Parrot 3.3.0 released | parrot.org | Log: irclog.perlgeek.de/parrot/today” | Accepted GSoC Students announced! | GSoC student information emails coming out soon
Set by moderator on 25 April 2011.
00:01 mtk left 00:08 pranq left 00:17 benabik left
davidfetter :) 00:20
well, it pretty much was what you said.
stop school. continue to get paid. do GSoC :)
00:22 davidfetter left 00:26 allison_ left
whiteknight blah, I was hoping dukeleto would come back 00:26
dalek rrot/tt1931-nci-parameters-deprecation: d7bc536 | plobsing++ | / (2 files):
eliminate PASM file with invalid syntax that has been superceded by equivalent PIR file
 00:28
rrot/tt1931-nci-parameters-deprecation: 20526e8 | plobsing++ | runtime/parrot/library/postgres.pir:
update PostgreSQL bindings to new NCI API
rrot/tt1931-nci-parameters-deprecation: b767bf7 | plobsing++ | src/nci/signatures.c:
fix minor thinko
rrot/tt1931-nci-parameters-deprecation: d70f47e | plobsing++ | src/nci/libffi.c:
change back to UnManagedStruct ptr wrapper in NCI (one change at a time)
cotto_work It's nice to have a 6guts post to look forward to when I get home. jnthn++ 00:29
whiteknight I was hoping it would be more about 6model 00:30
00:30 Caelum left 00:32 petdance joined
whiteknight I'm jonesing for some 6model 00:33
nopaste "kid51" at 192.168.1.3 pasted "Did this codingstd failure get fixed yet?" (22 lines) at nopaste.snit.ch/41814 00:43
cotto_work decommutes
dukeleto ~~ 00:44
whiteknight dukeleto!
dukeleto whiteknight: wazzup? 00:45
whiteknight how'd the thing with the people go?
kid51 apparently not fixed yet
whiteknight at the place, with the stuff?
kid51: oh damnit. That was me
one secont
second
my bad 00:46
nopaste "kid51" at 192.168.1.3 pasted "More c_arg_assert failures" (29 lines) at nopaste.snit.ch/41815
dukeleto whiteknight: very well
whiteknight: i am still backlogging, what needs doing? 00:47
whiteknight dukeleto: I wanted to send out that gsoc info email, with your blessing
I have it saved, can send it at any time
dalek rrot: e82f5f4 | Whiteknight++ | src/packfile/api.c:
fix codestd failure. kid51++ for pointing it out to me
00:49 jsut joined
dukeleto whiteknight: ok, i have some small edits 00:50
whiteknight see? this is why I waited
00:50 dmalcolm left
dalek rrot/tt1931-nci-parameters-deprecation: 4a2782a | plobsing++ | src/nci/ (2 files):
[codiingstd] cpp comments
 00:52
rrot/tt1931-nci-parameters-deprecation: ed17270 | plobsing++ | src/nci_test.c:
[codingstd] c function docs
rrot/tt1931-nci-parameters-deprecation: 0d8c40c | plobsing++ | src/datatypes.c:
[codingstd] line length
rrot/tt1931-nci-parameters-deprecation: 99ff475 | plobsing++ | / (3 files):
[codingstd] trailing space
00:54 jsut_ left
whiteknight my son dropped my laptop on the floor earlier today. Now my screen flickers and makes a clicking noise 00:55
kid51 And was that the laptop whose hinges you recently got fixed? 00:56
00:58 TiMBuS joined
dukeleto whiteknight: i am finishing my edits 00:58
whiteknight w00t 01:00
kid51: one and the same
kid51: my luck with this laptop has been low
dukeleto whiteknight: very nicely written. 01:03
whiteknight: i just changed a few details
whiteknight: my biggest change is to mandate that all blog posts go on the parrot.org blog 01:04
whiteknight you want to mandate that? I think any blog that we can aggregate on planet parrot should be acceptable
dukeleto whiteknight: in previous years, many students spun their wheels for weeks messing with their blogs, trying to figure out how to get a feed for their parrot tag, or installing a new blog for it, and other ridiculous things
whiteknight: one student actually starting coding a new blog engine for his parrot blog
whiteknight: it has generally been a nightmare 01:05
whiteknight okay. I suppose that's fine
dukeleto whiteknight: students change their blog URLs mid-summer and then their feeds get lost
whiteknight okay, what other big changes?
dukeleto whiteknight: it is a bit draconian, but I think it is the way of simplicity
whiteknight or small changes
dukeleto whiteknight: i made some wording stronger 01:07
whiteknight: as in "you will fail if you don't blog once a week". You were being too nice :) 01:08
whiteknight "you best be blogging fool, else you be trippin`"
dukeleto: I'm trying to allow for vacations and stuff
there are valid reasons to disappear for a week
dukeleto whiteknight: sure. They *must* tell us beforehand about that stuff.
whiteknight: The rule is "blog post every week or you fail" and exceptions must go through us.
whiteknight okay, fair 01:09
lesson learned. I'll never be nice again
01:09 ShaneC left
bubaflub i imagine the blogging requirement starts at the official GSoC start date, right? 01:09
whiteknight bubaflub: it started last week. You're in a hell of a lot of trouble 01:10
dukeleto whiteknight: can you add something about one blog post needed for the bonding period?
bubaflub: good quesion
whiteknight dukeleto: Sure
dukeleto bubaflub: you owe me 500 lines of code, 5 blog posts and 20 pushups, yesterday ;) 01:11
bubaflub dukeleto: doh 01:12
dukeleto: can i tag posts from my existing blog and have them fed to parrot.org blog? or should i blog directly on parrot? 01:13
cotto ~~ 01:16
whiteknight bubaflub: blog directly on parrot 01:17
set up a blog on parrot.org
kid51 whiteknight: I'm still getting c_arg_assert.t failures as nopaste.snit.ch/41815 (except for the last) 01:19
bubaflub whiteknight: ok, created a user on parrot.org
whiteknight kid51: yeah, I only fixed one
kid51: have GSoC business to deal with
kid51 I tried running headerizer.pl -- but that changed nothing. 01:20
01:20 soh_cah_toa left
petdance plobsing_: You want I should splint your branch? 01:22
plobsing_ petdance: that would be much appreciated
petdance what's the mojo to check out the branch? 01:23
oh, easy
git checkout tt1931-nci-parameters-deprecation
01:25 petdance is now known as alester
bubaflub dukeleto: also, after GSoC I am interested in doing some Postgres/GMP/DBIx::Class magic 01:26
dukeleto bubaflub: sounds good! let's concentrate on GSoC stuff for now, though :) Write down your ideas :) 01:27
bubaflub dukeleto: yeah, i think it would be cool to have a full stack for math experiments on the web 01:28
alester plobsing_: I have a big ol' compile error 01:30
oh, wait, pulled new stuff 01:31
whiteknight dukeleto: okay, sending now 01:33
alester plobsing_: You've got a consting error. Looking into it. 01:35
plobsing_ alester: is this a warning? a c++ error? 01:36
alester it's casting away const
well, it's not casting
just ignoring
01:38 soh_cah_toa joined
alester Also, there's a part in src/nci/signatures.c where you're trying to cast an int to a struct. 01:40
whiteknight if plobsing says the int is a struct, you can be damn certain it's a struct
alester oh, not a struct
an enum
whiteknight: I'm not doubting his skills. But people ARE human. :-)
01:41 sigue left
plobsing_ yeah, I do the int-enum thing far too often. if it walks like an int... 01:41
whiteknight plobsing does not produce syntax errors. The compiler is wrong
plobsing_ I wish that were true
alester Hmm, I wonder if there's a way around that 01:42
the int/enum thing is a C++ prob 01:43
dukeleto whiteknight++
plobsing_ alester. casting that is safe, but required to be explict by C++. enum has int-sized storage unless it can be guarranteed otherwise. 01:44
dalek rrot/tt1931-nci-parameters-deprecation: 156e409 | petdance++ | src/datatypes.c:
fixing constness of char*. Also consted a local
alester plobsing_: I undrestand that.
But I dunno how to get arond it short of adding -fpermissive. 01:45
soh_cah_toa whiteknight: i'm going through your gsoc email now. i thought #parrotsketch was at 20:30 utc? here it says 19:00. is it different during the summer?
whiteknight oh shoot, did I get the time wrong
I put something in and promised myself I would go look up the real numbers
but it's not my fault, dukeleto proof-read it! blame him
soh_cah_toa yup
hey, don't you be dissing my mentor yo 01:46
oh wait cotto my mentor
whiteknight cotto is your mentor
:)
I think we all have poor attention to detail tonight
soh_cah_toa i forgot b/c dukeleto is mentioned in the gsoc acceptance email
so just to make things clear, #parrotsketch is definitely at 20:30 utc? 01:47
whiteknight yes, 20:30
soh_cah_toa alright, i figured that was just a mistake
whiteknight normally I can tell time at a 9th grade level. Tonight was a fluke 01:48
soh_cah_toa i can't throw stones though. i still struggle when reading analog clocks. always have... :/ 01:49
whiteknight Okay, I've made enough of a fool of myself for one night. I'm going to bed 01:50
soh_cah_toa ha see you later
whiteknight later
01:50 whiteknight left
alester a-splinting we will go... 01:51
soh_cah_toa alester: you, mr. lester, have a ton of cpan modules! 01:53
01:53 woosley joined
alester soh_cah_toa: I believe they have no mass or weight. 01:53
soh_cah_toa alester: well, you made tap, right? or atleast contributed 01:54
alester Named. :=-)
I didn't even name it.
