignore the "fun" that is essential to the health of the open source projects that underlie their codebases and dev teams. They have a business to run and a way they want to run it and it doesn't include supporting "fun".

Fortunately the open source community is explicitly OK with businesses ignoring "fun".

But the "fun" *within those open source projects* (as against codebases that build on them) is the only truly sustainable fuel of long term healthy free (as in speech, not beer) maintenance and improvement by a diverse pool of devs.

Audrey Tang, who recently became Taiwan's youngest ever minister and who wrote the first partial Perl 6 compiler, invented the -Ofun meme about a decade ago and explained why "fun" is strategically important in open source projects in her "-Ofun: Optimizing for fun" presentation that she delivered a few years ago.

I'm not aware of any video recordings but the slide deck resonates for me and the devs I've shared it with. (Impatient LQers may wish to skip to slide 32 which gives a shout out to Linux before moving on to "Two Kinds of Fun", "Why Fun?" and how to increase "fun" in an open source project.)

> The concept of business risk does not seem to enter into the minds of language designers these days.

Perl 6 and Perl 5 are independent of each other, each with its own successful sustainability strategy, but each specially enabled to make good use of each others' capabilities. Having both options will ultimately *reduce* the business risk of being a Perl user even if Perl 6 makes basically zero difference at the moment and won't make much of a difference for another few years imo.

> They wrote (sometimes, quite by accident) a tool that became the linchpin of widely-deployed commercial applications

There's serendipity in all things in real life but you appear to be dismissing the entire early Perl community as an accident. I was around at the time and what I saw wasn't an accident.

> they don't seem to perceive a duty to what has been done using what they made.

What I see is a steadfast commitment to the position Larry staked out in 2000:

"We all like Perl 5 a lot. We all use it a lot. Many commercial interests will guarantee that Perl 5 continues to be well-maintained and stabilized for quite a few years to come ... We are determined to do the right thing by everyone. To this we pledge our lives, our fortunes"

and to repeat something I quoted earlier in this thread:

"It is our belief that if Perl culture is designed right, Perl will be able to evolve into the language we need 20 years from now. It's also our belief that only a radical rethinking of both the Perl language and its implementation can energize the community in the long run. In the long run means 10 and 20 years down the road. Finally, it is our belief that Perl 5 will be better supported [by developing Perl 6] than it would be if we merely tried to guard what we already have."

> When people say that "there is considerable business risk in using open-source software that is designed by committee," this(!) is what they are talking about.

"this(!)" seems to be your stories about what's happened.

My story is that both Perl 5, including CPAN, and Perl 6, inlining CPAN, are designed by anarchist forces that populate worlds designed by one mind, Larry Wall's. It's the opposite of designed by committee.

Bingo! So, call "Perl 6" by a different name because it isn't "Perl Version 6."

@Larry's new "pet project" is shamelessly riding on the coat-tails of a venerable name ... without actually being a successor of it. So be it. There has already been a pretty damned-good "successor to Perl," and the name of that language is "Ruby." There have been half-a-dozen code-names used in the Perl6 project, all of which would be good "official names." And the formal adoption of any one of them would eliminate vast amounts of confusion.

If you gather a thousand developers into a room and put up a slide that says -Ofun, most of those thousand developers might chuckle. (It seems to me mostly to be an odd sort of inducement, meant to persuade more people to participate in a project without getting paid, but nevertheless.) The bottom line, to me, is that "billions of dollars of on-line transactions, directly and indirectly supporting thousands of jobs," does not qualify as fun ... nor, for that matter, the lack thereof.

A programming language core ... this being @Larry's ostensibly-sacred contribution ... actually, these days, is "not that much." If Perl did not have CPAN ... if PHP did not have PEAR ... if Ruby did not have "Gems" within easy reach ... and so on ... no one would be paying attention. [i](Likewise, if "Linux" were "the Linux kernel.")

Programming languages (and, operating systems ...) are important because of what they facilitate ... and, have facilitated. Your "innovations" are destructive if they raise the threat of not guaranteed-successfully compiling (running ...) one hundred percent of that code: code that you didn't write.

If the "Perl-6" (sic) team had simply acknowledged the truth ... that they, like Ruby, were embarking on the design of an entirely new and different (and maybe better, we'll see ...) language ... then they would have at least been "truthful" rather than "deceitful."

It's perfectly fine to create a new language that will leave "Perl" in the dust. (Ruby already made a pretty damned good first-pass at that, by the way ...) Just, be truthful about it: don't call it "Perl."

I quit!

Interesting. Do you also get upset that gcc-fortran and gcc-pascal unllike gcc or gcc-c++ have nothing directly to do with 'C' as a language?

I see nothing deceitful. They clearly and publicly made these decisions when they started, and made it pretty clear what they were starting and why. That you, here and years later, are suddenly disturbed by the name conventions involved really speaks to your mindset, not any supposed deceit on anyone else's part.

