Da
Show enters and exits. Hide enters and exits.
| 00:02:37 | slava | brixen: going pretty well |
| 00:51:57 | evan | woop |
| 00:52:07 | evan | who wants to help uncle evan test the Rubinius OS X .pkg |
| 00:52:34 | brixen | me me me! |
| 00:52:39 | evan | ok |
| 00:52:44 | evan | it's uploading to elle now. |
| 00:52:48 | brixen | k |
| 00:53:29 | brixen | it's really weird having an uncle 9 years my junior heh |
| 00:53:43 | evan | it happens all the time with big families |
| 00:53:44 | slava | whoa you guys are old :P |
| 00:53:47 | brixen | heh |
| 00:53:49 | evan | we're a big happy family |
| 00:53:57 | brixen | oh slava, you kid you |
| 00:53:59 | evan | i have a friend that is the youngest of 12 |
| 00:54:11 | evan | and the gap between him and his next sibling is 10 years |
| 00:54:24 | evan | he was a bit of a... surprise! |
| 00:54:30 | brixen | heh, I bet |
| 00:54:41 | evan | he's got uncles that are younger than him. |
| 00:55:24 | evan | http://asset.rubini.us/Rubinius-1.0.0-rc1.pkg |
| 00:55:41 | evan | installs into /usr/local/rubinius/1.0.0-rc1 |
| 00:55:53 | evan | with a soft link from /usr/local/bin/rbx to 1.0.0-rc1/bin/rbx |
| 00:56:13 | brixen | woot, nice logo |
| 00:56:39 | slava | when's rc2 due? |
| 00:56:51 | brixen | only 29 MB, sweet |
| 00:57:02 | evan | slava: wednesday probably |
| 00:57:14 | evan | i'm not going to announce the .pkg's until I do an rc2 one |
| 00:57:22 | evan | you're free to play with the rc1 one though |
| 00:57:55 | brixen | hmm |
| 00:58:19 | evan | brixen: you're on 10.5 right? |
| 00:58:23 | brixen | yeah |
| 00:58:29 | evan | you're gonig to be guinne pig then. |
| 00:58:37 | evan | because that's a 10.6, 64bit build |
| 00:58:45 | evan | i've got a package restriction in there |
| 00:58:50 | evan | it will only install on 64bit available systems |
| 00:59:29 | slava | only 64 bits? |
| 00:59:32 | slava | jesus, I hope by the time you hit 1.0 you're at least 128 or 256 bit |
| 00:59:48 | evan | you mean in january? |
| 00:59:55 | evan | or february? |
| 00:59:56 | brixen | slava: how's that beer you're having? |
| 01:00:08 | brixen | evan: um, so it should or should not install? |
| 01:00:10 | slava | brixen: its only 2pm, haven't had my beer yet |
| 01:00:11 | brixen | because it did |
| 01:00:13 | brixen | http://gist.github.com/261403 |
| 01:00:19 | brixen | but it won't run |
| 01:00:27 | evan | ok |
| 01:00:29 | evan | thats fine |
| 01:00:32 | slava | adding some assertions to memory allocation code to track down why its making a 0-size block and hanging |
| 01:00:35 | evan | you were my guinne pig! |
| 01:00:40 | brixen | heh |
| 01:01:06 | brixen | I shall consider myself guineaed then |
| 01:01:13 | evan | :) |
| 01:01:42 | brixen | I bought SL this weekend for installing on my old mbp |
| 01:01:42 | evan | I suspect i need to pass some gcc flags |
| 01:01:48 | brixen | haven't done it yet though |
| 01:01:57 | evan | to get gcc to output and link for usage on both 10.5 and 10.6 |
| 01:02:04 | brixen | is this an XCode project? |
| 01:02:08 | evan | nah. |
| 01:02:10 | slava | could you just build on 10.5? |
| 01:02:12 | evan | thats way more work. |
| 01:02:18 | evan | slava: sure, but i'd prefer to not have to |
| 01:02:27 | brixen | oh, so you just need to make a fat binary then? |
