Slash Boxes
NOTE: use Perl; is on undef hiatus. You can read content, but you can't post it. More info will be forthcoming forthcomingly.

All the Perl that's Practical to Extract and Report

use Perl Log In

Log In

[ Create a new account ]

chromatic (983)

  (email not shown publicly)

Blog Information [] Profile for chr0matic []

Journal of chromatic (983)

Wednesday July 05, 2006
07:52 PM

Perl 6 Design Notes

[ #30187 ]
The Perl 6 design team has a weekly phone call to discuss the state of the project.  Here are my notes from today's meeting.

    - mainly going through PDDs
    - some auxiliary conversations going on around those
    - one interesting one with Audrey yesterday
    - Vishal is looking into some of the proposed IMCC refactorings
    - I'm glad to have him looking into it
    - hope to bless the exceptions PDD by the end of the week
    - that's one milestone to knock off of the list

    - starting to recover from Chicago
    - whacking away on the Synopses still
    - making clarifications
    - paying attention to the trainers when they carp about things that are
      hard to teach
    - first we paid attention to the original Perl users carping
    - now we're paying attention to the implementors
    - we are starting to pay attention to the trainers
    - things that are hard to pronounce
    - maybe rename the pipe operator so as not to conflict with Unix pipes
    - talking to Damian about using signature notation to process arguments
      coming into a process
    - rather than having multiple versions of Getopt
    - the exact syntax is indeterminate now
    - what if you want to multidispatch to your main()?
    - lots of things to consider there
    - just keeping my ear open on #perl6

    - managed to get Parrot embedded and working from Perl 5 and Ruby
    - need some fixes to Parrot to make it work for the test suite
    - I may just check in the Perl 5 bindings

    - some comments on my journal entry about Ponie

    - I need to write a Perl 6 project update to clarify some of the recent

    - chatted with Audrey about the semantics of Data::Bind
    - it does aliasing and read-only aliasing with magic
    - it won't work for the corner cases
    - the problem is Perl 5's scalar flag bits
    - there's a public and a private flag in Perl 5.6
    - in 5.8, I added new meanings to the private flag for the non-magic case
    - they have different meanings whether a scalar is magic
    - effectively you need three bits worth of information
    - I think the module will have corner cases with regard to large integers
      and blessing

    - will there need to be a core change for this to work?

    - yes, but it's a risk
    - some companies asked about speeding up Perl 5
    - not sure what the strategy is there
    - a lot of the problems are unbound
    - can plug in the Perl 5.10 regex engine
    - might dual-life that for Perl 5.8.1 and later soon

    - I've suggested that Rafael move to monthly 5.9.x releases
    - some Pugs folks have offered to help with delegatable tasks
    - oh, last night the Pugs folks managed to use to translate the Perl
      6 to somewhat idiomatic Perl 5

    - have we found someone to do Perl 6 summaries?

    - Ann Barcomb popped up to volunteer
    - Audrey will help her with that for a couple of weeks
    - we shall see summaries resume in the next little while
    - it should include things discussed on IRC

    - will there be pressure to fork a 5.11 off of a 5.9 before a 5.10 comes

    - I thought the same thing at times
    - you don't want to destabilize your 5.10

    - most of what the Pugs 5 people have said is that they're using loopholes
      as if they were APIs
    - they want canonicalization of those
    - they're optimizing for fun and not production readiness, so they're fine
      with using loopholes for now
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
More | Login | Reply
Loading... please wait.
  • Should be promoted to the front page.
  • Perhaps this should be in a wiki instead of a journal? I'd like to see an URL for Nicholas' comment about Ponie in his journal. I can't find the comments to which he refers...
  • I find it remarkable that in this day and age, in a group of talented programmers that use a programming language very well suited for building a text format converter, that people not only write their notes in plain text using a fixed-width font and hard wraps, but also publish it that way. With as a result, that some lines wrap badly, due to incompatible line lengths. (And it looks terrible. But that may be just my sense of esthetics.)

    The formatting you used reminds me a bit of YAML, but as the latter cla
    • What gives is that the more work it is for me to publish the notes, the less likely I'll do it.

      I'm sure it would be trivial to build a converter for this format, but I preferred to spend my time actually putting the notes online.

      I'm sure a really talented amenuensis could type the notes with lovely formatting in another format, but I just don't type that fast. This is a compromise between accuracy, completeness, and aesthetics. (Actually, I don't care about aesthetics beyond readability.)