But, enough of all this. Why do we not go code in Object Oriented COBOL for a while, and watch "Two Cellos" on YouTube. It will be more productive.

Nonetheless, they could have come up with another name, like Lrep or Pler or whatever.

All of these are portions of the "GCC Compiler Suite," which is a unified compiler that has a number of language-specific front ends and a fairly generic back end.

A better analogy would be to write an altogether new language vaguely based on COBOL and call it FORTRAN, either because you owned that trademark or because you thought that calling it FORTRAN make it more popular.

If, as others have observed, you decide that "~" is a much "keweler" string-concatenation operator than ".", such that you actually expect(!) us to adapt hundreds-of-millions of lines of installed code to suit what is nothing more than "your latest fancy," fuhgeddaboudit.

There's always room in the world of computing for one more programming language. Just don't call something by somebody else's name. (We sort-of have a word for human-beings who try to claim lineage or ancestry that they do not properly have, and that [unprintable] word begins with the letter "b" ...)

sundialsvcs wrote:

> We sort-of have a word for human-beings who try to claim lineage or ancestry that they do not properly have, and that [unprintable] word begins with the letter "b" ...

Your choice of analogy is revealing.

A "love child" (one of the synonyms google lists for bastard) has their natural (biological) parents irrespective of whether they are a bastard, irrespective of whether you or anyone else thinks they don't "properly have" the lineage and ancestry that they do actually have.

In fact, calling someone (or Perl 6) a bastard, intending its archaic derogatory meaning, reveals almost nothing about such a "natural child" (another synonym).

Instead, the main thing your bastard insult suggests is that you are willing to cast an empty slur based on an appeal to authority where the authority is the analogical equivalent of a *wedding registry* (the definition of a bastard is someone born out of wedlock).

But maybe there's something special about the authority to which you're appealing. I recommend always questioning authority and I powerfully distrust appeals to authority but who knows? If we can figure out to which authority you're appealing we might learn something.

Larry Wall designed, implemented, named, and shared Perls 1 thru 6. So he's a hot prospect for being *the* authority on the lineage and ancestry of Perls. But he says he sees Perl 6 as a Perl and has declared that v6.c is "official Perl 6" so you must have decided to question and reject his authority on what to call it.

The Perl Foundation "is dedicated to the advancement of the Perl programming language through open discussion, collaboration, design, and code." They also own the Perl trademark, so they're particularly apt as an authority that's an analogical equivalent to a wedding registrar. But TPF also sees Perl 6 as a Perl so presumably you also reject them.

perl.org's out too: "Perl 6 is a sister language, part of the Perl family, not intended as a replacement for Perl 5, but as its own thing - libraries exist to allow you to call Perl 5 code from Perl 6 programs and vice versa."

Just who or what is it that you trust more than Larry Wall, TPF and perl.org? Could it be you are *appealing to your own authority* and then leveling a bastard insult based on nothing more than that? Could it be that you don't even understand why that's a problem?

1 members found this post helpful.

You really don't have to psychoanalyze me, nor my comment. My use of "the b-word" was simply in reference to "a thing (person) who pretends to be what s/he is not." A pretender to someone else's rightfully-earned throne.

The developers of Perl, including the person who owns the trademark (and who can therefore both prevent the name from being attached to something and who can attach it to anything he pleases), have created a new language that is not related to its predecessor ... in syntax or in implementation ... but have confusingly chosen to refer to it as "version '+1.0'" of an entirely different language that they also own.

There are obvious marketing reasons for this. "Radoku" (or whatever it was ...) would have no more chance in the marketplace than "Haskell" and would probably be compared quite unfavorably to "Ruby," which is a ship that went-to-the-blocks only 5 years earlier than Perl-6 but which was launched quite promptly ... a simple fact which has made all the difference, commercially speaking.

Sixteen(!) years after the first discussions of a practical successor to Perl-5 first began (during which time the "Moose" set of extensions to Perl-5 appeared and blossomed), The Perl Foundation has created to great fanfare a language that ... cannot be practically and safely used by any "shop" that had standardized its operations on Perl-5.

No one puts new wine in old wineskins. You can't take an entirely new programming system, graft the name of a highly successful older system onto it (christening it as "version +1.0") when in truth it bears no relation to it.

PHP and Python recently wandered away from their installed-base and now find themselves "split" between a version that isn't being accepted and one that isn't going to go away. But neither of them attempted to do this, and to brazenly try to sell it on brand-name alone.

1. Whoever said that PERL 6 was a "new version" of Perl 5? Larry never said that, or anything LIKE that, as far as I know.

2. With all that there is going on in our world to cause outrage, you latch onto THIS?

This is not a problem, it is only your obsession. I respectfully suggest that there are better uses for your time and outrage.

I'm not particularly riled-up about it ... sixteen years after the fact, this language is not that important anyway ... but I do think that it is a serious mistake for the future of "the language that still is 'Perl'" to deliberately attach the same, venerable brand-name to an entirely unrelated thing.