| 01:02:29 | slava | do you still suppor 10.4? |
| 01:02:41 | evan | slava: should still work on 10.4 yes |
| 01:02:45 | brixen | we did recently, not sure if it's broken |
| 01:02:48 | evan | but i'm less concerned with our initial .pkg working on there |
| 01:02:51 | brixen | slava: you should try it! |
| 01:02:56 | slava | don't have 10.4 handy |
| 01:11:38 | evan | brixen: what version of g++ is the default for you? |
| 01:14:37 | brixen | 4.0.1 |
| 01:14:42 | evan | k |
| 01:14:44 | evan | thought so |
| 01:14:57 | evan | trying to figure out how to build "down" to 10.5 |
| 01:15:03 | evan | in a way that will work on 10.6 and 10.5 |
| 01:15:13 | brixen | ah hmm |
| 01:15:21 | evan | -isysroot /Developer/SDKs/MacOSX10.5.sdk |
| 01:15:25 | evan | seems to be the major magic. |
| 01:15:45 | evan | i'm not sure if LLVM will link with this though... |
| 01:16:02 | evan | yeah, looks like no. ok... |
| 01:16:47 | evan | good think we've got the prebuilts |
| 01:16:50 | evan | fudges the results. |
| 01:25:45 | evan | brixen: could you do me a favor? |
| 01:25:50 | brixen | yup |
| 01:26:11 | evan | ok, one sec. |
| 01:26:17 | evan | let me push sometihng to this easier. |
| 01:26:22 | brixen | k |
| 01:27:24 | boyscout | Style cleanup of StringIO - d885885 - Evan Phoenix |
| 01:27:24 | boyscout | Fix FFI numerical edge case - ac8f7e1 - Evan Phoenix |
| 01:27:24 | boyscout | Simplify the argument handler usage - 4d22ca1 - Evan Phoenix |
| 01:27:24 | boyscout | Add FAKEROOT support to install for packaging - 405489c - Evan Phoenix |
| 01:27:31 | evan | ok, do |
| 01:27:44 | evan | ./configure --prefix=/usr/local/rubinius/test |
| 01:27:50 | evan | the |
| 01:27:54 | evan | mkdir tmp |
| 01:28:01 | evan | rake install FAKEROOT=tmp |
| 01:28:13 | brixen | k |
| 01:28:25 | evan | that treats tmp as / for install |
| 01:28:30 | evan | so the internal paths are right |
| 01:28:33 | evan | and we can package |
| 01:28:40 | evan | tar up tmp and send it to me |
| 01:28:44 | evan | scp to elle is fine |
| 01:28:47 | brixen | k |
| 01:28:52 | evan | i wanna see if 10.5 builds work on 10.6 |
| 01:30:21 | boyscout | CI: 405489c success. 3019 files, 11617 examples, 35755 expectations, 0 failures, 0 errors |
| 01:34:24 | brixen | evan: do you want tmp/ in the path when I tar it or not? |
| 01:34:33 | evan | doesn't matter |
| 01:34:36 | evan | with is probably easier |
| 01:34:38 | brixen | k |
| 01:35:27 | brixen | rubinius-1.0.1rc1-OSX-10.5.8.tgz in my home dir |
| 01:35:51 | brixen | hmm |
| 01:36:00 | brixen | scp seems to be hanging... |
| 01:36:31 | brixen | le'me resend just in case |
| 01:36:54 | brixen | k, done |
| 01:37:41 | evan | k |
| 01:38:19 | evan | i LOVE remote completion |
| 01:38:28 | evan | scp e:/home/brixen/rubin<tab> |
| 01:38:40 | brixen | nice |
| 01:38:44 | brixen | how do you enable that? |
| 01:39:10 | evan | use zsh :) |
| 01:39:22 | evan | and it does it automatically |
| 01:39:32 | evan | so long as you have keys setup so you can login without a password |
| 01:39:53 | brixen | oh nice |
| 01:41:02 | evan | most people never try |
| 01:41:09 | evan | because they think "that will never work" |
| 01:43:59 | brixen | I need to play with zsh |
| 01:44:14 | evan | ok, a 10.5 build works on 10.6 |
| 01:45:21 | evan | i wonder why I can't get it to cross build. |
