Conversation with #inferno at Sun Mar 29 16:07:43 2009 on powerman-asdf@irc.freenode.net (irc)
(16:57:16) Fish left the room (quit: Remote closed the connection).
(17:35:06) npe [i=npe@naist-wavenet126-102.naist.jp] entered the room.
(18:29:09) Fish [n=Fish@ASte-Genev-Bois-151-1-42-94.w83-114.abo.wanadoo.fr] entered the room.
(19:17:12) Fish left the room (quit: Remote closed the connection).
(19:31:56) mennis [n=mennis@adsl-152-92-94.asm.bellsouth.net] entered the room.
(20:05:15) mennis left the room (quit: Client Quit).
(20:51:20) tombohannon [n=tombohan@h117.255.18.98.static.ip.windstream.net] entered the room.
(20:59:17) tombohannon left the room (quit: ).
(21:15:37) tombohannon [n=tombohan@h117.255.18.98.static.ip.windstream.net] entered the room.
(21:37:12) tombohannon left the room (quit: ).
(21:38:33) caerwyn [n=chatzill@c-76-119-5-71.hsd1.ma.comcast.net] entered the room.
(21:39:54) tombohannon [n=tombohan@h117.255.18.98.static.ip.windstream.net] entered the room.
(21:42:47) Fish [n=Fish@ASte-Genev-Bois-151-1-63-45.w82-121.abo.wanadoo.fr] entered the room.
(21:48:51) npe left the room (quit: ).
(22:03:43) tombohannon left the room (quit: ).
(22:08:21) uriel: sqweek: ndb? (wild guess)
(22:37:04) caerwyn left the room (quit: "ChatZilla 0.9.84 [Firefox 3.0.4/2008102920]").
(22:44:34) tombohannon [n=tombohan@h117.255.18.98.static.ip.windstream.net] entered the room.
(22:53:21) tombohannon left the room (quit: ).
(23:03:14) jas left the room (quit: leguin.freenode.net irc.freenode.net).
(23:10:03) eno__ [n=eno@adsl-70-137-164-249.dsl.snfc21.sbcglobal.net] entered the room.
(23:14:30) tombohannon [n=tombohan@h117.255.18.98.static.ip.windstream.net] entered the room.
(23:19:39) tombohannon left the room (quit: ).
(23:22:13) eno left the room (quit: Read error: 110 (Connection timed out)).
(23:34:53) caerwyn [n=chatzill@c-76-119-5-71.hsd1.ma.comcast.net] entered the room.
(23:39:37) tombohannon [n=tombohan@h117.255.18.98.static.ip.windstream.net] entered the room.
(23:45:03) tombohannon left the room (quit: ).
(00:06:09) tombohannon [n=tombohan@h117.255.18.98.static.ip.windstream.net] entered the room.
(00:07:27) bdheeman [n=bdheeman@122.173.27.235] entered the room.
(00:09:27) tombohannon left the room (quit: Client Quit).
(00:32:52) tombohannon [n=tombohan@h117.255.18.98.static.ip.windstream.net] entered the room.
(00:37:33) whorrel [n=whorrel@122.173.27.235] entered the room.
(00:42:31) aguent [n=andre@p4FF65AF9.dip.t-dialin.net] entered the room.
(00:45:03) whorrel left the room.
(00:46:30) Fish left the room (quit: Read error: 113 (No route to host)).
(00:47:15) bdheeman left the room.
(00:53:46) tombohannon left the room (quit: ).
(00:58:12) andguent left the room (quit: Read error: 110 (Connection timed out)).
(01:05:55) eno__ is now known as eno
(01:12:52) caerwyn_ [n=chatzill@c-76-119-5-71.hsd1.ma.comcast.net] entered the room.
(01:29:04) caerwyn left the room (quit: Read error: 110 (Connection timed out)).
(01:46:30) mjl-: uriel: do you have more info on that?
(01:46:38) mjl-: (pdf being hard to get right)
(01:48:46) mjl-: i have emu on the linksys wrt54gl btw
(01:48:58) mjl-: it was fairly easy, though not without hickup
(01:49:21) mjl-: somehow, a dynamically linked binary couldn't resolve some libc symbol...
(01:49:24) mjl-: static linking solved it
(01:49:42) mjl-: but now i'm still seeing %bd formatting is wrong
(02:45:24) newmanbe [n=btdn@138.74.131.25] entered the room.