I just said "Hey, let's name it."
soh_cah_toa haha, yeah i read the conversation
alester Becuase otherwise people can't use it as easily.
dalek rrot/tt1931-nci-parameters-deprecation: 73fc79e | petdance++ | src/nci/signatures.c:
const an immutable local
 01:55
alester plobsing_: That cast thing is going to be a bummer when you merge to anyone with a C++ compiler. 01:56
Is there a standard way around that? 01:57
Other than saying "Compiler, do not care?"
and really, -fpermissive just downgrades the error to warning.
plobsing_ alester: which switch thing? there's a lot of switches
recompiling w/ C++ 01:58
alester adding -fpermissive that I mentioned above
nopaste "kid51" at 192.168.1.3 pasted "tt1931-nci-parameters-deprecation branch: build failure with all g++ build" (14 lines) at nopaste.snit.ch/41816 02:05
dalek rrot/tt1931-nci-parameters-deprecation: 482b030 | plobsing++ | src/nci/ (2 files):
placate C++ and its silly attempt at type-safety
 02:09
plobsing_ C++ build should be fixed now
02:12 sigue joined
dukeleto blarg 02:18
cotto dukeleto, I agree with your blarg 02:20
alester yay 02:21
kid51 plobsing_ the all g++ build now completes: smolder.parrot.org/app/projects/rep...ails/15523 02:27
dalek rrot: ada2fff | petdance++ | / (2 files):
Update annotations for string functions that return STRINGNULL. They are now all PARROT_CANNOT_RETURN_NULL.
 02:33
rrot: aee7cc1 | petdance++ | src/string/encoding/latin1.c:
Don't treat strlen like a boolean
 02:35
dukeleto cotto: blarg^2
msg whiteknight did you send an email only to the students and not mentors? 02:36
aloha OK. I'll deliver the message.
02:42 sigue left, sigue joined
dukeleto any GSoC students lurking? 02:45
02:46 ShaneC1 joined
tcurtis is. 02:46
tcurtis can't really be said to be lurking now that he said, that, though. 02:47
nopaste "kid51" at 192.168.1.3 pasted "tt1931-nci-parameters-deprecation branch: darwin/ppc: build failure with all g++ build" (7 lines) at nopaste.snit.ch/41817
dukeleto tcurtis: aha!
kid51 must sleep
02:47 kid51 left
tcurtis dukeleto: why do you ask? 02:48
dukeleto tcurtis: just want to see if anybody had questions or whatever 02:50
02:50 bluescreen left
bubaflub dukeleto: i've created an account on parrot.org but can't seem to find the link to create a blog 02:51
dukeleto bubaflub: do you have a link to write a blog post?
bubaflub dukeleto: que?
02:52 ShaneC1 left, jrt4 joined
dukeleto bubaflub: i don't use the interface too much. It has the concept of just writing a blog post on the default blog, and then it has the concept of "personal blogs" 02:52
cotto: you know about the parrot.org blogs?
bubaflub dukeleto: ah, i thought i was suppose to setup a separate blog on parrot.org for GSoC stuff 02:53
dukeleto bubaflub: yes, that is the plan. But we might need for some adminy person to flip a bit somewhere 02:54
bubaflub dukeleto: ok 02:55
dukeleto bubaflub: write your blog posts in your favorite $EDITOR and we will figure out the parrot.org blog junk
plobsing_ msg whiteknight: (re: Parrot-Instrument) SFAICT, the problem is a corrupt/bogus context. the code segment (and the constant segment to which it refers) are fully populated with consistent data. nopaste.snit.ch/41818 avoids using the context for constant lookup. it hits a snag further along with register lookups (which go through the context again). I don't see how this relates to packfiles at all.
aloha OK. I'll deliver the message.
dukeleto bubaflub: i keep drafts of blog posts and other writing in a git repo, of course :)
bubaflub: it has saved me from losing work from web browsers crashing and other horrific things 02:56
02:56 jrt4__ left
dukeleto bubaflub: github.com/leto/writing/ 02:57
bubaflub: you might want to print out the Bene Gesserit Test-Driven Litany and frame it ;)
dalek rrot/tt1931-nci-parameters-deprecation: cc8b461 | petdance++ | / (2 files):
Update annotations for string functions that return STRINGNULL. They are now all PARROT_CANNOT_RETURN_NULL.
 02:58
02:59 jrt4__ joined
cotto dukeleto, what about them? 03:03
03:03 jrt4 left 03:12 bubaflub left 03:15 sigue left 03:36 soh_cah_toa left
dukeleto cotto: bubaflub wants to get set up with his own blog on parrot.org 03:37
cotto dukeleto, I'm sure we can accommodate that 03:38
dukeleto cotto: does some adminy person need to flip a bit to add a personal blog to parrot.org ? 03:45
cotto dukeleto, not sure. If the user can go to "create content" -> "blog post", no 03:46
03:48 hudnix left
dalek rrot: 2213372 | (Gerd Pokorra)++ | docs/parrothist.pod:
add empty line
 03:49
rrot/tt1931-nci-parameters-deprecation: 8ab5168 | plobsing++ | / (30 files):
Merge branch 'master' into tt1931-nci-parameters-deprecation
 03:52
rrot/tt1931-nci-parameters-deprecation: 8cc38ff | plobsing++ | / (2 files):
Merge branch 'tt1931-nci-parameters-deprecation' of github.com:parrot/parrot into tt1931-nci-parameters-deprecation
alester It would sure be nice if parrot.org and rakudo.org could run together 03:55
since they're both Drupal
PerlJam so that they could suffer the same problems? ;) 03:56
cotto rakudo.org doesn't look too good atm 03:58
PerlJam indeed
alester argh 03:59
cotto well, the page saying that Drupal can't connect to its db looks great, but it's not what I was hoping for
03:59 sigue joined
alester Maybe the perlbuzz traffic knocked it over. 03:59
19,318 hits yesterday 04:00
04:01 benabik joined 04:02 alin left
PerlJam alester: btw, your "slipping away" post called your book to my attention (again). I was recently slightly upset at a couple of my student workers because they got mundane jobs after graduation rather than the outstanding jobs I know they could have gotten. I intend to encourage future student workers to read your book. 04:02
alester Heh, thank you. 04:03
That's kind of the point.
To any naysayer who says "Yeah, like you can find a job you love," I say "Yes, you can."
cotto andy++ 04:04
alester Half of it is deciding that you CAN love your job.
I bounced the rakudo.org 04:07
Who wants to run Nagios on it for me? :-)
04:29 sigue left
dalek rrot: d48f5d9 | (Gerd Pokorra)++ | ports/fedora/ (4 files):
update to packages for 3.3.0
 04:37
04:41 plobsing_ left 04:42 sigue joined
dalek rrot/jit_prototype: cd19a6f | bacek++ | compilers/opsc/src/Ops/ (2 files):
Move deep_clone into Utils
 04:45
rrot/jit_prototype: aad9e1f | bacek++ | compilers/opsc/ (5 files):
Move Preprocessor into own class
rrot/jit_prototype: f67c1bf | bacek++ | compilers/opsc/src/Ops/ (2 files):
Add ability to set CPP for Actions
rrot/jit_prototype: ef59717 | bacek++ | compilers/opsc/src/Ops/Compiler/Actions.pm:
Fix check for existence of Preprocessor.
rrot/jit_prototype: ef15bdc | bacek++ | compilers/opsc/src/Ops/ (2 files):
Add couple more macros
rrot/jit_prototype: 7e97c5d | bacek++ | compilers/opsc/src/Ops/JIT.pm:
Add handling of pirop<!>
rrot/jit_prototype: 7b874a6 | bacek++ | t/compilers/opsc/21-jit-files.t:
Use preprocessor in tests
04:46 jrt4 joined 04:50 jrt4__ left 04:55 jrt4__ joined 04:59 jrt4 left 05:07 sigue left 05:15 dukeleto left 05:39 bacek left 05:43 ShaneC joined, alester left 05:44 jrt4__ left 05:57 sigue joined 06:04 alin joined 06:32 theory left 07:05 mj41 joined, alin left 07:16 BloodyCoin left 07:26 dod joined 07:54 sigue left, sigue joined 07:59 sigue left, sigue joined 08:06 sigue left, sigue joined 08:15 JimmyZ joined 08:22 lucian joined
lucian does GSoC acceptance victory dance 08:24
congratulations everyone else too!
