Beefy Boxes and Bandwidth Generously Provided by pair Networks
Don't ask to ask, just ask

Re^7: Perl 6: Managing breakages across Rakudo versions

by raiph (Chaplain)
on Jul 17, 2012 at 22:09 UTC ( #982345=note: print w/replies, xml ) Need Help??

in reply to Re^6: Perl 6: Managing breakages across Rakudo versions
in thread Perl 6: Managing breakages across Rakudo Star versions

I think what you are talking of is number of test cases failed or spec coverage affected with a particular refactoring effort.


Eg: If you break the addition operator, you are likely to break all while loops in a program, that is likely to break nearly everything. Therefore a stability policy for any such non backwards compatible release doesn't make much sense.

Imagine that Perl 6 and Rakudo have reached the point that it's very unlikely that such a basic problem would occur but might. More generally, issues are much more likely to arise than with a mature product such as Perl 5.

If you were writing Perl 6 code that runs under Rakudo, wouldn't you want to know what the Rakudo team thinks about such things? Wouldn't you want to know whether they'd consciously break the addition operator and how much notice you might get if they did, and what happens if it occurs accidentally, and so on? Wouldn't you want to know what explicit commitments they are making in this regard? Wouldn't you want to participate in discussions about this?

  • Comment on Re^7: Perl 6: Managing breakages across Rakudo versions

Replies are listed 'Best First'.
Re^8: Perl 6: Managing breakages across Rakudo versions
by jdporter (Canon) on Jul 18, 2012 at 01:59 UTC
    Wouldn't you want to know whether they'd consciously break ...

    I might like to know it, but I shouldn't be surprised and wouldn't be heartbroken to learn it.

      I think we all agree with Larry's position: no guarantees of stability:

      "Until 6.0.0, there is *no* guarantee of stability. You are all Early Adopters. If you can't figure out how to snapshot a particular version and keep it around, you probably shouldn't be Early Adopting. The whole point of Rakudo * is to give us lots of Early Adopters so we can put lots of final polish onto 6.0.0 as rapidly as possible after that. Don't start asking us to leap across the chasm before we're ready."

      ~~ Larry Wall, March 2010.

      That said, stability (contrasted with a guarantee of stability) is still a significant issue for most Early Adopters. So it makes sense to me that the Perl 6 and Rakudo teams are considering how they might best accommodate those who care about stability; and that Patrick was considering writing a document to formalize stability policy (even if that policy starts with "there are no guarantees of stability!"); and that Patrick was asking for input from users and other interested parties about these issues.

        As usual, Larry cuts to the heart of the matter, brilliantly.

        But, if I may be so bold, I believe you misinterpret him when you focus on a distinction between stability and a guarantee of stability. I believe what Larry is saying is that instability is a fact, things break, you (as a "user") don't know what or when, and as an Early Adopter you're foolish to expect any stability. And therefore, you, as the "vendor", should not be worrying about what your Early Adopters might think about this or that level of stability.

Re^8: Perl 6: Managing breakages across Rakudo versions
by Anonymous Monk on Jul 18, 2012 at 04:47 UTC

    If you are guaranteed to break stuff(which is no backwards compatibility) a stability policy has little meaning.

    I cannot imagine anyone, breaking something for just fun. I know there are valid reasons why anybody would break anything.

    If you want to discuss about a stability policy you must first work towards a backward compatible release

Log In?

What's my password?
Create A New User
Node Status?
node history
Node Type: note [id://982345]
and all is quiet...

How do I use this? | Other CB clients
Other Users?
Others lurking in the Monastery: (5)
As of 2018-06-22 04:06 GMT
Find Nodes?
    Voting Booth?
    Should cpanminus be part of the standard Perl release?

    Results (121 votes). Check out past polls.