| 01:57:50 | benschwarz | evan: no silly, fish :) |
| 01:58:11 | evan | ?? |
| 01:58:13 | benschwarz | shell |
| 01:58:34 | evan | never need it. |
| 01:58:51 | benschwarz | why zsh over bash? |
| 01:59:02 | evan | more power. |
| 01:59:05 | evan | things I like. |
| 01:59:19 | slava | evan: power corrupts |
| 01:59:23 | evan | like %2~ |
| 01:59:26 | slava | that's why I use the original bourne shell |
| 01:59:32 | evan | i can't live without %2~ now. |
| 01:59:37 | benschwarz | I don't know much about zsh, but I do know that bash has some of the worst UX a shell could have |
| 01:59:48 | evan | its. |
| 01:59:49 | evan | a shell. |
| 01:59:57 | evan | what UX do you EXPECT it to have? |
| 02:00:08 | slava | scantly-clad ladies |
| 02:00:25 | benschwarz | slava: its the aussie way |
| 02:00:53 | benschwarz | evan: thats like saying a dsl / interface isn't a user expereience |
| 02:00:57 | benschwarz | we both know better :) |
| 02:03:53 | evan | it's more about expectations |
| 02:03:57 | evan | what do you want that it doesn't do? |
| 02:10:49 | evan | brixen: i think the thing to do is we'll create 10.5 and 10.6 pkgs seperately for now |
| 02:10:59 | evan | i found the cli for creating a .pkg |
| 02:11:04 | evan | so i'm going to wrap it up in a rake task |
| 02:11:14 | evan | so it should be easy |
| 02:13:52 | brixen | k |
| 03:09:39 | boyscout | Add package:osx to create .pkgs - fed666f - Evan Phoenix |
| 03:09:51 | evan | brixen: so, if you configure --prefix |
| 03:09:52 | evan | then do |
| 03:09:56 | evan | rake package:osx |
| 03:10:02 | evan | it will do the right thing |
| 03:11:12 | evan | it seems to make a flat, directory based .pkg |
| 03:11:16 | evan | rather than the self containted one |
| 03:11:25 | evan | so i guess you have to zip it up |
| 03:12:30 | boyscout | CI: fed666f success. 3019 files, 11617 examples, 35755 expectations, 0 failures, 0 errors |
| 04:10:36 | brixen | pulling |
| 04:19:54 | brixen | evan: rubinius-1.0.0-rc1.pkg.zip in my home dir |
| 04:40:51 | brixen | I like how easy it is to configure rbx for install or not and switch back and forth |
| 04:41:12 | brixen | with just a 'rake' in between to build |
| 08:28:23 | dbussink | evan: i got a tad bit of experience with building universal binaries, or do you already worked on that/ |
| 08:28:23 | dbussink | ? |
| 08:28:41 | dbussink | evan: you can pass gcc some flags so you get a 32 and 64 bit build in one, would be useful for the installer |
| 13:13:16 | rue | It should be just a matter of passing both arch flags. |
| 15:25:13 | dbussink | rue: yeah, it should yeah, although if you have some pregenerated constants that depend on cpu arch would be annoying |
| 15:25:26 | dbussink | rue: but don't think rubinius has that issue |
| 15:25:43 | dbussink | rue: annoying this is that all external libs need to be universal too |
| 17:30:00 | brixen | buenos dias muchachos y muchachas |
| 17:30:58 | BrianRice-work | hay muchachas aqui? |
| 17:31:10 | brixen | I wish |
| 17:31:13 | brixen | um, I mean, dunno |
| 17:31:16 | rue | Si |
| 17:31:16 | brixen | heh |
| 17:31:23 | brixen | oh yeah, rue's here |
| 17:31:35 | rue | I was thinking more yugui, but sure. |
| 17:31:45 | brixen | heh, just keeeeding rue |
| 17:31:47 | BrianRice-work | good. monocultures suck |
