% allison has joined #parrotsketch
<obra> Parrot sketch in 30 minutes
% smash has joined #parrotsketch
% pmichaud has joined #parrotsketch
* pmichaud will be late to parrotsketch (10-15 mins)
<obra> ok
% jonathan has joined #parrotsketch
% chromatic has joined #parrotsketch
<chip> yo d00ds
<obra> hi chip
<chromatic> It's 1800.
<jonathan> hi
<obra> Hi other folks
<particle> heya
<allison> hey
<leo_> hi all
<obra> parick said he'll be running a bit late
<obra> but let's go in order of announcement. so: chip, chromatic, jonathan, particle, allison, leo
<chip> heh
<chip> I'm working on the more-than-one-Integer problem.
<chip> parrot;Integer vs. hll;Integer
<chip> and I'm really happy to see that :vtable() has sprung to life
<chip> looking forward to hackathon, and I'll be at LISA06 for anybody in the DC area
<chip> ^Z
<chromatic> No code progress this week.
<chromatic> I do have some thoughts on concurrency I'll post.
<chromatic> I've also been thinking about the object model.
<chromatic> Finally, I plan to sketch out the types of transformations Pheme uses in the hope that a nice syntax will appear.
<chromatic> EOT
<jonathan> I had some Parrot time this week.
<jonathan> * Checked in an initial REQUIREMENTS section for the Objects PDD and requested feedback from HLL folk. So far only got a response from tewk++ though.
<jonathan> * Discovered a segfault in Key.pmc and fixed it.
<jonathan> * Implemented :vtable...well, kinda. * The holes in my Parrot internals knowledge were exposed...bugs++ :-( * Fixed some of 'em * Some still to fix.
<jonathan> * Removed the Flags thing from string_append as requested in a leo ticket.
<jonathan> * Discovered issue with properties not being serialized stopping attaching :vtable stuff as a prop on the ns
<jonathan> * Will try and resolve that this week and implement :vtable as a prop.
<jonathan> EOR
<particle> ~ i spent my parrot time on wednesday performing coding standard cleanups. parrot source is *way* out of conformance to the standards, so it's a long row to hoe. any help, from anyone reading this, is *greatly* appreciated.
<particle> ~ friday i rewarded myself by working some on the smartlinks code. didn't make much progress on features, but i did fix some lingering bugs.
<particle> ~ i'd like *very much* to convert our test suite from perl5 to pir. but until the proper i/o ops are available, i can't test expected vs actual for stdout from a registered pir compiler. if this can be available by chicago, my mission will be to lead a small army of newbies to convert many, many tests.
<particle> .end
* obra queues allison 
<allison> On a small note of personal victory, I have read, responded to, archived or deleted every message on parrot-porters this week. It's been a long time since I managed that. (My other email is suffering as a result, but I'm working on technological solutions to that.)
<obra> (and cues. I suppose you're already queued)
<allison> At first I thought it would also keep me from finishing off the threads PDD, but I've found being actively involved in the list actually boosts my motivation.
<allison> It's also very satisfying to say "X should work like Y" and have it suddenly work days later. 
<allison> jonathan++
<allison> Next for me is following threads discussion, and filling in a couple more points to I/O PDD.
<allison> EOR (with question for later)
<leo_> * a few :vtable patches and simplification
<leo_> * nothing else except for the usual mail/irc contribs due to ENOPARROTTIME
<leo_> ^D
% rafl has left rafl!~rafl@armitage.dreker.org
<obra> who else is here?
<obra> ok. questions until patrick is here
<obra> chip: you're up?
<chip> ok
<chip> particle: do you have a ticket# for the I/O ops that are holding up test harness?
<particle> hrmm, no i don't
<chip> (I know there are I/O tickets but I mean for this particular application)
<chip> if you could find 'em and/or make 'em I'd appreciate it
<particle> rokay
<chip> kthnx
<obra> next?
<chip> also, Allison in privmsg just suggested that the long-overdue release should be before hackathon, so I'll be working on that
<chip> [done]
<particle> allison++ # her own motivation and chip's
<obra>  indeed
<obra> allison++
<chromatic> That means we need to figure out the status of the failing tests.
<obra> next?
<particle> c: indeed
<obra> er. sorry. chromatic: go ahead
<allison> It seems like the failing tests have been piling up over the last couple of months. I now have over 100, including some that fail because I don't have Moose.pm installed (it has some nice features, but it's not a requirement for building Parrot, and should smartly skip when not available).
<chromatic> Just that we need to triage those.