(03:15:45) uriel: mjl-: on what?
(03:18:51) tombohannon [n=tombohan@h117.255.18.98.static.ip.windstream.net] entered the room.
(03:19:34) mjl-: 22:37    uriel: mjl-: sadly handling pdf is just a few notches below acpi in dificulty..
(03:19:34) mjl-: 22:37    uriel: (ie. pointlessly difficult)
(03:26:47) uriel: info on what?
(03:26:55) uriel: pdf or acpi?
(03:27:15) mjl-: pdf
(03:27:37) mjl-: perhaps a doc like the one on http and what's wrong with it ;)
(03:30:04) newmanbe: http://c2.com/cgi/wiki?PdfSucks
(03:35:13) mjl-: hah, big page.  will look more tomorrow
(03:35:17) mjl-: need sleep now :)
(03:35:25) mjl-: thanks
(03:35:35) newmanbe: Goodnight.
(03:39:53) anothy_x: that page says very little about why pdf, as pdf, sucks. it's mostly about why page-oriented presentation formats suck.
(03:40:25) newmanbe: *nod*
(03:44:32) tombohannon left the room (quit: ).
(03:52:49) KillerX left the room (quit: ).
(04:07:56) uriel: mjl-: actually, I'm not sure all the new crap is documented at all, it is a bunch of xml and embeded crud on pseudo-postcript
(04:08:37) uriel: doing postscript would be hard enough (I got a pretty good manual, that only covers an old version manually and runs into close to a 1000 pages)
(04:09:13) uriel: but pdf is a whole other ball game, it does even 3d, javascript scripting, form input, and this days I think even video
(04:09:26) tombohannon [n=tombohan@h117.255.18.98.static.ip.windstream.net] entered the room.
(04:09:30) uriel: (oh, and much of it with xml sprinkled IIRC)
(04:10:14) uriel: and as gdiaz pointed out, supporting a sane/old subset is not quite good enough, because new pdfs use new crap for do simple shit'
(04:27:25) newmanbe: PostScript, at least, is documented.
(05:06:31) uriel: yup, and built on a fairly solid fundation
(05:10:32) caerwyn_ left the room (quit: Read error: 104 (Connection reset by peer)).
(05:10:35) caerwyn [n=chatzill@c-76-119-5-71.hsd1.ma.comcast.net] entered the room.
(05:14:34) tombohannon left the room (quit: ).
(05:31:44) caerwyn left the room (quit: "ChatZilla 0.9.84 [Firefox 3.0.8/2009032609]").
(05:45:54) newmanbe left the room (quit: "Leaving").
(05:50:19) aguent left the room (quit: "This computer has gone to sleep").
(06:41:59) npe [i=npe@naist-wavenet126-102.naist.jp] entered the room.
(07:02:52) eno__ [n=eno@adsl-70-137-131-24.dsl.snfc21.sbcglobal.net] entered the room.
(07:15:25) eno left the room (quit: Read error: 110 (Connection timed out)).
(08:33:03) eno [n=eno@nslu2-linux/eno] entered the room.
(08:37:27) eno__ left the room (quit: Read error: 104 (Connection reset by peer)).
(09:04:15) aguent [n=andre@p4FF65AF9.dip.t-dialin.net] entered the room.
(09:18:44) aguent: omg that pdf hate page is totally ridiculous and doesn't get the point at all
(09:19:27) aguent: html is a big pile of crap for presenting text...and that for the following reason: you can't have line breaks at syllables
(09:21:36) aguent: there's a reason why people have a day job of formatting text job all day or knuth spending half of his life optimizing tex layout... 
(09:40:16) eno__ [n=eno@adsl-70-137-148-237.dsl.snfc21.sbcglobal.net] entered the room.
(09:50:50) eno left the room (quit: Read error: 110 (Connection timed out)).
(10:18:51) C-Keen: pdf is documented. And at some point in production you need page oriented layout / presentation formats. That's what pdf has been developed for
(10:33:02) gdiaz_ios [n=none@217.14.40.23] entered the room.
(10:43:41) aguent: C-Keen: +1
(10:47:27) C-Keen: mjl-: great! any caveats while building emu apart from that linking issue?