| 17:32:10 | brixen | yes, the awesome yugui is here, and we are honored |
| 17:33:02 | brixen | I would think Ruby would attract more women than say, python |
| 17:33:10 | brixen | we obviously suck at marketing |
| 17:34:22 | brixen | calls for proposals for OSCON is open! |
| 17:34:45 | brixen | I want to do an *interesting* talk this year... |
| 17:37:30 | Defiler | brixen: u.s. rails devs are a disaster around women |
| 17:37:43 | Defiler | There is zero mystery to me in our gender stats |
| 17:37:48 | brixen | :( |
| 17:38:11 | brixen | Defiler: we must brainstorm to rectify this unfortunate circumstance |
| 17:38:35 | brixen | Defiler: btw, 1/21-25 |
| 17:38:44 | Defiler | We should. The more women we have in the community, the lower the odds of any given one being jacqui |
| 17:38:52 | Defiler | brixen: gonna be in NYC? |
| 17:38:53 | brixen | Defiler: that's a lot of advance notice, you better be around *one* of those days |
| 17:38:56 | brixen | yes |
| 17:39:06 | Defiler | Yeah, I should be around for that whole span of days |
| 17:39:11 | Defiler | Let's have ramen |
| 17:39:16 | brixen | indeed |
| 17:46:55 | probablyCorey | I miss NYC ramen |
| 17:49:33 | evan | morning. |
| 17:50:04 | brixen | morning |
| 17:50:58 | evan | read the interesting post-mortem on Duke Nukem Forever |
| 17:51:40 | brixen | link? |
| 17:51:59 | evan | http://www.wired.com/magazine/2009/12/fail_duke_nukem/all/1 |
| 17:52:57 | Defiler | probablyCorey: If you need someone to escort you to ramen while you are visiting, let me know :) |
| 17:53:42 | Defiler | I still have an unopened pack of Duke Nukem Forever playing cards from E3 1997 |
| 17:53:51 | Defiler | I wonder what I could get for them on ebay heh |
| 17:55:05 | evan | probably a decent ammount |
| 17:55:18 | evan | since the graphics on the cards probably look nothing like would have come out |
| 17:55:35 | evan | the dev for DNF sounds like your typical death march |
| 17:55:48 | evan | management with no discipiline |
| 18:02:03 | probablyCorey | Game Development has the worst software/management practices. It's a real life parody of software development. |
| 18:03:29 | evan | probablyCorey: that doesn't surprise me at all. |
| 21:59:48 | evan | man, Kim Peek died. |
| 22:01:53 | brixen | :( |
| 22:02:19 | evan | i wonder how much they figured out about how he worked |
| 22:03:09 | brixen | good question |
| 22:12:12 | evan | well this is a fun bug! |
| 22:12:28 | evan | rack works when there are NOT .rbc files |
| 22:12:36 | evan | but does NOT work when there are |
| 22:12:52 | evan | puts on his pith helmet |
| 22:13:01 | evan | once more into the breach! |
| 22:14:07 | brixen | hmm |
| 22:14:26 | evan | the little debugger behind my ear tells me that it might be a string duping issue |
| 22:14:32 | evan | or rather, a lack of duping. |
| 22:15:01 | evan | hey look at that. |
| 22:15:09 | evan | the little debugger was right |
| 22:15:14 | brixen | heh |
| 22:15:38 | evan | the constant "/" it's being destroyed somehow |
| 22:15:49 | evan | and the literals table is probably not duping the string |
| 22:19:13 | brixen | lypanov!!! |
| 22:19:28 | evan | holy shit. |
| 22:19:30 | lypanov | ozmog! |
| 22:19:36 | evan | pours popcorn over lypanov's head |
| 22:19:44 | lypanov | vane! |