moritz lucian++ 08:26
08:39 alin joined 08:46 dod left 08:52 UltraDM joined 09:00 UltraDM left 09:03 dod joined, alin left, alin joined 09:11 UltraDM joined 09:12 sigue left 09:18 sigue joined 09:22 JimmyZ left, theory joined 09:36 theory left 09:47 dod left 10:00 sigue left, sigue joined 10:04 dod joined 10:06 woosley left 10:19 ambs joined 10:22 alin left 10:25 sigue left 10:26 sigue joined 10:28 alin joined 10:37 alin left, alin joined 10:45 alin left 10:47 alin joined 10:49 alin left 10:57 alin joined 11:00 woosley joined 11:03 sigue left, sigue joined 11:08 Caelum joined 11:13 alin left, Caelum left, Caelum joined 11:16 alin joined 11:46 sigue left, sigue joined
moritz who develops smolder? 11:46
11:53 sigue left, Patterner left, sigue joined 11:56 Psyche^ joined, Psyche^ is now known as Patterner 11:57 birdwindupbird joined 12:10 whiteknight joined 12:16 bluescreen joined
whiteknight good morning, #parrot 12:18
msg dukeleto I sent the email to all the students and mentors except you. I was using a list of addressed I had pre-assembled, from the time before you become a mentor. Sorry about the omission. I'll fwd to you 12:20
aloha OK. I'll deliver the message.
12:20 hudnix joined 12:23 jsut_ joined
moritz wonders if he'll receive two GSOC t-shirts this year, one for each mentoring organization (PaFo and TPF) :-) 12:25
whiteknight I never received a GCI shirt 12:26
so I'm hopeful that I get at least one GSoC shirt
moritz /o\\
coke_ . 12:27
whiteknight I was meaning to email carol smith about it, but she's busy with GSoC so I didn't want to bother her
coke_ seen rohit?
aloha Sorry, I haven't seen rohit.
12:27 jsut left
coke_ moritz: smolder is on CPAN 12:28
seen rohit_nsit08?
aloha rohit_nsit08 was last seen in #parrot 17 hours 19 mins ago saying "whiteknight: thanks, going for the party :-)".
coke_ aloha, msg rohit_nsit08 - my IRC will be spotty for a while -you can always reach me at will@coleda.com though. 12:29
aloha coke_: OK. I'll deliver the message.
12:46 bubaflub joined 12:47 woosley left 12:48 woosley joined 13:12 particle1 joined 13:15 particle left
particle1 moritz: mpeters, michael peters, at plusthree, develops smolder 13:20
13:20 particle1 is now known as particle
moritz coke_, particle: thanks 13:21
13:23 dod left 13:33 lucian_ joined, dod joined 13:35 birdwindupbird left 13:37 lucian left
whiteknight msg plobsing Oh, I didn't realize the context had changed, it was looking like the bytecode was changing somehow. Somewhere along the line somebody is corrupting the current context it seems. I'll look at it more today 13:44
aloha OK. I'll deliver the message.
13:47 particle left
mikehh whiteknight: I am sort of a testing/test mentor for GSoC, but am quite happy to be backup mentor for the GMP project 13:51
13:52 PacoLinux joined 13:55 particle joined
whiteknight mikehh: okay, I'm going to pencil you in for that. Like I said, we just want to have names in the slots. 13:55
13:58 lucian_ left
moritz whiteknight: fwiw I'm busy mentoring tadzik++ (through TPF), so I'm not available as a backup mentor for PaFo :/ 14:00
whiteknight moritz: oh, congratulations to tadzik++! Don't worry about it, I think we have plenty of people available to backup 14:02
I just sent you an email because you were on the list of mentors
14:02 lucian joined
moritz aye, I know 14:03
whiteknight did p6 get any other students? 14:05
dalek rrot: ed6f23c | mikehh++ | MANIFEST:
re-generate MANIFEST
whiteknight One thing I liked from years past was the ability to see all the students accepted for Perl 14:06
moritz mdk.per.ly/2011/04/25/gsoc-the-tpf-...-students/
no other p6 projects
at least none worth mentoring 14:07
whiteknight oh, that's something of a shame
moritz some good ideas, but with poor feedback and timelines
whiteknight yeah, we had plenty of those too 14:08
I'm surprised TPF only had 6 projects. That seems low
Is that because of a small total number of applications, or just a small number of "good" ones? 14:09
14:10 mtk joined 14:13 Andy_ joined
moritz the latter 14:14
PaFo turned down one or two slots too, right? 14:17
whiteknight We had 16 or 17 proposals total 14:21
about 7 of them were absolutely unacceptable. 1 application was good but we couldn't find a suitable mentor for 14:22
It's a shame too. Like you said some of those had really good ideas at the core of them, but so many students never took any feedback, never updated the proposals, never made them into good proposals 14:30
and we aren't going to devote our resources to a good idea with a bad plan
14:30 ambs_ joined, ambs left, ambs_ is now known as ambs 14:41 birdwindupbird joined 14:42 darbelo joined 14:43 birdwindupbird left 14:51 theory joined
benabik Morning, #parrot 14:53
whiteknight good morning, benabik 14:54
darbelo ~~ 14:56
whiteknight good morning darbelo 14:58
15:01 UltraDM left 15:03 birdwindupbird joined, hercynium joined
mikehh All tests PASS (pre/post-config, make corevm/make coretest, smoke (#15561) fulltest) at 3_3_0-51-ged6f23c 15:06
Ubuntu 11.04 beta amd64 (g++)
benabik Is there a magic incantation to create a blog on parrot.org? I appear to only be able to create "Scratch" pages. 15:18
whiteknight let me look. it might be a permissions thing 15:19
benabik: try now 15:20
benabik whiteknight++: Oooh. Now I can make four kinds of things. Including blogginess. Thank you. 15:21
I supposes that somewhat obligates me to make a post in the near future. :-D 15:22
whiteknight what other things?
benabik Blog, Page, Scratch, Story
whiteknight okay, awesome
blog is what you want
benabik I figured. :-) 15:23
whiteknight let me know when you do make a post, so I can check the links and make sure your posts are going to the aggregator
benabik whiteknight: Will do. I'll probably try for a quick one after brunch. 15:24
dalek rrot-instrument: 6336df9 | Whiteknight++ | src/dynpmc/instrument.pmc:
we goofed. Use the supervised interp to compile and run the file, not the current interp. Set up IMCC compiler PMCs in the supervised interp so we have that when we need to look for it
whiteknight msg cotto I committed a fix to parrot-instrument that gets us much further than we were getting. We were using interp instead of the supervised child interp for some operations, which was causing a problem. 15:26
aloha OK. I'll deliver the message.
whiteknight msg soh_cah_toa: I fixed parrot.org permissions so you can write blog posts. Check it out 15:29
aloha OK. I'll deliver the message.
whiteknight msg rohit_nsit08: I fixed parrot.org permissions so you can write blog posts. Check it out
aloha OK. I'll deliver the message.
whiteknight msg bubaflub: I fixed parrot.org permissions so you can write blog posts. Check it out
aloha OK. I'll deliver the message.
bubaflub whiteknight++
whiteknight Those are all the students who have created parrot.org accounts, that I can find 15:30
tcurtis already has permissions from last year 15:31
darbelo Are parrot.org accounts linked to trac accounts?
Or was that svn.
benabik darbelo: Nope
darbelo I can never remember which things are linked together.
15:32 plobsing joined
whiteknight parrot.org is unhelpfully not linked to anything 15:32
trac and svn shared credentials 15:33
now, all of them are different
and of course, I don't think any of our sites allow OpenID. 15:34
I'm actually surprised that Github doesn't allow OpenID logins, much less that it wouldn't serve as an OpenID provider 15:35
that seems like a natural fit 15:36
of course, even if they did it would be a huge uphill battle to get our drupal and trac instances to accept OpenID logins too
lucian is hugging github more closely an option? 15:39
i hear their issue tracker sucks a lot les
whiteknight it has gotten much better, yes 15:40
still not as great as I think it should be
lucian checks it out
bah, still sucks
whiteknight yeah, nowhere near what we would need if we wanted to migrate Parrot to there 15:41
lucian they had a prototype written in Objective-J i think, that was very slow but had useful things, like milestones and various states
could drupal at least be replaced with github's websites?
or w/e the feature is called 15:42
whiteknight github pages don't have the features we would need
close, but it's all not integrated enough 15:43
lucian: speaking of which, create a parrot.org account so I can give you bloggy goodness
lucian does 15:44
is using a different blog an option? 15:45
whiteknight no
15:45 dukeleto joined
dukeleto hola 15:45
whiteknight I wanted to allow it, but dukeleto had some depressing anecdotes about students creating problems in the past
dukeleto still with the blog posts?
