Beefy Boxes and Bandwidth Generously Provided by pair Networks
Problems? Is your data what you think it is?
 
PerlMonks  

Re^2: Waiting for a Product, not a Compiler

by chromatic (Archbishop)
on Nov 28, 2011 at 06:58 UTC ( #940318=note: print w/ replies, xml ) Need Help??


in reply to Re: Waiting for a Product, not a Compiler
in thread Moose - my new religion

Again, I have no interest in telling volunteers what to do, and I'm not Ian Hague or a leader of TPF or a grant manager, so I have no interest in telling paid developers what to do, but I'll tell you the honest truth as I see it: given the history of every Perl 6 implementation so far (and Pugs and Rakudo in specific), I see a lot of interest in writing a compiler but very little evidence of desire to bundle that compiler into a product that real people can use to write and maintain real programs.

(Before you tell me I'm a hypocrite for complaining instead of volunteering, I started contributing to Parrot in 2001 and Perl 6 in 2003. I've paid enough sunk opportunity costs, thank you.)

I believe the current Perl 6 implementations have, as you adroitly put it, "fundamental flaws that aren't being addressed by any future plans for ... development that I've seen so far".

Does the most recent Rakudo Star release represent a worthwhile point for someone to write a serious program in Perl 6? Does the most recent compiler release?

How much work would the average non-committer have had to do to keep a serious program running on the monthly releases? How much research does this entail? (How much work would this have required since the first Rakudo Star release?)

How many non-core modules pass their tests (or run at all) on the most recent compiler release? On HEAD?

Given the Morton's fork of using a stable but buggy release of Rakudo (for which all development has stalled) and an unreleased version with admirable improvements but serious regressions, what is your average user interested in Perl 6 but not interested in writing a compiler to choose?

What's Rakudo's bus number?

Was Rakudo Star a "useful release"? Was it what people hoped? Did it meet expectations? (Did it meet promised expectations?)

What are the plans to address this, if any? (Has anyone asked this question?)

Given the history of the project (and how long it's been "right around the corner now, this time we mean it!"), why should I expect anything different now?

Which parts of this situation suggest to you the words "usable" or "useful"?


Improve your skills with Modern Perl: the free book.


Comment on Re^2: Waiting for a Product, not a Compiler
Re^3: Waiting for a Product, not a Compiler
by moritz (Cardinal) on Nov 28, 2011 at 09:47 UTC

      Both: it's not the root cause, but it's an important and obvious symptom.

      Will that be fixed any time soon?

      How about my questions to you?


      Improve your skills with Modern Perl: the free book.

        Thanks for your (relatively) straight answer.

        Will that be fixed any time soon?

        You don't seem to have noticed that, but the Rakudo Star releases are our attempt at fixing that. The "star" releases aim at providing modules, documentation and stability to the user.

        Since the current development branch ("nom") has had quite a few regressions against the old master branch, we haven't made any new star release from it so far (though 4 months aren't that much, compared to the release cycles of many other projects). So people who use the star releases enjoy much more of the stability you seek. We'll take care to make the next star release as compatible with previous star releases as we reasonably can, though of course module authors and end users still have to track spec changes.

        You seem to oppose the big rewrites that Rakudo has gone through, though so far I haven't seen you proposing any viable alternatives. I can see how annoying the breakage is that comes with such a rewrite, but we don't do them for fun; we do them because we see no other way to implement large-scale changes that need to be implemented in order to advance Rakudo. What would you do instead?

Re^3: Waiting for a Product, not a Compiler
by raiph (Hermit) on Nov 28, 2011 at 17:45 UTC
    Hey chromatic,

    It is sobering to see how the 2 year Perl 6 project has turned in to 12 years, and still counting before a 6.0.0.

    I've followed p6 and parrot, and your path alongside that, for the last 12 years. I've read several of your posts about this this year, and now this one.

    Before getting into head space, I'd appreciate you trying to be explicit about your underlying feelings about p6 and parrot, separate from justification of them. I hear alienation, disappointment, frustration, confusion, exhaustion. (I don't hear anger or contempt.) Is that about right?

    To ensure some balance in the force, I want to also note that I imagine that many in the Perl 6 community will feel uneasy and even alienated when they read some of the posts you've made this year. Maybe I'm wrong, but I'm especially sensitive to what's going on with Pm. (I don't know.) I don't care if he never writes another line of code; he will remain a p6 hero in my book forever, as will you, and I want his life to be as wonderful as it could possibly be. OK, balance restored, I hope.