<particle> whoops, i'll fix the Moose problem
<chromatic> A few are platform-specific, but a lot look like unimplemented features.
<chip> does triage involve marking TODOs where appropriate?
<chromatic> If they have bug numbers, yes.
<chip> interesting.  ok
<obra> ok. next is chromatic, i think
<chromatic> Does Parrot::Embed build anywhere but my machine?
% coke has joined #parrotsketch
<chip> does, or can?  /me ducks
<chromatic> In theory it can, but I haven't heard yay or nay since my last attempt.
<obra> (after chromatic's answer, coke goes next)
<particle> c: not on win32. i'll co a fresh parrot and get you a ticket with failure results
<chromatic> Thanks!
<chromatic> Unix user reports welcome too.
<obra> ok.
<obra> coke: you're up
<obra> then me, again
* pmichaud returns (reviewing log)
<obra> Or i can interject: So far, only jerry has given me his take on what's complete. http://fsck.com/~jesse/parrot.xls is the canonical file (for now)  Please, please,please tell me which bits you think are done.
<coke> I haven't had much parrot time past few weeks. managed to get it compiled and tested on my new intel imac. hopefully be able to get back to the build process improvements I was working on last month. Still have to do the .xls work that obra mentions. 
<coke> ERO
<coke> or EOR. whoops. =-)
<obra> pmichaud: you're up
<pmichaud> hello, all
<pmichaud> I put together a text-version draft of the .xls that obra provided last week, it's in the repository and can be seen at http://svn.perl.org/parrot/trunk/docs/dev/status_matrix.txt
<pmichaud> this is because some felt a svn-managed version of the document might be easier to work with (personally I'll go with whatever the rest of the group decides)
<pmichaud> I also worked quite a bit with :vtable and trying to figure out why PGE wouldn't work with it; I posted my results this morning to p2p and an rt ticket #40626
<chip> svn++
<pmichaud> it took a fair amount of time to track down, as it has to do with classes inheriting from parrot core classes
<obra> (the text version in svn is going to be hard to work with when I'm tracking actual initials in each cell
<pmichaud> in looking at the parrot.xls I notice that there are perl6 milestones there, but as written they seem "big" relative to the parrot ones -- i.e., they're not chunks that can be easily completed
<pmichaud> so, I'm going to come up with some intermediate milestones for that document that we can track -- smaller, bite-size chunks
<pmichaud> I'm still working on past, that's my task for the week (unless other perl6-related items come up)
<pmichaud> eor
<particle> allison: moose error is now fixed (r15068)
<pmichaud> obra: the text version tracks initials as well, unless you mean something other than what I did
<obra> pmichaud: ...fitting all the initials in the squares
<allison> particle++
<particle> i suggest new rows be appended for each set of initials
<particle> ...with no row title
<obra> how about this:
<obra> I'll manage the document and generate a text version on update
<obra> But. somebody will have to give me a parrot committ bit
<obra> commit, even
<pmichaud> I don'tknow that the text version has to live in the repository -- if it's not the master document then it perhaps *shouldn't* live there
<chip> obra: what's your rt name? "obra"?
<obra> chip: jesse
<chip> k
<pmichaud> a periodic post to p2p would be useful, in lieu of committing to svn
<particle> paul cochrane needs his commit bit as well
<pmichaud> (I'm just tossing out ideas here, I don't have a real preference.)
<allison> particle: I put in a request to robert for paul
<obra> I'm also happy to revisit it after I get an initial version with at least 5-6 of you submitting comments
<particle> great, thanks allison
<allison> obra: I'm really not sure which of those are "done"
<pmichaud> is Parrot::Embed in the matrix somewhere? 
<pmichaud> that might help chromatic :-)
<allison> obra: I'm also not sure what the various levels mean
<obra> I'd be thrilled to take suggestions for what should be tracked, though I know that it got discussed last week
<allison> obra: there's "Subsystem Stubs" but doesn't seem to be any category for "implementation complete"
<obra> allison: indeed not
<obra> "oops"
<obra> I believe that happened because I cribbed from the milestones document
<obra> which had "beta release" and "final release" as the items for implementation
<obra> But yes. adding a row now
<obra> what should it read? "Complete implementation"?
<chromatic> Sounds good.
<pmichaud> obra:  did you perhaps transpose the rows/columns?  Last week it was mentioned that it might be easier to manage if the individual items were vertical and the completion stages were horizontal
<obra> (and yes, there's an "embedding" column in the listing)
<pmichaud> because there are going to be more things to milestone than milestone states
<obra> pmichaud: no. I'm still working from the master .xls
<obra> fair enough.