whiteknight we just started with the blog posts
lucian asked where his blog can be, and I said parrot.org 15:46
dukeleto whiteknight: i am fine with students putting a blog post on their blog *and* the parrot.org blog
whiteknight: having them on parrot.org makes it easy to find all our gsoc blog posts and saves a lot of grey hair 15:47
whiteknight yeah, that's not an issue
benabik If parrot.org can give me a personal RSS feet, I'll probably get tumblr to read it. But I wouldn't want to use my tumblr for GSoC since it's full of irrelevant stuff. 15:48
*feed
dukeleto benabik: yep, that is possible
lucian wouldn't it be possible for parrot.org to use a feed instead? wordpress can create a feed based on tags for me, i think 15:49
whiteknight lucian: is it a problem to just create a blog on parrot.org, if only for the summer? 15:50
lucian whiteknight: no, i'll just have to post things twice
dukeleto lucian: it might, but for the sanity of the org admins, having your blog posts directly on parrot.org is easiest. You are of course free to put the same blog on any other number of blogs, blags and micro-irrelevancy devices
lucian: the same blog post, that is
lucian dukeleto: sure, it's just slightly less easy 15:51
whiteknight the problem is that if we open up the requirement, we're going to end up with several students who go off in exactly the wrong directions
students who waste hours and hours building and tweaking new blogs, changing blogs, etc
We aren't trying to fund people to set up a personal blogging infrastructure to survive beyond GSoC, we're looking for an information stream from the student to interested developers 15:52
lucian whiteknight: sure, i was thinking of people who already have one set up 15:53
whiteknight lucian: right, but what happens if your blog provider goes offline this summer? Or changes URLs?
or ....
lucian shrugs 15:54
whiteknight I know, it's not perfect
lucian wordpress reliability > parrot.org reliability i guess
whiteknight but we have to be fair. And fair means realizing that not all our students are at the same level as Lucian is
lucian whiteknight: "level"? 15:55
made an account, eponymous
whiteknight the same amount of knowledge of things blog-related
benabik When sending the CLA to legal@, do I need to scan/send more than just the pages I wrote on?
whiteknight okay, you have the ability to blog now
lucian i'm supposedly very "backwards" when it comes to "social" "media"
whiteknight: thanks 15:56
whiteknight benabik: if possible, send all pages
lucian; when I did GSoC, all the students had to use use.perl.org
15:56 alin left
whiteknight so be thankful we don't do that anymore 15:56
lucian heh
whiteknight lucian: I think you would bleed from the eye sockets if you had to go there once a week
15:57 dukeleto left
benabik Hah. I got Preview to include the unchanged pages from the original PDF. 15:57
lucian whiteknight: i think a planet.parrot.org might be useful though. probably longer term than GSoC, though 15:58
15:58 dukeleto joined
whiteknight lucian: yeah, and I think parrot.org blogs get aggregated there automatically 15:58
I may be wrong about that
lucian uh, afaict planet.p.o doesn't exist 15:59
whiteknight planet.parrotcode.org, I think
moritz planet.parrotcode.org exists
whiteknight our old website was parrotcode.org
lucian i see 16:00
not very pretty :)
whiteknight the aggregator? No
lucian i guess planet.sugarlabs.org spoilt me. it aggregates + caches, so no one was jumpy about losing posts 16:01
also, either you or melange got my name wrong
whiteknight I thought I copied it from melange
what should it be?
lucian Lucian Branescu Mihaila
or Lucian Branescu-Mihaila 16:02
whiteknight gotcha
lucian the passport office and the national id office can't agree which
dukeleto lucian: which country do you live in? just wondering
lucian i guess i prefer the second, makes it obvious that all that crap is my last name
dukeleto: that happens in Romania
right now i'm in the UK, though
dukeleto whiteknight: i think we may want to have a document that lists students, mentors and their timezones or something like that. It will come in handy. 16:03
whiteknight lucian: fixed. Sorry about the mixup
cotto whiteknight, thanks! It's surprising that there was a bug like that.
lucian whiteknight: no problem
whiteknight cotto: not surprising. I updated some of that code hastily
cotto Ah. I thought it was in some of khairul's original code. 16:04
whiteknight no. We've had too many updates to bits of the code since khairul left
lucian hints at planetplanet.org 16:05
benabik IIRC, bacek is on Sydney time... Being 14 hours different than my mentor is going to be interesting. Good thing I tend to stay up late anyway. :-D 16:06
dukeleto benabik: yes. should be very fun :) 16:07
lucian whiteknight: oh, and if my name doesn't fit that long somewhere, don't bother 16:08
whiteknight lucian: we'll make it all fit everywhere
lucian shrugs
whiteknight giving proper credit and proper attribution to the correct name is an important thing
lucian my co-nationals have great trouble with my name, my expectations are low 16:09
whiteknight tell your co-nationals that Whiteknight said to get it straight
they'll listen to me
:) 16:10
lucian in their defence, names tend to be written differently, like "Branescu Mihaila Lucian" 16:11
16:12 darbelo_ joined
whiteknight In America, we call you "That Lucian guy or whatever his name is from IRC" 16:12
lucian "Mihaila" is an extremely uncommon last name, so people tend to think it's a weirdly/misspelled version of "Mihai" or "Mihaita", which would be a first name
whiteknight one day we are going to have to skype, so you can tell me the correct way to pronounce that 16:13
lucian loo chee uhn, roughly 16:14
16:14 darbelo left
lucian people call me Luci usually, though 16:15
whiteknight ah, okay
lucian the english-style pronunciation is close, but i enjoy tormenting native english speakers a little
lucian still doesn't get how most brits barely speak a second language 16:16
whiteknight I don't speak a second language
I barely speak english
lucian but didn't you have to learn it in school at least? 16:17
whiteknight I did take latin for a few years in school
can't remember more than a scant handful of words
dukeleto lucian: the US has very weak second language requirements
cotto whiteknight, perfect. You can go to Latin America.
lucian dukeleto: well, apparently US as well
dukeleto lucian: usually you have to take 2 language classes in each of high school and college, which doesn't quite add up to fluency :) 16:18
lucian i speak few languages compared to the average Dutch, but some of what i was taught still stuck
16:18 woosley left
dukeleto kann Deutsch 16:18
16:19 mj41 left
lucian this reminds me, i have to learn spanish better by the end of this summer :) 16:19
16:19 Andy_ left
dukeleto cotto: LFNW is *this* weekend. How did that happen? 16:25
16:27 fperrad joined
dukeleto fperrad: howdy 16:28
16:28 lucian left
cotto dukeleto, I know, right? 16:29
dukeleto cotto: what dates are you planning on being there? 16:31
16:32 lucian joined
lucian now hates his laptop a little 16:33
cotto dukeleto, I'll probably take off a little early on Sunday. I'd like to get back home by 5-ish. 16:39
time for $dayjob
16:44 davidfetter joined
dukeleto cotto: what time you planning on getting there? 16:46
whiteknight lucian: why hate your laptop?
16:46 ilbot2 joined
moderator Parrot 3.3.0 released | parrot.org | Log: irclog.perlgeek.de/parrot/today” | Accepted GSoC Students announced! | GSoC student information emails coming out soon
davidfetter dukeleto, so about your PL/concerns... 16:46
whiteknight lucian: because I will trade you in a heartbeat
lucian whiteknight: well, a little. it's very, very slow, even though the hardware is pretty good 16:47
and it's heavy, and it's annoying because it's apple
whiteknight ah, all you had to say was "it's apple"
lucian it's not exactly bad, but there are much better laptops to buy at much lower prices 16:48
it's a MBP 4.1. at least newer MBPs have nicer hardware
i'll try to sell it again, hopefully for a decent price 16:49
davidfetter has a somewhat creaky mbp
dukeleto davidfetter: what now?
davidfetter dukeleto, well, there's going to be a PL summit in ottawa
dukeleto davidfetter: i made a pgxn meta.json for pl/parrot
davidfetter: yes, i've heard about it
davidfetter great!
anyhow, i heard you weren't going to make it, so i'd like to get your concerns on the table 16:50
dukeleto davidfetter: ok
davidfetter: specific concerns, or general concerns?
davidfetter both
dukeleto davidfetter: the documentation for writing/mainting/bug fixing PL's leaves much to be desired and actually mocks the reader 16:51
davidfetter: at least, the last time I read them
davidfetter go on :) 16:52
coke_ any feather admins here? 16:54
whiteknight still doesn't have a feather account, not for lack of trying 16:55
coke_ seen juerd?
aloha juerd was last seen in #perl6 3 days 5 hours ago saying "Is that the secret channel? :)".