| 22:19:57 | evan | it is I! |
| 22:19:57 | lypanov | nxreib! |
| 22:20:02 | evan | rotated by 1! |
| 22:20:28 | lypanov | naah, random shuffle man |
| 22:20:35 | evan | i just got lucky then. |
| 22:21:44 | lypanov | how goes conquering the world? |
| 22:22:29 | evan | oh, good. |
| 22:22:30 | evan | you know |
| 22:22:36 | evan | how's scala? |
| 22:29:38 | brixen | dbussink: ping |
| 22:30:37 | lypanov | evan: i'm stuck in a crazy inbetween world |
| 22:31:37 | evan | TYPE LIMBO |
| 22:31:41 | lypanov | evan: migrating massive amounts of crazy erb code to pure javascript with html injection |
| 22:31:55 | lypanov | next step, moving from js / html to gwt / html |
| 22:32:15 | lypanov | after that moving clean pure ruby backend to something in scala |
| 22:32:30 | lypanov | but "inbetween world" hosting is a bitch |
| 22:32:47 | lypanov | basically just need to go pure ec2 afaict |
| 22:33:18 | lypanov | erb is the most evil thing ever thrust upon this world imo |
| 22:33:30 | lypanov | rails == pretty php |
| 22:33:41 | lypanov | all the same fuckups. just prettier. shorter. more complex |
| 22:34:36 | evan | thankfully you don't have to use erb |
| 22:36:59 | lypanov | evan: with rails you have little other choice |
| 22:37:08 | evan | plenty of ones |
| 22:37:14 | evan | haml, moustache, etc |
| 22:37:17 | evan | lots of them |
| 22:37:22 | evan | new ones every month |
| 22:37:25 | lypanov | mustache is a few months old |
| 22:37:25 | lypanov | haml is a joke |
| 22:37:32 | lypanov | mustache is the first serious one |
| 22:37:43 | lypanov | and you can throw away basically 50% of the point of rails if you use it |
| 22:37:47 | lypanov | rails3 is going the right dir |
| 22:37:58 | lypanov | but rails 2.3 is a complete joke imo from maintainability standpoint |
| 22:38:34 | lypanov | lack of serious support for unobtrustive js makes it simply unusable |
| 22:38:51 | lypanov | sive* |
| 22:39:11 | evan | thankfully, rubinius runs them both |
| 22:39:15 | evan | and whatever comes after them. |
| 22:44:10 | lypanov | without type checking / good reliable refactoring i'll be long gone anyway. |
| 22:44:18 | evan | by then! |
| 22:44:20 | evan | bye rather. |
| 22:44:26 | evan | we'll miss you. |
| 22:44:30 | lypanov | 5 years of being forced to use java has injected sanity into my brain ;) |
| 22:44:37 | lypanov | i still love the runtime, don't get me wrong. |
| 22:44:43 | lypanov | just wouldn't use it. |
| 22:45:56 | lypanov | i feel test case induced type checking possibly is the way to go. |
| 22:46:11 | lypanov | either way, whatever way it is that its done. |
| 22:46:22 | evan | ew. |
| 22:46:32 | brixen | mega ew |
| 22:46:35 | brixen | what's that get you? |
| 22:46:48 | lypanov | refactoring. |
| 22:46:54 | brixen | since when does type-checking imply maintainability? |
| 22:47:01 | lypanov | type checking. |
| 22:47:10 | lypanov | sorry. typo* checking |
| 22:47:24 | brixen | lost |
| 22:47:29 | lypanov | figured |
| 22:47:45 | lypanov | "ew" + "insanely cool idea" -> didn't explain well |
| 22:47:50 | lypanov | :P |
| 22:48:09 | brixen | I'm not ignorant of types |
| 22:48:19 | lypanov | i think the code coverage tools currently available completely blow |
| 22:48:21 | brixen | I'm highly, extremely skeptical of all claims made about them |
