Beefy Boxes and Bandwidth Generously Provided by pair Networks
Pathologically Eclectic Rubbish Lister
 
PerlMonks  

Re^2: The Future of Perl 5

by raiph (Chaplain)
on Aug 19, 2018 at 22:04 UTC ( #1220659=note: print w/replies, xml ) Need Help??


in reply to Re: The Future of Perl 5
in thread The Future of Perl 5

"In ten years Perl will have this and that and for that we have to break compatibility in a few places. Only as necessary of course!"

Here's what he actually said in his speech announcing the current P6 project in 2000:

"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. Itís also our belief that only a radical rethinking of both the Perl language and its implementation can energize the community in the long run."

Replies are listed 'Best First'.
Re^3: The Future of Perl 5
by chromatic (Archbishop) on Aug 19, 2018 at 23:58 UTC

    He also actually said:

    Now, this is not going to happen quickly. We expect to have alpha code a year from now, or some definition of alpha. We might even ship it, but we expect it to be well-designed alpha code.

    ... and:

    Basically what we are saying at this point is if we are going to bite the bullet and require translation of Perl 5 to Perl 6, that really means that we can consider anything that still allows us to translate most scripts. Now we do not expect to be able to translate a 100 percent, but if we can translate with 95-percent accuracy 95 percent of the scripts, and 100-percent accuracy 80 percent of the scripts, then thatís getting into the ballpark....
      Larry said these words within hours of the conception of Perl 6.

      Like much of what Larry has said over the years about his thinking about the long term big picture, the stuff I quoted has always made sense to me.

      In contrast his talk of an alpha was essentially him relaying an estimate from the core group, a group that had already decided they wanted Larry to stay out of project management or implementation.

      The plan allotted him just 4 weeks to go from initial design concept to finished specification once the RFCs were in. Then, instead of the anticipated couple dozen RFCs, a few hundred of them rolled in.

      By mid September 2000 anyone paying attention should have known that the estimate from the group for an alpha (which Larry had simply relayed on that first day) was going to be years off.


      The translation narrative still makes sense to me. I also see Inline::Perl5 as an appropriate solution in many scenarios.

        By mid September 2000 anyone paying attention should have known that the estimate from the group for an alpha (which Larry had simply relayed on that first day) was going to be years off.

        Would you then suggest that subsequent communications from the core team about an 18 month schedule (take, for example, the original project manager upholding his 18 month timeline from mid-September 2000) were made in bad faith?

        That's not helping your argument.

        I suppose you could point to this end of September update that the schedule would move by *two weeks*, but that's also not helping your argument.

        Seems weird to hew so closely to "what Larry actually intended in the announcement" and then immediately walk it back with "but everyone knew that was a lie by September" while jumping way over what Larry actually said in September.

        I am continually amused and disappointed by P6 advocates trying to revise history when that history is so easily searched.

      So we expected some definition of alpha in a year, did we get it? Or did the feature creep only just started to gain speed a year after the announcement with the specs getting (some definition of) finished only several years later?

      Jenda
      Enoch was right!
      Enjoy the last years of Rome.

        So we expected some definition of alpha in a year, did we get it?

        We did not. It took about 10 years or so.

        Or did the feature creep only just started to gain speed a year after the announcement with the specs getting (some definition of) finished only several years later?

        I never realized this until I re-read the link when raiph posted it, but Larry's observation about the time between releases getting longer and longer applies even more so to the Apocalypse documents, which reached 6 or 7 before being replaced with something else. I find that amusing.

Re^3: The Future of Perl 5
by Jenda (Abbot) on Aug 20, 2018 at 21:56 UTC

    Either the Perl culture was not designed right or that belief was wrong then.

    And there was more than this speech back then. While it was clear from the start that total compatibility was not a goal, the resulting hodgepodge of ad hoc changes is not what was promised back then.

    Jenda
    Enoch was right!
    Enjoy the last years of Rome.

      Either the Perl culture was not designed right or that belief was wrong then.