16:57 bacek joined, birdwindupbird left
dukeleto davidfetter: hmmm 16:57
davidfetter what things do you feel you needed to build that should have just been there? 16:58
dukeleto davidfetter: trusted vs. untrusted are very confusing topics, even to PG veterans. That means something is wrong.
davidfetter never mind whether you've actually gotten around to building them
16:59 bacek left
dukeleto davidfetter: i had to read all the internal PG headers and source code to figure basic things out 16:59
davidfetter such as?
dukeleto davidfetter: but that is more of a reflection of docs
davidfetter: mostly to figure out what all the internal PG datatypes were and what macros were used to access them 17:00
davidfetter good to know
dukeleto davidfetter: data marshalling was and still is the hardest part of PL/Parrot
davidfetter: having a map that converts any PG datatype to a Parrot datatype, and vice versa
davidfetter dukeleto, do you think that's because pg lacks facilities for "playing nicely with others?" 17:01
i.e. is it missing APIs you'd want to have?
dukeleto davidfetter: the PG docs are very nice, but it seems that the docs for PL's are not of the same quality
davidfetter also good to know
dukeleto davidfetter: yes, there probably is, let me brood for a minute
davidfetter hopes dukeleto doesn't mind if he just saves this conversation off for reference 17:02
dukeleto davidfetter: sure, you can document all this and let them know
davidfetter that's the plan :) 17:03
dukeleto davidfetter: one issue is that PL/Parrot coredumps postgres a lot, through no fault of postgres. It is usually that PL/Parrot has a bug and postgres freaks out. This makes me think that postgres could have more error-checking when interacting with PL's
cotto_work ~~
dukeleto davidfetter: i core dump postgres A LOT :)
davidfetter: but parrot is probably to blame around 50% of the time. hard to tell :) 17:04
davidfetter would some kind of pg facility make it easier to tell?
dukeleto davidfetter: if there was one place I could do a security review of postgres source, it would be the boundary between PL's and the postmaster
davidfetter: not sure. possibly 17:05
davidfetter: i only every coredump individual pg backends, very rarely the postmaster, just to be clear
davidfetter right 17:06
dukeleto davidfetter: postmaster is quite robust
davidfetter have you managed do do the postmaster?
17:07 bacek joined
dukeleto davidfetter: possibly once or twice, but they were probably because I was doing something Very Bad. If I ever coredump postmaster, I will keep track of it. 17:07
davidfetter to do*
that's very helpful
dukeleto davidfetter: i am trying to think of API's and such that would help PL authors
davidfetter: i haven't been hacking on PL/Parrot enough lately to have it fresh in my mind
davidfetter will you have any time for this before mid-may? 17:08
dukeleto davidfetter: ok, here is one
davidfetter: different PL's all implement their own caching layers
davidfetter: so if PG internals had a caching solution, all PL's could use it and you wouldn't have a dozen partially implemented cache solutions
davidfetter ah, excellent!
dukeleto davidfetter: PL/Perl seems to have one of the better caching techniques
davidfetter <3 concrete proposals 17:09
dukeleto davidfetter: I mostly stole it for PL/Parrot, the easy parts at least
davidfetter convincing people they should move to a new infrastructure is no fun :P
but i'll deal with that
dukeleto davidfetter: and by caching I mean stored procedure caching, but other types of caching would be useful as well 17:10
davidfetter k
any others in particular?
dukeleto davidfetter: if you look at the struct called plperl_call_data, you can see another type of caching that PL/Perl does
davidfetter: i think if there was new optional caching infrastructure, the most-used PL's would migrate to it quickly, and the other would do it when they get around to it, if at all 17:11
davidfetter: shouldn't need much convincing of people
davidfetter: is _PG_fini supported yet? 17:12
davidfetter: last time I read the source, it was getting ready to be implemented
davidfetter: it is the handler for when things are unloaded
davidfetter There is nothing more difficult to take in hand, more perilous
to conduct, or more uncertain in its success than to take the lead
in the introduction of a new order of things, because the innovator
has for enemies all those who have done well under the old
condition, and lukewarm defenders in those who may do well under
the new. 17:13
Niccolo Machiavelli (1513)
dunno, but i'll ask
dukeleto good quote
17:13 darbelo_ left
dukeleto davidfetter: another thing. perhaps they are addressed in other parts of the docs, but a short guide about writing postgres exceptions (PG_TRY and friends) for PL and extension authors would be very useful 17:14
davidfetter this is all excellent stuff :)
dukeleto davidfetter: SPI still seems magical to me. I am sure I have not RTFM'ed hard enough, but there could probably be improvement on SPI+PL docs 17:15
davidfetter k 17:16
dukeleto davidfetter: that is it, for now. I just read through all of plparrot.c to remind myself of stuff :) 17:17
davidfetter: go forth and bring about positive change! Best of luck.
davidfetter i'll be pestering you about what you find in plparrot.c soon ;) 17:18
17:19 rohit_nsit08 joined
dukeleto rohit_nsit08: howdy 17:19
rohit_nsit08 dukeleto: hello, awesome.. just came back from a grand treat :-)
mailing the dates for meeting on skype 17:20
17:20 bluescreen left
tcurtis aloha, msg darbelo Hey. I see that you're my GSoC mentor. 17:21
aloha tcurtis: OK. I'll deliver the message.
17:21 ShaneC left
tcurtis oops 17:22
rohit_nsit08 dukeleto: do studends need to submit any report for today's design meeting ?
tcurtis aloha, msg darbelo When would be a suitable time for you for us to talk each week about my project's progress? 17:23
aloha tcurtis: OK. I'll deliver the message.
coke_ rohit_nsit08: hio. 17:24
rohit_nsit08 coke_: hello, how are u?
coke_ good. Busy with $DAYJOB this week. 17:26
rohit_nsit08 coke_: I think I should do some little more work on Project timeline details ( specially the milestones ) and pls review this post rohitnsit08.blogspot.com/2011/04/bo...piler.html I have tried to mention in full detail the steps I'll be working on. If this is fine than I'll proceed with the steps 17:27
coke_ rohit_nsit08: image in step six borked. 17:29
dukeleto rohit_nsit08: it is optional today, since GSoC just started yesterday, but feel free to, since you have been doing stuff :)
coke_ (and 10)
aloha 10
coke_ aloha, 9 and 10 and 11
aloha coke_: 30
dukeleto rohit_nsit08: awesome! you can just post a link to your blog post there, that would be sufficient
rohit_nsit08: cool diagrams! 17:30
atrodo feels the bonding
rohit_nsit08 coke_: I hv been coding in PIR from last week and the API also, My understanding of JavaScript Object System and prototypes is I believe is also fine now.
dukeleto: thanks, made them in google docs :-) 17:31
coke_ rohit_nsit08: glad to hear.
rohit_nsit08 coke_: Will the understanding of PCT help us in this project? I was following the squaak tutorial and I think it will be good if I am able to extend some winxed like object support into it. What say? 17:35
17:35 bluescreen joined 17:36 darbelo joined 17:40 bubaflub left, darbelo left
dukeleto rohit_nsit08: benabik has the best PCT tutorial, imho 17:41
benabik: where does it live again?
benabik github.com/Benabik/cish 17:42
It's not quite what I would call a tutorial, but it's (hopefully) a decent introduction. 17:43
rohit_nsit08 benabik: It's great.
I was earlier following the one included in parrot docs 17:44
17:45 mj41 joined
whiteknight rohit_nsit08: if you have any status to share, you can make a report for #parrotsketch 17:47
17:47 dodathome joined
whiteknight rohit_nsit08: if you have nothing to share, you don't need a report 17:47
rohit_nsit08 whiteknight: hmm.. I have already updated my latest status on IRC, I don't think there is something new rightnow. I should better do some more work and than put it in the next design meet 17:49
whiteknight rohit_nsit08: that's okay. We're just getting started so most students probably haven't done anything yet 17:51
benabik: we have your CLA. Thanks 17:52
17:52 ShaneC joined 17:54 bubaflub joined 17:58 soh_cah_toa joined 18:04 dmalcolm joined
benabik whiteknight: I blogged: www.parrot.org/content/hello-my-name-... 18:22
dukeleto benabik: URL truncated
benabik++ for blogging!
rohit_nsit08 benabik: unable to access it
benabik dukeleto: Actually, Drupal put a ... in the URL for some reason.
whiteknight Hi, my name is...frika frika benabik 18:23
yeah, the dots in that url are obnoxious
dalek website: benabik++ | HELLO! My name is...
website: www.parrot.org/content/hello-my-name-...
dukeleto lulz
benabik I think I can modify the URL to be less stupid.
dukeleto there is a bug where drupal makes blog posts for parrot x.y.z. have many dots in it as well, and many broken URL rendering algorithms barf on it, like twitter's web interface 18:24
benabik New URL: www.parrot.org/content/hello-my-name-is-benabik
dukeleto benabik: awesome 18:25
dukeleto can feel the GSoC bump in productivity already
benabik will be less clever with blog titles in the future, or at least avoid too much punctuation. 18:26
whiteknight yes, one blog post is more than we've had on parrot.org since last GSoC!