    Now to head space.

    > given the history of every Perl 6 implementation so far (and Pugs and Rakudo in specific), I see a lot of interest in writing a compiler but very little evidence of desire to bundle that compiler into a product that real people can use to write and maintain real programs.

    Perhaps the evidence is lacking, especially while Pm is away, but don't you feel it? I'd be amazed if there wasn't tremendous pent up desire to do exactly as you suggest. I suspect there is delayed gratification of superhuman dimensions going on.

    > I believe the current Perl 6 implementations have, as you adroitly put it, "fundamental flaws that aren't being addressed by any future plans for ... development that I've seen so far".

    Ha! Touche. :)

    I will understand if you prefer to pass on this, but what would your #1 pick be for a fundamental p6 flaw not being addressed by a plan (other than not being production ready)?

    > Does the most recent Rakudo Star release represent a worthwhile point for someone to write a serious program in Perl 6? Does the most recent compiler release?

    As you know, one hacker's serious is another's supercilious. I'm serious about that.

    You jumped in early enough to get early adopter burn out. But, from my perspective, I would guess that most people who jumped in since Rakudo Star shipped, with clear eyes, to write code they defined as "serious", would be impressed at what's going on, and delighted by what's in the latest compiler.

    > Given the Morton's fork of using a stable but buggy release of Rakudo (for which all development has stalled) and an unreleased version with admirable improvements but serious regressions, what is your average user interested in Perl 6 but not interested in writing a compiler to choose?

    A classic dilemma, common to users of any project when it's in the midst of making a painful but ultimately smart break with the past. No easy answers, but a smart break is a smart break. In this case, the dilemma is only lasting a few months, and is clearly paying off very handsomely for the project as a whole.

    > What's Rakudo's bus number?

    I'm more confident of catching the Rakudo bus than I was when it was really just Pm driving it.

    > Was Rakudo Star a "useful release"? Was it what people hoped? Did it meet expectations? (Did it meet promised expectations?)

    I'd like to see a poll on those three questions. And a follow up one asked in about April next year of the upcoming Rakudo Star release. And then a comparison of the two polls. (I have hopes and expectations about how that comparison will go, but make no promises.)

    > Given the history of the project (and how long it's been "right around the corner now, this time we mean it!"), why should I expect anything different now?

    If you've been hearing that it's right around the corner, and this time we mean it, for a long time, you can't be expected to expect anything good.

    Otoh, if you look at what's going on (not what people say is going on), well, I'm very excited by what I see...

      (I don't hear anger or contempt.)

      Right. I'm disappointed. I think the current directions of the various implementations are misguided, but they're fairly organic and depend on what individuals individually choose to work on.

      In this case, the dilemma is only lasting a few months...

      The same as with every rewrite for Rakudo, right? How long will the dilemma last for the next rewrite? (I know that some people assert this is the last one, they really mean it this time. I go by history.)

      ... what would your #1 pick be for a fundamental p6 flaw not being addressed by a plan (other than not being production ready)?

      What's the plan to make and support a product usable for end-users who aren't themselves Perl 6 developers? ("Oh, we should probably make a Rakudo Star release again sometime!" isn't a plan.)

      Look, I understand that there's very little funding and a low bus number and that attracting volunteers to do work most people consider hairy and scary is very difficult. I really do. I know things like this take a while. (I worked on it for quite a while myself.)

      I'm just not interested in being told "It's really right around the corner, this time. It's really usable this time, we mean it." Stop telling me that and start showing me that people use it for something productive other than writing static blog pages and solving math puzzles on Rosetta Code. Show me that if I write another library, it'll stay working for more than a release or two. Show me that I can upgrade between Star releases and not pay an extraordinary upgrade tax. Show me a compelling reason why I should use Rakudo over something else (and no, "Perl 6 is inevitable, see right there in the version number!" is not a compelling reason.)


      Improve your skills with Modern Perl: the free book.

Log In?
Username:
Password:

What's my password?
Create A New User
Node Status?
node history
Node Type: note [id://940318]
help
Chatterbox?
and the web crawler heard nothing...

How do I use this? | Other CB clients
Other Users?
Others rifling through the Monastery: (6)
As of 2014-08-28 01:28 GMT
Sections?
Information?
Find Nodes?
Leftovers?
    Voting Booth?

    The best computer themed movie is:











    Results (255 votes), past polls