| 22:48:33 | lypanov | i feel taking typing into account when doing coverage would help a huge amount |
| 22:48:52 | evan | does scala have good tools? |
| 22:49:06 | lypanov | and i also feel that taking typing constraints into account when *programming* would aid program execution optimization no end |
| 22:49:10 | lypanov | no. it stinks. |
| 22:49:19 | evan | so typing doesn't improve your tools |
| 22:49:21 | evan | humans do. |
| 22:49:23 | lypanov | slow as fuck compiler. crap refactoring. |
| 22:49:35 | lypanov | scala is possible to refactor. |
| 22:49:37 | lypanov | ruby is not. |
| 22:49:43 | lypanov | java has amazing tools. |
| 22:50:16 | lypanov | so amazing in fact that i'd rather code in gwt than in js/gwt |
| 22:50:22 | lypanov | sorry. js/jquery* |
| 22:50:47 | lypanov | (don't underestimate how frustrating a ring finger cast can be) |
| 22:51:01 | lypanov | can't type |
| 22:51:25 | lypanov | anyway... my idea is crazy |
| 22:51:26 | evan | sorry lypanov |
| 22:51:27 | evan | you're wrong |
| 22:51:32 | evan | i refactor ruby fine. |
| 22:51:54 | lypanov | but i once explained it to mr factor and he agreed. and he's always right. |
| 22:52:06 | evan | mr. factor? slava? |
| 22:52:07 | lypanov | evan: automated? |
| 22:52:09 | lypanov | aye |
| 22:52:14 | evan | slava hates ruby |
| 22:52:19 | lypanov | right. |
| 22:52:25 | lypanov | for a good reason :P |
| 22:52:27 | evan | *eyerloll* |
| 22:52:35 | evan | i like ya |
| 22:52:38 | evan | but are you in here just to troll us? |
| 22:52:50 | lypanov | dude. i've hated coding ruby for 4-5 years... |
| 22:53:06 | evan | why were you writing your own ruby environment then? |
| 22:53:20 | lypanov | because i felt a good vm might improve things. |
| 22:53:23 | lypanov | and i still feel that. |
| 22:53:49 | evan | so your problem is with typing or tools? |
| 22:53:53 | lypanov | but without a fast runtime, non of what i want is worthwhile anyway. |
| 22:53:55 | lypanov | tools. |
| 22:54:07 | lypanov | i don't give a crap about the method. i just care about quality. |
| 22:54:17 | evan | smalltalk has automated refactoring tools |
| 22:54:24 | evan | and smalltalk has less type info than ruby |
| 22:54:24 | lypanov | right. |
| 22:54:27 | lypanov | aye. |
| 22:54:33 | evan | so it's not a language problem |
| 22:54:35 | lypanov | as i said. typing has little to do with it. |
| 22:54:42 | evan | it's people getting off their asses and writing some tools. |
| 22:54:48 | lypanov | but i'm crap at explaining myself. |
| 22:55:14 | brixen | sounds simple to me: ruby needs better tools |
| 22:55:19 | brixen | there, I said it :) |
| 22:55:22 | lypanov | aye. |
| 22:55:24 | evan | i think we'd all agree there. |
| 22:55:29 | brixen | you can build them on rubinius btw |
| 22:55:34 | lypanov | right. |
| 22:55:49 | lypanov | that was the point in rubinius for me, the possibility of better tools. nothing more. |
| 22:55:57 | brixen | and speed |
| 22:55:59 | brixen | and reliability |
| 22:56:02 | brixen | and a real GC |
| 22:56:02 | lypanov | better tools without a faster runtime is impossible. |
| 22:56:07 | brixen | etc... |
| 22:56:22 | lypanov | overhead of adding tracing was insane with mri |
| 22:56:25 | brixen | lypanov: come build some tool then |
| 22:56:30 | evan | lypanov: thats why we're here |