18:27 bluescreen left 18:30 alin joined, jsut joined 18:35 jsut_ left 18:38 bluescreen joined 18:47 bacek left 18:52 darbelo joined 18:53 Coke left, Coke joined
soh_cah_toa hey guys, could someone tell me how to start a blog on parrot.org? i've already created an account 18:54
cotto_work create content -> blog
soh_cah_toa oh okay, last night the only option was scratch
cotto_work er, blog entry
yeah. whiteknight fixed it
whiteknight I done did dat already 18:55
soh_cah_toa oh yeah, there it is
18:55 darbelo left
soh_cah_toa speaking of mistakes...i noticed that the gsoc acceptance letter had the wrong time for #parrotsketch as well 18:55
not just the one from whiteknight
cotto_work That's the secret #ps that we don't tell anyone about. 18:56
soh_cah_toa haha
the parrot illuminati gathering
cotto_work time for noms 18:57
18:57 rohit_nsit08 left
cotto_work #ps in 92 18:57
soh_cah_toa agh, #parrotskech today. that's right
benabik soh_cah_toa: You were just talking about when it was. Did you forget it was Tuesday?
soh_cah_toa my mind has been having trouble processing anything non-gsoc 18:58
18:58 rohit_nsit08 joined
soh_cah_toa anything other than "yay, gsoc!" has not entered my mind 18:59
hmmm...i'm in a bit of a pickle here. 19:00
b/c i have an hour workshop at school today right at #ps time
whiteknight soh_cah_toa: it's no big deal. You can post a report, if you have anything worthy of sharing, at any time 19:01
I've already posted my report for the day
and you don't have to attend every meeting. I probably won't attend any of them until daylight savings time
soh_cah_toa alright. i'm definitely gonna start a blog post today 19:02
once the semester ends, there shouldn't be anymore conflicts w/ parrot stuff
i decided not to take that intership so i could focus on gsoc 19:03
and no classes
19:04 lucian_ joined
soh_cah_toa dukeletto: btw, i'm available for a conference any time on the weekends and any time after 1:00pm on fridays 19:04
*dukeleto, whoops...
dukeleto: oh, and anytime monday 19:05
dukeleto is back
soh_cah_toa dukeleto: didn't realize you were away. you got that? 19:06
dukeleto soh_cah_toa: roger. 19:07
soh_cah_toa alright
dukeleto soh_cah_toa: don't worry about attending a #ps if you have school stuff, just paste a short report
dukeleto writes a report
soh_cah_toa will do 19:08
19:08 lucian left
soh_cah_toa whiteknight: did you get a chance to look at my email about those tests? 19:08
whiteknight soh_cah_toa: briefly 19:09
19:09 Coke left, Coke joined
soh_cah_toa whiteknight: okay. do you think that's enough? i wanted more but i explained why i didn't think it was possible 19:09
dukeleto soh_cah_toa: i am still planning on responding to your feedback email, just haven't gotten that deep into my email stack yet :) 19:10
whiteknight soh_cah_toa: yeah, I think that should be enough for now 19:11
if you're worried about it, leave a comment note explaining it
soh_cah_toa whiteknight: okay, then i'll submit a patch after class later tonight
dukeleto: okay, good
19:14 darbelo joined, bubaflub left
whiteknight soh_cah_toa++ # patches for the win 19:20
soh_cah_toa that reminds me, when do i get a commit bit? i submitted my cla a while ago 19:22
whiteknight soh_cah_toa: depends. They don't just come automatically 19:25
somebody will nominate you at the #ps meeting, and then there will be a vote
dukeleto soh_cah_toa: have you submitted at least 2 patches already?
soh_cah_toa hmm...let me think
whiteknight it's actually not super-urgent that you have a commit bit right now. You can work in a fork and submit pull requests very quickly if you want
And your gsoc work should be done in a fork 19:26
soh_cah_toa whiteknight: that's right
whiteknight soh_cah_toa: I'm not against you having a commit bit, I'm just saying it's not a big deal
soh_cah_toa whiteknight: right, i get it 19:27
cotto_work Yup. It's really nice that you could theoretically go the whole summer without a commit bit and not lose any productivity because of it.
git++
whiteknight yeah, that's a big benefit of git
soh_cah_toa dukeleto: after tonight will be two. soon three after i finish release manager script
lucian_ i was wondering, would a AL 2.0 / PSL dual license be ok for a HLL?
whiteknight doing my GSoC project in svn was a huge waste. I lost so much productivity keeping my branch up to date with trunk, and all that garbage
lucian_ cotto_work: i'd rather say dvcs++ 19:28
whiteknight soh_cah_toa: ah yes, the release manager script. Definitely send a copy of that to kid51. He's one of our resident perl experts and would like to see that
lucian_: a dual license should be fine. I don't know a lot about PSL
soh_cah_toa whiteknight: sure
lucian_ whiteknight: similar to MIT license 19:29
dalek rrot/tt1931-nci-parameters-deprecation: 346910c | plobsing++ | / (3 files):
provide more descriptive error messages for missing NCI thunks
19:29 lucian_ is now known as lucian
rrot/tt1931-nci-parameters-deprecation: ea81969 | plobsing++ | src/nci/extra_thunks. (2 files):
add thunk for "ipP" nci signature used in tests
rrot/tt1931-nci-parameters-deprecation: f62e3ec | plobsing++ | src/nci/ (2 files):
handle non RIA types as NCI signatures
whiteknight lucian_: The biggest draw with AL2.0 is that we could conceivably bundle your compiler in with Parrot without jumping through any hoops
lucian yes, and i also rather like AL
whiteknight lucian: do you have a link to the PSL text?
19:29 soh_cah_toa left
lucian whiteknight: docs.python.org/license.html 19:29
afaict it's not quite as loose as MIT, but doesn't have the BSD weak copyleft 19:30
whiteknight lucian: text of the license looks like it's explicitly between the PSF and the end-user. That suggests to me that the code would have to be owned by or licensed to the PSF in order to use that license 19:31
at least, to use it verbatim 19:32
lucian whiteknight: yes, i've thought of that
whiteknight: i'm not sure the PSF would want my code, though
afaict, the two licenses are compatible, though
dukeleto lucian: do you want companies to be able to use your code for a profit? 19:33
whiteknight Yes, I don't see any discrepancies between them
lucian and AL doesn't seem to impose extra limitations
dukeleto: sure
whiteknight and most releases look like they are GPL compatible too
lucian whiteknight: they are
whiteknight so you could dual license AL2.0/GPL2.0 to be safer if you want
lucian so incorporating an AL bit into a PSL project should be ok
whiteknight I suspect so, yes 19:34
lucian allison: ping
dukeleto lucian: then you could just dual license with AL and MIT or something like that.
whiteknight Again, you might want to find an expert to ask about this kind of stuff
lucian whiteknight: yes, i might
allison lucian: pong
lucian allison: do you happen to know anything more about the PSL thing?
whiteknight lucian: Keep in mind that as the author, you can grant licenses to users for your work under any terms at any time
lucian whiteknight: yes, i know. i'm not sure i'll own a copyright to everything, though
whiteknight lucian: so if you need to modify things later to be friendly to PSF, you can do that
allison the PSF was working on getting a straight PSL license approved by the OSI
cotto_work There's always the wtfpl
allison I would generally say the PSL is better for adoption of a Python implementation 19:35
whiteknight lucian: true, but for GSoC you probably will be the only contributor to the repo, at least for the duration of the summer
lucian cotto_work++
whiteknight allison: the text of the PSL I am seeing looks like it's explicitly between the PSF and the end user
allison lucian: and also desirable to donate the copyright to the PSF, if possible
whiteknight allison: and that implies that the PSF needs to be the owner, or executor of the license for that to work
allison lucian/whiteknight: yup, which is also good for adoption of a Python implementation 19:36
the PSF is generally happy to take contributions of code bases
19:36 bubaflub joined
lucian they can't actually do anything with it, though 19:36
whiteknight okay, that's what I was worried about. We should talk to them about the contribution before using a license which suggests the code has already been contributed
allison the Chairman has specifically said they'd be happy to accept contributions of implementations of Python on Parrot 19:37
whiteknight allison: ah, that's good information to have. Who is the current Chairman?
allison Steve Holden
I'm also on the board
whiteknight oh nice. Ace in the hole
allison and can raise it as a resolution in the next meeting if desirable
but, it's up to you lucian
dukeleto Steve Holden just moved to Portland, OR, as well 19:38
whiteknight I suggest that the more we can work together with PSF for this project, the better.
dukeleto whiteknight: indeed
allison lucian: as whiteknight says the copyright of the GSoC project is entirely yours, so you can keep it for now and we can arrange with the PSF later if you'd rather
lucian allison: i don't have a problem with donating it, i just don't see what they could possibly do with it
dukeleto allison: do you think Steve would like to meet for a chat about python on parrot?
