Perl 6 Design Minutes for 18 October 2006

brian_d_foy on 2006-10-21T18:54:00

The Perl 6 design team met by phone on 18 October 2006. The attendees were Larry, Damian, Allison, Patrick, Jesse, Nicholas, and chromatic. These are the minutes.

Allison:

• threads
• my two-hour review of the threads turned into something longer
• there are only two hours left

Larry:

• and we'll have fusion in twenty years

Allison:

• I'm just about to put that up
• it's not an apocalypse on threading
• it's just a collection of thoughts so far

Jesse:

• background reading, not spec

Allison:

• some of the background reading includes partially finished specs
• after people talk about it a little bit, I'll chunk it into something closer to an actual spec
• either objects or events next
• events are close to threads
• it might make sense to take a break from that while people muddle over threads
• I also know Patrick is working on something that needs review
• probably a small addition to the existing subroutine flags

Larry:

• things proceed apace
• helped usher the new Pugs release out the door
• hacking away on my Perl 6 program for work
• having a lot of fun with that
• starting to think about my talk in Brazil
• also holding forth on Perl Monks

Jesse:

• now that Sage is back in school, what's going on with 5-to-6?

Larry:

• as far as I know, nothing

c:

• fixed up as much of Parrot::Embed as possible
• now trying to get it to build everywhere possible
• should be able to call Parrot subs from Perl 5 as if they were normal Perl 5 subs in a few days
• that has an interesting technical problem with regard to typed signatures, but I hope it reveals some good solution

Patrick:

• working on PGE namespaces
• hope to have that today or tomorrow
• requires lots of internal changes
• I previously used a flat namespace
• looks like Kevin Tew implemented the :init flag per my message yesterday
• still reviewing the namespace PDDs
• happy to invite people to contribute tests and features
• they'd be immediately useful to the tools I'm writing
• will pass off some of the PGE converstion to smash from IRC who wanted something to do
• can't make it to the Hackathon phyiscally, but I can be on IRC for most of the weekend

Damian:

• continuing to work on the POD spec and implementation
• that's going okay
• still would like feedback on the proposed extention to =use
• if that's okay, I'm almost ready for another release of the spec and parser
• otherwise preparing for more work in Australia in the next week or two
• have had some feedback on the lists, but it's very quiet
• either we're being Warnocked, or we're being Warnocked

Jesse:

• talked to Chip briefly yesterday
• also turned the original milestone document into a matrix
• added in additional PDDs and things like testing
• poking some non-designers for their opinions
• just trying to see if external opinions line up with internal opinions
• should have more information by the next meeting

c:

• my only comment on the POD spec was wondering if the single-colon demarcation of the language might be confusing when used with the double-colon namespace separator

Damian:

• that's a language issue in general
• Larry needs to rule on that

c:

• that's fine; it just came to mind when I saw it in the =use spec
• would perldoc6 let you do ruby:Some::Module?

Damian:

The answer is: Yes indeedly-doo, neighbour!

If the Ruby module happened to be in your library path (it's not clear to me whether it would have to be in $PERL6LIB or $RUBYLIB; I assume the latter) then it would be required, presumably loading the necessary class to add some block functionality to the Perldoc parser. But that magic would be entirely handled by the default cleverness of the Perl 6 require, presuming the eventual parser is written in Perl 6.