(10:50:00) mjl-: C-Keen: no, i don't think so.
(10:50:27) mjl-: well, i forgot to put spim in the mkconfig, so i ended up linking against 386 libraries at the first attempt :)
(10:50:33) mjl-: but that's my sillyness
(10:51:02) mjl-: C-Keen: i do still have problems with print-style-formatting bigs though
(10:51:21) C-Keen: mjl-: probably the byte order?
(10:51:24) C-Keen: ah no
(10:51:38) mjl-: well, the code looks byte-order independent
(10:51:53) mjl-: it just uses %'s en /'s to go from vlong to string
(10:52:15) mjl-: no direct byte manipulation
(10:52:30) mjl-: but it does look like two 32-bit ints are swapped, and then some value added.
(10:52:31) C-Keen: hm
(10:52:45) mjl-: i sent mail to the inferno list yesterday evening, but it hasn't showed up
(10:53:03) C-Keen: is a vlong really 64 bits?
(10:53:36) mjl-: it should be...
(10:53:50) mjl-: or:  i expect it to be
(10:53:51) C-Keen: I have architectures sitting here where 64 bit data types are actually less
(10:53:53) mjl-: but perhaps it isn't
(10:54:02) mjl-: hmm, how does that work out?
(10:54:13) C-Keen: you have to live with it
(10:54:16) mjl-: then you'll have no file offsets/lengths in your programs?
(10:54:19) mjl-: or only less
(10:54:22) C-Keen: less
(10:54:38) C-Keen: a lot of DSPs have this limit for speed reasons
(10:55:07) C-Keen: because the haven't been designed for that or it is not that common in the main field of applicaton
(10:55:17) mjl-: myeah, i see
(10:55:27) mjl-: but don't the compilers then emulate real 64-bits?
(10:55:34) mjl-: i thought the p9 compilers did?
(10:55:40) C-Keen: heh some do some don't
(10:55:42) mjl-: or is that an exception to the rule
(10:55:45) mjl-: ok
(10:55:52) C-Keen: most don't :)
(10:55:56) C-Keen: it is easier (tm)
(10:56:03) C-Keen: and everyone knows (tm)
(10:56:43) mjl-: ☺
(10:56:47) mjl-: i do now :P
(10:56:48) mjl-: unix: -rw-rw----  1 mjl  mjl  25902 Mar 13 16:47 CHANGES
(10:56:48) mjl-: emu-wrt: --rw-rw---- M 2 mjl mjl     111248242901092 Mar 13 16:47 CHANGES
(10:56:48) mjl-: 10r25902 is 16r652E
(10:56:49) mjl-: 10r111248242901092 is 16r652E00000064
(10:56:57) mjl-: that's what i'm seeing
(10:57:23) mjl-: directories are 10r100 bytes, or 16r64.  they should be 0
(10:58:24) mjl-: i'll put in a print for sizeof vlong
(10:58:30) mjl-: that should at least give the right result, right?
(10:59:01) mjl-: btw, when i the two int halves in limbo, it does show correctly
(10:59:04) gdiaz_ios: hello
(10:59:20) mjl-: b := big 1; print int (b>>32); print int b; (pseudo)
(10:59:21) mjl-: hi gdiaz_ios 
(10:59:43) gdiaz_ios: what are you playing with?
(10:59:53) C-Keen: mjl-: rigth
(10:59:56) mjl-: gdiaz_ios: inferno on the wrt54gl (linksys)
(11:00:00) gdiaz_ios: hum that's sounds like yoda speaking. . . 
(11:00:19) mjl-: E6BBE2B45A     2577 Mon Mar 30 02:33:29  mechiel@xs4all.nl
(11:00:19) mjl-: (host smtp.vitanuova.com[217.34.29.140] said: 450 4.7.0 deferred: smtp ping: smtp: connection refused (/net.alt/tcp!xs4all.nl) (in reply to DATA
(11:00:19) mjl-:  command))
(11:00:27) mjl-: (that's my mail not coming through)
(11:01:12) gdiaz_ios: oh, :) the emu version someone posted this weekend i guess :) nice
(11:01:19) gdiaz_ios: or a new one?
(11:02:20) mjl-: gdiaz_ios: i'm using that one, from richard miller
(11:02:23) mjl-: though i had to recompile
(11:03:46) gdiaz_ios: nice, i have a linksys but not the wrt54gl :(
(11:04:02) mjl-: if you have a really old one you might be in luck...