allison dukeleto: very possibly, but he will be in the SFO area for most of the next few months 19:39
dukeleto lucian: the biggest reason for you to be part of a foundation (parrot or python) would be so that some patent troll doesn't try to sue you
allison lucian: well, you'd still make the releases, they'd just be the copyright holders, like we often do with Parrot Foundation
lucian: the Parrot Foundation is the other option I'd recommend, but I think the PSF is better in this case 19:40
19:40 ShaneC left
dukeleto lucian: i agree with allison++ 19:40
it would hopefully create a nice relationship between PaFo and PSF
whiteknight Lucian can easily grant licenses to both foundations too
no reason we need to make either/or exclusivity
granting licenses is cheap 19:41
allison whiteknight: aye, or really the PSF license would grant us all the rights we need
lucian right, then for now i'll go with AL 2.0 (and dual with GPL if they're not compatible)
allison whiteknight: no need for anything special
whiteknight and we're desperate to work nicely with the PSF, so we won't cause any trouble
allison AL 2.0 is GPL compatible, you don't need the dual
lucian: would you like me to post a quick note to the PSF board asking about this? 19:42
lucian allison: yeah, sounds useful
whiteknight allison: could we set up like an informal kind of meeting between some of them and some of us?
allison lucian: one thing I always wished is that Pynie was PSL from the start, it would have been so much easier
lucian: btw, what are you calling the GSoC implementation? 19:43
lucian: can I call it "the new Pynie"?
lucian allison: or Puffin
whiteknight "Lucian's Awesome Freaking Python Compiler On Parrot" (LAFPCOP)
allison lucian: sounds good
lucian allison: Python includes a bunch of things with lots of licenses 19:44
i don't see how AL would be any harder to integrate than those
dukeleto lucian: Puffin has a nice ring to it
whiteknight lucian: the python distribution is probably treated as an aggregate 19:45
lucian: that's only necessary if you are shipping a bundle with independent components
lucian dukeleto: google it, it's the cutest thing on earth
allison lucian: yes, that's why you want the "Python Software Foundation License" (just the top) and not the "Python License" (the whole ugly stack)
lucian allison: so adding a AL 2.0 paragraph would not be desirable? 19:46
dukeleto lucian: i would say that you should worry 99% about code now, and 1% about licenses
allison lucian: is Puffin Python 3 or Python 2
dukeleto lucian: it will "come out in the wash" as they say
allison lucian: I'd skip AL 2.0 and do just PSFL 19:47
lucian: remember, the Artistic License screams "Perl"
lucian allison: i was thinking py3, it's easier
allison lucian: which isn't exactly Python-friendly
lucian allison: meh, the license is quite nice :)
allison lucian: cool, agreed py3 is simpler (just wanted to double check before posting)
lucian: why, thank you! (I wrote most of it)
lucian: took up two years of my life, that license did 19:48
lucian dukeleto: yeah, i know. i'm just so busy with my dissertation now, my background cycles don't get scheduled for programming
allison: wow. i did read you wrote it
here comes praise: of all the 'loose' licenses, the AL 2.0 is my favourite 19:49
'loose' meaning MIT/BSD-like
whiteknight does python 3 use a different license from python 2.7?
lucian whiteknight: nope 19:50
whiteknight ok
lucian allison: i'm totally there with python devs getting creeped out by perl, but hating the license is going too far 19:52
whiteknight if the artistic license screams perl and that causes a problem for HLLs, Parrot is in deep trouble
allison lucian: I wouldn't say they hate the license, it's more of a subconcious thing
lucian: They're trying out a Python implementation and the PSF license sends subtle "this is home" signals 19:53
whiteknight: I've seriously considered proposing BSD license for Parrot, for this reason
lucian allison: perhaps. but staying away from a good license really is going too far 19:54
allison whiteknight: but, it's not as relevant for core Parrot as it is for language implementations
lucian i'll be the first to admit that the path between running a process, running a regex on its output and using it on another is shorter in perl than in python
allison lucian: it's like wearing native garb when traveling
lucian allison: ah, but it always looks odd on stragers 19:55
allison lucian: sure, but at least you tried to be local :)
lucian: though, I'll never understand Americans who try to be "local" by wearing Hawaiian shirts in Africa ;) 19:56
lucian allison: heh. i've noticed i'm disconnected with american culture lately anyway, i stopped trying :)
allison lucian: hah, me too! :) 19:57
benabik We have culture? ;-)
allison lucian: maybe there's a whole culture of Americans who don't get American culture
lucian allison: i'd be a real oddball there, not even being American in the first place :)
allison lucian: I find I have more in common with open source people around the world than I do with anyone in any specific country 19:58
lucian i simply don't get it when they say things in new American movies. "Ahhhh! you say me naked!" or something. what?
allison lucian: more than any specific culture in any specific country
lucian allison++ 19:59
allison lucian: sounds baffling (American movies)
lucian nods
i'm running out of classics to watch, though :) 20:00
well, there's House. not too bad
tadzik hello parrots
allison lucian: aye, but he's English, just doing an American accent 20:01
lucian: where are you hosting Puffin?
lucian allison: i suppose
allison lucian: source control, I mean
lucian allison: probably bitbucket, unless there are objections
allison lucian: makes sense
(again, mercurial is familiar to Python devs) 20:02
lucian and to me, too :) 20:04
whiteknight allison: I suspect it would be a huge headache at this point to change the license of core parrot 20:07
davidfetter MIT!
whiteknight I don't think it would even be possible
davidfetter i'm pretty sure the CLA guarantees that changing the licenses is a matter for TPF
lucian whiteknight: i thought there was a license agreement
allison davidfetter: yes, that's in the CLA intentionally 20:08
davidfetter :)
allison whiteknight: it would be possible, but it doesn't seem particularly necessary at the moment
whiteknight: especially since AL 2.0 allows relicensing under GPL
davidfetter working on a reason or two to make it a good thing
allison whiteknight: so anyone who wants can use it as GPL instead
davidfetter it's not the GPL that's a problem. it's the permissive ones 20:09
whiteknight ah, that's interesting
I would actually prefer GPL, if I had my druthers
lucian whiteknight: for parrot?
whiteknight lucian: yes
lucian why? it'd make it useless as a library
allison whiteknight: we don't really have any reason to restrict proprietary versions of Parrot
whiteknight it's a personal preference 20:10
lucian i generally agree with using gpl for applications
whiteknight but I wouldn't foist that on the rest of the community
allison whiteknight: really, I'd be perfectly happy if a proprietary version popped up, just as long as *someone* was using it :)
cotto_work I'm excited for someone to start making money using Parrot.
lucian but i think libraries/runtimes should be LGPL at worst
whiteknight lucian: yes, LGPL would be preferrable
allison cotto: exactly
lucian cotto_work: i have a few ideas, they just require a lot of work 20:11
cotto_work lucian: most ideas do
lucian sobs
allison whiteknight: AL 2.0 is a good compromise when you have a community that's kind of half BSD-style and half GPL-style 20:12
20:12 jrtayloriv joined
allison whiteknight: kind of the pragmatic idealists :) 20:12
whiteknight like I said, personal preference. I'm not suggesting we change
lucian one of the reasons i like AL is that it preserves creative control to some degree, but in a LGPL-style, rather than GPL 20:13
it hasn't been tested in court afaik, so i can't be sure
allison whiteknight: yah, I'm just reflecting on the ideas (I've been working a lot lately on legal stuff again, this time for harmonyagreements.org)
lucian but it reads ok
allison lucian: Artistic 1.0 has been tested in court, really the only open source license that has 20:14
davidfetter ?
lucian allison: what about the GPL suits?
davidfetter GPL's held up in .de courts
allison lucian: and Artistic 2.0 sticks pretty closely to the 1.0
lucian allison: but anyway, sounds good then
whiteknight lucian: I'm interested in how you are going to get started with your project. Are you forking an existing project? 20:15
allison lucian/davidfetter: I consider GPL enforcement a different kettle of fish then a general infringement case that just happens to use an open source license, but, I'm probably splitting hairs :)
davidfetter k 20:16
lucian whiteknight: not intending to, no. just using a parser
whiteknight okay
lucian CPython's 'ast' module is pretty good
and it might again get brownie points with PSF 20:17
and maybe nudge them into rewriting it in pure python
whiteknight lucian: if nothing else, that might be a great outcome of your project 20:18
I wonder what that would take?
lucian whiteknight: i suppose so. a bit of waste, though
rewriting it? not too much
PyPy's parser isn't quite API-compatible, but close 20:19
once they port it to py3, it might be an easy job of merging them
20:20 kid51 joined
lucian kid51: re your report, i'd hoped arm/linux would be similar, but all tests pass! dammit :) 20:20
kid51 Yes, I occasionally get errors on one Linux/i386 box that I do not on the other. 20:22
dukeleto kid51: same gcc versions?
kid51 On the second box, I rarely have time to debug further.
dukeleto: I think probably 4.3.n on the first; 4.4.1 on the second 20:23
cotto_work: At YAPC, whether we should have a hackathon on the Thursday of that week depends in part on your availability. Can you be there on the Thursday (1st day after conference itself)? 20:25
cotto_work kid51: sure. I haven't booked my flight yet.