| 22:56:37 | brixen | tools even |
| 22:56:39 | evan | building a great platform for future work |
| 22:56:43 | brixen | why limit yourself to just one :) |
| 22:56:44 | evan | as fast as we can. |
| 22:56:55 | lypanov | how speed lately? |
| 22:57:03 | evan | 114x faster on one benchmark |
| 22:57:04 | lypanov | global lack of s's |
| 22:57:04 | evan | :D |
| 22:57:08 | lypanov | hehe |
| 22:57:13 | lypanov | i had a 3 |
| 22:57:15 | lypanov | oops |
| 22:57:20 | lypanov | 2000x bench speed up |
| 22:57:28 | Zoxc | 54x slower on others :D |
| 22:57:35 | lypanov | i hope your bench didn't blow as much as mine :P |
| 22:57:36 | evan | Zoxc: ICE BURN |
| 22:57:43 | evan | lypanov: nope, they all run fine |
| 22:57:46 | evan | and they're all compliant |
| 22:57:46 | lypanov | only 54? lame |
| 22:58:02 | lypanov | when i started i was 2 million times slower than mri |
| 22:58:10 | lypanov | after 2 weeks i was 2000 times faster |
| 22:58:25 | rue | That is pretty impressive. |
| 22:58:26 | lypanov | (1..2000 +=1 loop) |
| 22:58:36 | lypanov | it was fun, yes. |
| 22:58:39 | evan | thats why i'm sad you're not helping anymore. |
| 22:59:30 | lypanov | should get evan to chat with his wife so that his wife says "did you work on rubinius yet today?" every day |
| 22:59:42 | evan | sure! |
| 22:59:49 | evan | or we could have our wives discuss it together |
| 22:59:52 | evan | let them work it out. |
| 22:59:59 | lypanov | :P |
| 23:00:16 | evan | solved the webrick problem, it was a cached hash value problem again |
| 23:00:19 | evan | btw |
| 23:00:40 | lypanov | evan: have you done any ultra 1337 mem opt things yet btw? |
| 23:00:48 | lypanov | eg homogenous array opts? |
| 23:00:49 | Zoxc | wonders why the Boehm GC shipped with Ubuntu 9.10 crashes with mirb </hijack> |
| 23:00:53 | evan | work on the GC |
| 23:01:03 | evan | lypanov: no, talked with slava about them a little |
| 23:01:10 | evan | but thats about it |
| 23:01:18 | evan | we haven't yet it somewhere that needed them |
| 23:01:22 | evan | so we haven't put any work into it. |
| 23:01:25 | lypanov | slava is doing f'n awesome things with factor. blown away. |
| 23:01:45 | lypanov | shame concatenative coding has to suck so much. |
| 23:01:49 | lypanov | i'm too stupid apparently ;) |
| 23:02:17 | evan | me too. |
| 23:02:24 | evan | too stupid that is. |
| 23:03:48 | evan | factor is a nice green field |
| 23:03:49 | evan | too :) |
| 23:06:02 | evan | was reading my VM book last night |
| 23:06:17 | evan | and thinking about a way to moving object allocation inline |
| 23:06:21 | evan | i think i've got it |
| 23:06:39 | lypanov | you coded much c++ in past? |
| 23:06:52 | evan | some in the way past |
| 23:07:02 | lypanov | not years worth? |
| 23:07:03 | evan | before I started up on c++ rubinius a year ago |
| 23:07:08 | evan | nah |
| 23:07:11 | lypanov | k |
| 23:07:14 | evan | never professionally |
| 23:07:17 | evan | i consider that a good thing |
| 23:07:18 | lypanov | ah |
| 23:07:26 | evan | my mind wasn't clouded with boastian crap. |
| 23:07:35 | lypanov | ah no |
| 23:07:41 | lypanov | me == ex kde devel |
| 23:07:46 | evan | ah yes |
| 23:07:48 | evan | i recall. |
| 23:07:49 | lypanov | qt. no insane shit. |
| 23:07:59 | lypanov | stl == teh lame. |
| 23:08:06 | evan | moc is a little bit crazy. |