<pmichaud> just a suggestion
<obra> When I generate readable items, I'll gen it the other way
<pmichaud> works for me
<obra> (has anyone added data to the version in svn?)
<pmichaud> I don't believe so
<obra> ok
<obra> what else is missing from the document?
<pmichaud> (nothing has been done to the version in svn since it was checked in)
<pmichaud> (and I didn't make any content modifications in my version)
<allison> how do we track iterations on a cycle?
<chromatic> Continuations and coroutines?
<allison> like, the fact that we often go back and revisit earlier implementations in light of later implementations
<obra>  allison: at the moment, I wasn't planning to go that granular on this. it's assumed that nothing is done until we ship ;)
<obra> chromatic: what PDD do those fall under?
<chromatic> I don't know, but they're fairly important for various of the other categories.
<obra> are they supposed to have a PDD?
<obra> note that this really is intended to help have a view of what's done, not to supplant RT for "what needs doing"
<pmichaud> .oO(would we have a critical feature without a pdd spec?)
<chromatic> Maybe, maybe not.
<allison> obra: ah, I guess I just keep hoping I'll get an "at-a-glance snapshot" that will tell me what state every subsystem is in, how far along it is, and what it's blocking on
<allison> obra: but I'm not sure there is any way to really represent that much information simply
<pmichaud> perhaps a "blocking on" column or row?
<obra> allison: I was going to start with the at-a-glance-snapshot of what state each subsystem is in.
<chromatic> It would be nice to have some sort of official ruling about what PMC methods can do when they're C and when they're HLL, or what kind of boundaries exist, or if they're stack-consuming.
<obra> Right now, I just want to get the data entry done
<obra> Then we can bikeshed about formats for a few weeks
<allison> obra: oooh, can we skip that part? :)
<obra> allison: Only if I try out my new iron glove
<pmichaud> "I'm happy to skip the bikeshedding phase as long as the bikeshed is purple."
<obra> fair enough. I just care about the gun turrets.
<obra> so. for next week, your homework is to tell me what you think is done on that spreadsheet.
<obra> anybody got antyhing else for now?
<allison> chromatic: that's a good point to add, but I'm not sure which PDD to put it in
<particle> as long as they're 50mm i don't care what color the gun turrets are
<coke> ... this parrot is not going metric.
<allison> chromatic: how about a misc_questions file in pdds/clip?
<chromatic> Why is the sky blue?
<chromatic> Why can't I recurse infinitely?
<pmichaud> the sky is blue due to the oxygen in the atmosphere
<particle> you can't? i can!
<obra> the sky is blue because I ran out of black crayons
* obra prepares to bang the gavel and close the meeting
<allison> chromatic: call it vir_verskeidenheid
<chromatic> That's my luggage combination!
* pmichaud eagerly awaits a gavel bang
% obra is now known as gavel
* gavel bang
% gavel is now known as obra
<obra> catch y'all next week
<pmichaud> next week!
<particle> in chicago?
<pmichaud> I thought the hackathon was the following weekend
<mdiep> there's one more week before chicago
<allison> thanks, jesse
<jonathan> I'll be at Chicago too. :-)
<obra> thanks all
<particle> oh, right. mental calendar is all screwy
<pmichaud> I'll be virtually in chicago
<pmichaud> it depends on how many people have their irc windows open and don't have my nick masked :-)
* jonathan is the UK invader of a US hackathon. :-)
<jonathan> cya @all
% pmichaud has left #parrotsketch
<particle> btw looks like mdiep found the answer i need, pushing an io layer on stdout
<chromatic> Funny, I thought I implemented that.
<particle> yep, you did. but i missed it.
<particle> mdiep pointed me to your impl
<mdiep> chromatic: you may have implemented it, but you didn't document it!
<chromatic> I did test it.
% smash has left smash!~smash@gil.di.uminho.pt
% chromatic has left #parrotsketch
% allison has left allison!~chatzilla@ppp-71-139-5-15.dsl.snfc21.pacbell.net
% allison has joined #parrotsketch
% allison has left allison!~chatzilla@ppp-71-139-5-15.dsl.snfc21.pacbell.net
% rafl has joined #parrotsketch
% coke has left #parrotsketch
% particle has left particle!~particle@144.81.84.178
% particle has joined #parrotsketch
% particle has left particle!~particle@144.81.84.178
% particle has joined #parrotsketch