Sticking around for a hackathon is fine for me. 20:26
kid51 Then we can advise atrodo to secure that room (though if it's a large room, we won't need it all for ourselves)
#parrotsketch in 2 20:27
20:28 whiteknight left
cotto_work dukeleto: privmsg ping 20:29
20:31 Andy_ joined 20:34 rohitnsit08 joined 20:36 rohit_nsit08 left, rohitnsit08 left 20:37 rohit_nsit08 joined 20:52 dodathome left
cotto_work tadzik: what timezone are you in? 20:53
tadzik cotto_work: utc + 2 I think, 22:53 for me now
cotto_work so you're at me +9 20:54
aloha: clock?
aloha cotto_work: LAX: Tue, 13:54 PDT / CHI: Tue, 15:54 CDT / NYC: Tue, 16:54 EDT / UTC: Tue, 20:54 UTC / LON: Tue, 21:54 BST / BER: Tue, 22:54 CEST / TOK: Wed, 05:54 JST / SYD: Wed, 06:54 EST
cotto_work tadzik: what do your mornings look like? 20:55
tadzik cotto_work: depends on the morning
every day is different, I have no daily routine (almost)
cotto_work What time are you used to waking up?
6? 9? 12? 20:56
20:56 rohit_nsit08 left
cotto_work If you're an early riser, we can review the Select when you get up tomorrow. 20:56
20:57 Andy_ left, Andy_ joined, Andy_ left, kid51 left, rohit_nsit08 joined, Andy_ joined
tadzik more like 9 20:57
20:58 mj41 left
tadzik I'm rather a night person 20:58
cotto_work ok. That'll be a bit late for me. 20:59
tadzik my 9 is your 0:00, no?
cotto_work yup
I might be up, but not very useful. 21:00
tadzik when is your free time usually, after like 6 PM, 8?
cotto_work after 6-ish 21:02
tadzik so that's my... 27, right? 3 AM, damn
anyway, what are you planning at? 21:03
cotto_work Select review
tadzik well, I'm not much a reviewer, I just look forward to having it merged, so I can give some async IO to Rakudo 21:04
I just know that you commented on the ticket with "I'd like to look more into it"
cotto_work right 21:05
21:05 rohit_nsit08 left
cotto_work I'll just do my review in the ticket or in the pull request 21:05
21:05 rohit_nsit08 joined 21:21 ambs left 21:24 bluescreen left 21:31 jrt4 joined 21:34 jrtayloriv left 21:40 jrt4__ joined, hercynium left 21:41 hercynium joined 21:43 rohit_nsit08 left, jrt4 left 22:02 whiteknight joined 22:03 davidfetter left
tcurtis benabik++ 22:05
22:07 lucian left 22:11 hercynium left 22:16 sigue left, sigue joined 22:22 fperrad left
dalek rrot/tt1931-nci-parameters-deprecation: bd631cb | plobsing++ | src/nci/extra_thunks. (2 files):
add thunk for "ppS" signature used by PCRE
 22:23
22:28 sigue left, sigue joined 22:32 soh_cah_toa joined 22:34 jrt4__ left, jrtayloriv joined 22:35 bubaflub left
dalek p/ctmo: 98eb233 | jonathan++ | src/ (2 files):
Add attribute meta-objects to the HOW exports.
 22:49
p/ctmo: 591efd8 | jonathan++ | src/Regex/Cursor-builtins.pir:
Fix FAILGOAL.
p/ctmo: 4b81716 | jonathan++ | src/stage0/ (7 files):
Update bootstrap to get FAILGOAL fix.
p/ctmo: 0c07f7d | jonathan++ | src/ (3 files):
Add attributes to compile time meta-object for all packages except knowhow (got a circularity issue to resolve there).
22:54 darbelo left 23:00 mtk left 23:04 sigue left 23:07 mtk joined 23:12 bubaflub joined
soh_cah_toa how can i update the branch i forked to make sure it includes all the changes in parrot/parrot? 23:14
23:15 cosimo joined
dukeleto soh_cah_toa: it is described in docs/project/git_workflow.pod 23:15
soh_cah_toa: git pull --rebase
soh_cah_toa: you are using branches, correct?
soh_cah_toa ah, rebase. right
yeah, forked on github
dukeleto soh_cah_toa: github.com/parrot/parrot/blob/mast...rkflow.pod 23:16
soh_cah_toa: you are creating a topic branch and not commiting directly to master, correct?
soh_cah_toa: read that git_workflow.pod, it is the source of much knowledge
soh_cah_toa yeah, i can't commit to master anyway
dukeleto soh_cah_toa: you are confused
soh_cah_toa: master is a branch in a repo 23:17
23:17 sigue joined
dukeleto soh_cah_toa: parrot/parrot.git is a repo with a master branch, but your fork has a master branch as well 23:17
soh_cah_toa: makes sense?
soh_cah_toa oh right
then no, i think
dukeleto soh_cah_toa: :)
23:17 dmalcolm left
dukeleto soh_cah_toa: read those docs, and then send an email to parrot-dev with your question 23:17
soh_cah_toa: if you have it, then other gsoc students will too 23:18
soh_cah_toa okay
dukeleto soh_cah_toa: i am going to work out, and will answer it when I get back (if somebody else hasn't already)
soh_cah_toa dukeleto: alright, thanks
benabik is very unlikely to have git questions, but is happy to try to provide answers.
23:20 bacek joined
dukeleto soh_cah_toa: benabik++ most likely knows a lot more about git than me, so feel free to bother him :) 23:21
dukeleto really goes
benabik should have kept his mouth shut. ;-)
soh_cah_toa benabik: ha, well your lucky that i have class in a few minutes so i won't bother you for too long 23:22
it seems that 'git pull --rebase' is not working as i had hoped
i'm working on a copy that i forked from parrot/parrot on github 23:23
forked pre-3.3.0
benabik It should take your changes on the current branch and put them on top of the original branch in parrot.git
What problem are you having with it? 23:24
soh_cah_toa well there are patches that i applied to parrot/parrot that i don't have in soh-cah-toa-/parrot
benabik The patches are in your local repo? 23:25
soh_cah_toa yeah. they're wrong though. after those patches i made some fixes to them and then submitted. now the correct patches are in parrot/parrot but the "immature" ones are in soh-cah-toa/parrot 23:26
i'm thinking maybe just delete the repo and re-fork but i'm not sure if i'd loose anything 23:27
23:27 Andy_ left
benabik That's not needed. 23:27
soh_cah_toa but there's gotta be a way to do this
i'm sure there's just some git command that i'm unfamiliar w/ that's the solution 23:28
benabik If there's nothing in your local branch you need, you can `git fetch; git reset --hard @{u}`. (@{u} means upstream branch and was added in git 1.7
(Note that --hard may lose uncommitted changes)
soh_cah_toa blah, yeah it did 23:29
benabik I shouldn't have added the --hard, or advised stashing first. Stupid finger memory. 23:30
I usually either have everything committed or stash before pulling.
benabik--
soh_cah_toa i can recommit in about 6 seconds
benabik should remember not to tell people to do things that might lose data. 23:31
soh_cah_toa alright, back to where i started 23:32
benabik \\o/ 23:33
soh_cah_toa git reset w/o the --hard?
benabik "git reset @{u}" will set your current branch to be identical to the upstream branch.
The --hard also changes your files to match.
23:34 Kulag left
soh_cah_toa hmm...i think someone already applied the patch i submitted to parrot/parrot 23:34
b/c git reset @{u} has the changes i submitted to trac about 20 mins ago 23:35
and that's what i wanted to do w/ the fork
except i'm not in the log. hmm...
benabik Check "git status". 23:36
If the changes were in your working copy, they'll still be there if you didn't use reset --hard
soh_cah_toa ah. there they are
agh. i really gotta get going though
benabik You can now recommit them or blow them away as you see fit.
soh_cah_toa this is what i wanted (i think). thanks for the help 23:37
soh_cah_toa heads off to school :(
benabik Yay school.
dalek p/ctmo: cdae853 | jonathan++ | src/NQP/Actions.pm:
Oops, fix that type lookup.
 23:40
p/ctmo: 509ee04 | jonathan++ | src/HLL/SerializationContextBuilder.pm:
Fix an attribute access that was wrong, caught by the in-progress undeclared attribute detection.
p/ctmo: 1a79be7 | jonathan++ | / (4 files):
Add .type to PAST::Var nodes. Also twiddle the build a bit to make sure we build changes to the PAST extensions.
p/ctmo: eafc258 | jonathan++ | src/NQP/ (2 files):
Detect undeclared attributes at compile time and complain about them. Possible now that we register them with the compile timemeta-object. Also, pass any declared type along to the PAST::Var node, though it's not used yet.
 23:41
p/ctmo: cf88f9e | jonathan++ | src/how/NQP (2 files):
Standardize introspection interface implemented in the role meta-objects. Unregresses the roles.t breakage in the last commit.
23:43 sigue left, sigue joined 23:47 Kulag joined 23:51 sigue left, sigue joined

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!