      Or your analysis is wrong.

      From my perspective the haters were winning the war in 1999/2000 but now the war is essentially over, haters have limited impact in the Perl community, and Perl is set to evolve nicely over the next decade.

      And there was more than this speech back then. While it was clear from the start that total compatibility was not a goal, the resulting hodgepodge of ad hoc changes is not what was promised back then.

      You have misunderstood the scope of change that was talked about at the start for Perl 6; the coherent design that has emerged; and the way the latter is indeed consistent with what was promised back then.

      The whole point was that Perl 6 needed to be a radical break if Perl was to remain attractive to ultra creative thinkers like Damian Conway and hundreds like him:


      Anyway, this thread should be about the bright future of Perl 5 and I'm going to switch my focus to that.

      I reject the thesis that Ovid's discussion of improving Perl 5 is old hat, uninteresting, and doomed to failure. I accept that it took uncomfortably long to get to where we are for Perl 5 as well as Perl 6 but I don't buy that it's worthwhile debating whether it could or should have been done any other way and I don't buy that the thing to do right now is to pour cold water on good ideas like the ones Ovid suggested.

        From my perspective the haters were winning the war in 1999/2000

        My perspective is actually the complete opposite. At the Y2K, Perl was pretty much at its zenith. It was attracting some of the best minds, was widely deployed, CPAN was expanding apparently exponentially and mod_perl (v1) was arguably the jewel in the crown of IT at the time. The only "haters" in evidence were the ColdFusion advocates as their baby was being utterly thrashed by mod_perl.

        It was later on that Perl started to fade and division and hostility emerged. After the release of 5.6 there was a bit of a brain drain as some of the brightest were lured away to work on P6. The Perl core stagnated for a good few years (although not completely - unicode and threads made some good gains but those were mostly considered esoteric by the majority of the then perl user pool unfortunately) and PHP, the great pretender, achieved FPM and started the takeover. It wasn't until the release of 5.10 that we started to get the impetus back in terms of development but by then a lot of the user base had left. The opportunity had gone - maybe forever.

        The future of Perl is that it will continue to improve in terms of efficiency and features but that fewer and fewer people will care. But I will still be one of them.

        What haters? There were a few trolls back then, some unix lovers showed a bit of hostility to Windozers, but generally the community was way bigger and way livelier back then. I do not see Perl being set to evolve nicely over the next decade. I see Perl6 continuing to attract tens and Perl losing ground steadily.

        "coherent design". Mkay. Sure.

        The radical break may have succeeded in making the not-at-all-Perl remaining attractive to creative thinkers like Damian Conway and (more like) tens like him at the expense of tens of thousands of others. Good job!

        Also you are mixing up things that were written with things you made up. Yes the discussion is old hat, but it is interesting to quite a few and no one but you claimed it's doomed to failure. Debating whether it could or should have been done any other way is of course pointless. We don't have a time machine to go and convince Larry to, at the very least, NOT assign the next version number to the project.

        Jenda
        Enoch was right!
        Enjoy the last years of Rome.

        "in my experience the Perl world seemed by 1999 to host a lot of unempathetic unproductive hostile arguing"

        That was my experience, as well, but let's call that subjective.

        What is objective, is that Perl peaked around the millennium, and today is generally considered outdated. And people generally think that, regardless of the numeral that follows the name perl.

Log In?
Username:
Password:

What's my password?
Create A New User
Node Status?
node history
Node Type: note [id://1220659]
help
Chatterbox?
and the web crawler heard nothing...

How do I use this? | Other CB clients
Other Users?
Others making s'mores by the fire in the courtyard of the Monastery: (3)
As of 2019-06-18 00:41 GMT
Sections?
Information?
Find Nodes?
Leftovers?
    Voting Booth?
    Is there a future for codeless software?



    Results (81 votes). Check out past polls.

    Notices?
    • (Sep 10, 2018 at 22:53 UTC) Welcome new users!