Conversation with #inferno at Fri Mar 25 06:29:18 2011 on powerman-asdf@irc.freenode.net (irc) (06:55:13) bugQ left the room (quit: Quit: but don't be sad.). (07:31:27) perdix [~mkhl@sxemacs/devel/perdix] entered the room. (07:34:41) tensai_cirno [~cirno@194.154.66.97] entered the room. (08:09:19) Guest28129 [~algol@188.24.73.8] entered the room. (08:10:01) Guest28129 left the room (quit: Read error: Connection reset by peer). (08:12:54) fddfoo [~algol@188.24.73.8] entered the room. (08:12:54) fddfoo left the room (quit: Changing host). (08:12:54) fddfoo [~algol@unaffiliated/fdd] entered the room. (08:21:11) cirno_ [~cirno@194.154.66.97] entered the room. (08:21:11) tensai_cirno left the room (quit: Read error: Connection reset by peer). (09:35:14) cirno_ left the room (quit: Quit: Leaving). (09:44:14) perdiy [~mkhl@f055051028.adsl.alicedsl.de] entered the room. (09:44:14) perdiy left the room (quit: Changing host). (09:44:14) perdiy [~mkhl@sxemacs/devel/perdix] entered the room. (09:44:26) perdix left the room (quit: Ping timeout: 255 seconds). (10:09:35) perdiy is now known as perdix (10:21:50) mjl-: howdy (10:35:29) robot12 [~kazzhilki@proxy10.ts.fujitsu.com] entered the room. (10:53:18) me_ left the room (quit: Ping timeout: 250 seconds). (10:53:24) me_ [~me@c-68-48-11-23.hsd1.md.comcast.net] entered the room. (11:29:33) tensai_cirno [~cirno@77.232.15.216] entered the room. (12:22:09) GriffenJBS: mjl-: hello (13:35:47) robot12 left the room (quit: Ping timeout: 240 seconds). (13:46:56) fddfoo left the room (quit: Quit: 10100011010101000011100101.00.). (13:51:04) arvindht left the room (quit: Quit: Page closed). (13:59:33) me_ left the room (quit: Read error: Operation timed out). (14:35:02) vpm left the room (quit: Quit: Take me to your leader!). (14:37:51) vpm [~vpm@reverse-94.fdn.fr] entered the room. (15:00:27) tensai_cirno left the room (quit: Quit: Leaving). (15:59:27) Fish left the room (quit: Ping timeout: 260 seconds). (16:02:20) mjl-: hej GriffenJBS (16:03:55) GriffenJBS: what is going on with vita nuova? at times they seem to be the keepers of inferno, but so much of their info is not current (16:07:53) mjl-: they are the keepers (16:07:57) mjl-: copyright holders :) (16:08:08) mjl-: but i'm pretty sure they are not a big company (16:08:14) GriffenJBS: is there a liason? (16:08:15) mjl-: so their resources are also limited (16:08:24) mjl-: what do you mean by that? (16:08:34) GriffenJBS: a company person to contact? (16:08:54) mjl-: charles forsyth is the one you want to talk to (16:09:13) GriffenJBS: ... ?! your very involved with inferno, may I ask your connection? (16:09:39) mjl-: of course (16:09:44) mjl-: i don't have any official connection (16:10:09) mjl-: came into contact with plan 9 around 2000, as a CS student, used it a bit, then found inferno and developed for it a bit (16:10:22) mjl-: that's mostly it :) (16:10:35) mjl-: i've gone to a few plan 9 workshops (16:10:44) GriffenJBS: I still don't exactly get the difference between plan 9 and inferno (16:10:46) mjl-: so i've met some of the people involved (16:11:10) mjl-: the most obvious ones: programs in inferno are written in limbo (16:11:29) mjl-: and inferno can also run on a host OS as an application (16:11:43) GriffenJBS: and so much info I read is plan 9 that doesn't exactly translate to inferno, and vice versa (16:12:18) mjl-: inferno development split from plan 9 development (16:12:26) mjl-: so there was some divergence (16:12:31) mjl-: some is merged back, but not all (16:12:56) mjl-: inferno originates from somewhere mid-1990's, and both have evolved since that (16:13:08) GriffenJBS: doesn't