| 23:08:08 | evan | a little bit. |
| 23:08:17 | lypanov | not really. |
| 23:08:28 | evan | a C++ preprocessor? |
| 23:08:30 | evan | thats a little crazy. |
| 23:08:32 | evan | not bad crazy. |
| 23:08:37 | evan | but a little. |
| 23:08:42 | lypanov | its the reason why gtk apps segfault but qt apps say "boo boo" on the console |
| 23:09:00 | evan | oh for sure |
| 23:09:07 | evan | it's a loose, late binding mechanism |
| 23:09:09 | evan | and it's pretty damn cool |
| 23:09:13 | lypanov | moc is just a good enough work around for a crapness in c lang |
| 23:09:17 | lypanov | aye |
| 23:09:24 | lypanov | signals concept rocks |
| 23:09:43 | lypanov | messaging. |
| 23:09:53 | lypanov | not really dyn vs static |
| 23:10:04 | lypanov | but distrib vs completely lame |
| 23:10:22 | evan | heh |
| 23:10:58 | lypanov | is watching a fucking wonderful talk by mr tangasol aka cameran purdy |
| 23:11:09 | lypanov | can't spell his name |
| 23:11:29 | lypanov | evan: actually. did you watch... *digs up* |
| 23:13:26 | lypanov | http://www.youtube.com/watch?v=5uljtqyBLxI |
| 23:13:59 | lypanov | this guy is f'in scary |
| 23:14:09 | brixen | cliff click is fun to watch |
| 23:14:30 | lypanov | aye. genius. |
| 23:15:31 | lypanov | grid computing is certainly the way to go for the coming years. |
| 23:15:33 | evan | hm, don't think i've seen this one |
| 23:16:01 | lypanov | needs to give gridgain.org a try |
| 23:16:32 | lypanov | evan: have you looked at swarm? |
| 23:16:43 | evan | nope |
| 23:16:49 | lypanov | or the scala 2.8 feature "delimited continuations"? |
| 23:16:50 | lypanov | k |
| 23:17:14 | lypanov | basically, swarm lets you continue do remote continuation |
| 23:17:42 | lypanov | s/continue// |
| 23:18:13 | lypanov | automatically transfers control back to the machine that has the data thats being operated on the most |
| 23:18:40 | lypanov | terracotta tech is related |
| 23:18:57 | lypanov | basically a multi machine multi gb shared memory space |
| 23:19:31 | lypanov | imo these are the sort of things that can make rubinius kick alt langs ass in the coming years |
| 23:19:50 | lypanov | but "200 x slower than java" was never enough |
| 23:19:57 | lypanov | i think rubinius can be 5x slower |
| 23:20:02 | lypanov | 5x slower is completely doable |
| 23:21:01 | lypanov | i thought jruby would have a major advantage for our app actually but apparently image processing on java platform completely stinks |
| 23:27:50 | evan | i'm glad click is getting all this figured otu |
| 23:27:53 | evan | so I don't have to. |
| 23:29:12 | brixen | heh |
| 23:32:24 | lypanov | ;) |
| 23:34:08 | boyscout | Fix the double strerror() output in the message - 34e19ca - Evan Phoenix |
| 23:34:08 | boyscout | Reformatting - 4ad0346 - Evan Phoenix |
| 23:34:08 | boyscout | Use FFI to get a file descriptor - 20ce0d0 - Evan Phoenix |
| 23:34:08 | boyscout | Fix stale String hash bug - 3120cab - Evan Phoenix |
| 23:34:08 | boyscout | Turn off ri and rdoc generation by default - 6764004 - Evan Phoenix |
| 23:34:09 | boyscout | Add -s to 'gem install' to show whats going on - 515333b - Evan Phoenix |
| 23:37:42 | boyscout | CI: 515333b success. 3019 files, 11617 examples, 35755 expectations, 0 failures, 0 errors |
| 23:51:22 | lypanov | meh. gotta run. |