that duplicate a lot of effort? (16:13:08) mjl-: acme came to plan 9 but has been ported to inferno (c -> limbo) (16:13:24) mjl-: yes (16:13:42) mjl-: well, a lot of the time in development goes to thinking of the right ideas (16:13:47) GriffenJBS: with so few resources, it's sad to see them split lie that (16:13:49) mjl-: porting the code is less work (16:14:00) mjl-: yeah, especially for the kernels (16:14:01) GriffenJBS: *like that (16:14:09) mjl-: i mean, limbo vs c is a pretty big difference (16:14:14) mjl-: i prefer limbo for userspace things (16:14:29) mjl-: but i can understand why people do not want to ditch all the plan 9 userspace code (16:15:21) mjl-: but the kernels are so similar (16:15:43) GriffenJBS: I really dislike limbo, no fault of limbo, but after asm, c, javascript, perl and lua (and their dialogs/environments) limbo is another language I'd rather not have to add/learn (16:16:06) mjl-: :) (16:16:18) GriffenJBS: have you ever played with node.js? (16:16:21) mjl-: nopes (16:16:30) mjl-: i've seen the term dozens of times (16:16:37) mjl-: but i haven't programmed javascript (16:16:55) mjl-: well, i did way back when i started using computers and started with html+javascript :) (16:17:00) mjl-: but it's different now (16:17:25) GriffenJBS: v8 javascript engine, pulled out of the browser and coupled with non-blocking IO, very fast, rather simple (16:17:36) mjl-: anyway, i don't really like languages that don't have type checking and such for bigger projects (16:18:07) mjl-: i don't really like non-blocking i/o either in general :) (16:18:17) mjl-: i really like the threading style in inferno (and plan 9) (16:18:23) GriffenJBS: while it doesn't force type checking, it does exist, and works rather well (16:18:40) mjl-: for javascript? (16:18:47) GriffenJBS: mjl-: I agree it' easier to work with / think about, but it is much harder to get right, and usually slower (16:19:52) mjl-: for many cases i've found it not so hard to get right (16:19:57) GriffenJBS: yeah, do you really care if data is a 8bit or 16 bit int? (16:20:16) mjl-: not really (16:20:19) mjl-: why? :) (16:20:30) mjl-: i mean, why do you ask? (16:20:34) GriffenJBS: javascript removes details like that (16:20:45) mjl-: that's usually fine by me (16:20:55) mjl-: having no overflows would be good in many cases (not all) (16:20:57) GriffenJBS: and while there isn't native support for binary data, there are libs for it, that work rather well (16:21:29) mjl-: i'm mostly worried that the "compiler" (interpreter) does not warn me if i pass the wrong type or even number of parameters (16:21:46) GriffenJBS: I've started do a lot of C/C++ code in node, much faster to program, less corner cases, gotchas and suprises (16:21:54) mjl-: well, it warns me when (if ever) it reaches that point of execution (16:22:12) mjl-: do you write unit tests? (16:22:32) GriffenJBS: mjl, I used to feel that way, but in practice it happens a lot less than you would think (16:23:03) GriffenJBS: :-) lots of test :-( that usually work and make you question why you wrote them (16:23:12) mjl-: yeah :) (16:23:14) GriffenJBS: but when something breaks they are handy (16:23:45) mjl-: i've written in python and ruby, i prefer python, but i've written the bigger program in ruby (16:23:49) mjl-: a network daemon (16:23:57) mjl-: i found i spent huge amounts of time testing code (16:24:13) GriffenJBS: python is another one I'm trying to stay clear of, and php (16:24:21) mjl-: only to find out some function was being called wrong (16:24:44) GriffenJBS: oh and the debuggers for v8 are GREAT (16:24:58) mjl-: so you start writing unit tests for things that even c compilers could catch early (16:25:01) GriffenJBS: change code on the fly to see if it works right (16:25:09) mjl-: ok, a good debugger is nice to have (16:25:16) mjl-: i'm also staying away from php :) (16:25:28) mjl-: i've programmed a bit in it, and it's ugly (16:25:35) mjl-: i don't like python so much either nowadays (16:25:51) sl: if you try to pronounce php as a regular word it sounds like you are trying not to throw up. :) (16:25:52) mjl-: i don't like the classness/objectiness of it (16:26:07) GriffenJBS: lol, or on drugs, pcp!? (16:26:55) mjl-: GriffenJBS: so what attracts you to inferno? (16:27:09) GriffenJBS: the same thing that brought me to linux (16:27:36) GriffenJBS: being able to do 'cat /vmlinux > /dev/dsp' and hear the voice of God (16:27:43) mjl-: hehe (16:27:49) GriffenJBS: I can connect anything to anything (16:28:03) GriffenJBS: windows can't come glose (16:28:05) GriffenJBS: *close (16:29:08) GriffenJBS: I have a NMEA usb gps, I'm currently trying to capture and send via tcp to a server (16:29:21) GriffenJBS: but usb in hosted inferno is proving problematic (16:29:59) GriffenJBS: windows reads it as a com port but infero has problems with that as well :-/ (16:30:18) GriffenJBS: :-) I'm guess the problem isn't inferno, but rather windows (16:30:30) mjl-: hmmz, that is annoying (16:30:36) Fish [~Fish@exo3753.pck.nerim.net] entered the room. (16:30:38) mjl-: i would expect windows to detect the usb device (16:30:42) mjl-: and give it a driver (16:30:53) mjl-: gps is just nmea over serial (over usb in this case) right? (16:31:11) mjl-: so it' sjust reading & writing text, provided you have the right serial line details set up (baudrate and such) (16:31:24) GriffenJBS: well windows sees it as a com port, so I can bind #t to see it (16:31:36) GriffenJBS: but reading, writing, status, or ctrl crashes (16:32:06) mjl-: ai, nasty (16:32:16) mjl-: bugs in inferno (16:32:51) mjl-: the com5 is listed in emu/Nt/deveia.c (16:33:09) mjl-: did you try finding where exactly it crashes? (16:33:51) GriffenJBS: not sure how, I get 'panic: decref, pc=0x407c78' every time (16:34:22) mjl-: mjah, need a debugger on windows :) (16:34:25) mjl-: or some print statements (16:34:30) mjl-: i have no idea how to do the first on windows (16:35:16) GriffenJBS: wow, more info now 'alloc: D2B: 19707c0 not in pools', 'panic: alloc:D2B 19707c0 (from 431eff/524d72)' (16:36:54) GriffenJBS: strange the last two lines are from 'emu.exe -?' (16:37:34) mjl-: the usage info? (16:37:58) mjl-: corruption in the malloc pools... (16:39:35) GriffenJBS: ... I know how to catch and debug in windows ... but emu and inferno aren't compiled by a ms compiler so what debug info is available? (16:40:16) mjl-: they are compiled by the ms compiler! (16:40:20) mjl-: cl.exe (16:40:55) mjl-: see mkfiles/mkfile-Nt-386 (17:04:13) GriffenJBS: I attached windbg to it and ran it, it stopped on a ret in nt.dll!KiFastSystemCallRet (17:19:39) GriffenJBS: so now what ..!? http://i55.tinypic.com/eit7is.jpg (17:20:36) mjl-: does that come with a back trace? (17:29:02) GriffenJBS: I'm not sure, :-) first time using the windows debugger (17:29:16) GriffenJBS: so far I'm not impressed (17:43:20) robot12 [~kazzhilki@proxy10.ts.fujitsu.com] entered the room. (17:52:41) mjl-: perhaps you can lookup the pc's printed by the error messages? a debugger should be able to tell you which instruction (& function) that is (18:01:53) robot12 left the room (quit: Quit: Leaving.). (18:34:10) The account has disconnected and you are no longer in this chat. You will be automatically rejoined in the chat when the account reconnects.