(11:04:27) mjl-: but the later wrt54g's were not as easy to upgrade.  if at all
(11:04:42) gdiaz_ios: btw i retried the ircfs commands of the other day and seems to work, so there was something uncontrolled here. . .it works now without -d
(11:04:58) mjl-: ah, cool
(11:58:57) gualteri [n=unknown@crespins.disca.upv.es] entered the room.
(12:05:10) bdheeman [n=bdheeman@122.173.27.235] entered the room.
(12:48:12) bdheeman left the room (quit: "ircII EPIC4-2.0 Plan9 4 -- Are we there yet?").
(13:57:20) gdiaz_ios left the room (quit: Remote closed the connection).
(13:59:16) gdiaz_ios [n=none@217.14.40.23] entered the room.
(14:12:25) tombohannon [n=tombohan@h117.255.18.98.static.ip.windstream.net] entered the room.
(14:20:04) nickgr left the room (quit: "Leaving").
(14:55:49) KillerX [n=anant@gentoo/developer/KillerX] entered the room.
(15:16:26) tombohannon left the room (quit: ).
(15:21:36) tombohannon [n=tombohan@h117.255.18.98.static.ip.windstream.net] entered the room.
(16:01:25) anothy_x: njl-: how did you actually stick it on the linksys? what firmware are you using? what storage?
(16:05:21) mjl-: anothy_x: i downloaded a firmware from openwrt.org, then used the wrt54gl firmware upgrade page on the web interface to upgrade
(16:05:42) mjl-: (then i did another upgrade to a more recent firmware, the wiki explained to do it like that)
(16:06:00) mjl-: and then you login to telnet, set a password, reboot the device, and you can ssh into it
(16:06:10) mjl-: so i scp'ed the emu, logged in over ssh, and started the emu
(16:06:14) mjl-: and that was it
(16:08:03) anothy_x: which model do you have?
(16:08:04) mjl-: the device has 4mb of rom
(16:08:07) mjl-: and 16mb ram
(16:08:11) mjl-: a wrt54gl v1.1
(16:08:33) mjl-: that's an (almost) identical to an earlier wrt54g version.  i.e. it runs linux and has more memory than later wrt54g models
(16:08:56) mjl-: (the l stands for linux)
(16:09:06) anothy_x: i have an older openwrt-based firmware (what's it called again... dd-something?), but i have some vague memory from when i last looked of not having enough space to get the writable space.
(16:09:29) mjl-: i think dd-wrt is a similar project
(16:09:40) anothy_x: yeah, i have a v2.
(16:10:01) mjl-: i think soul9 mentioned a few days ago that it isn't as easy to run your own programs on dd-wrt
(16:10:23) mjl-: and perhaps the dd-wrt was a bit larger (in firmware)?  leaving less room for other stuff
(16:11:15) anothy_x: okay, time to re-learn the current state of openwrt, i guess.
(16:12:08) anothy_x: except openwrt.org has gone unreachable. nice.
(16:12:16) mjl-: annoyingly, today seems to be the day openwrt is upgrading their servers :D
(16:13:14) mjl-: fwiw, they really are upgrading their servers according to #openwrt
(16:13:26) mjl-: it's not just random offlineness we've come to lve
(16:13:27) mjl-: love
(16:13:53) anothy_x: well, that's something, at least.
(16:16:21) C-Keen: mjl-: thanks for the ssh client
(16:18:14) soul9: yeah, it doesn't really give you the type of access openwrt does
(16:18:32) soul9: there is ssh access, but only something like 10 commands available and stuff
(16:18:45) soul9: really, it's for using the wrt as a home router and nothing more :-)
(16:19:28) anothy_x: when i did the initial firmware load on this thing (~3 years ago) i remember there being something dd-wrt had that openwrt didn't... but now i can't remember at all what it was.
(16:20:31) soul9: hum, it has a few things iirc, like multi-essid
(16:20:46) soul9: but even that's possible in openwrt, it's jsut that you need to hack it
(16:23:28) soul9: aaaah
(16:23:35) soul9: so you crosscompile hmmm
(16:24:30) anothy_x: ?
(16:24:32) soul9: that's why i failed with emu on the wrt is my guess :-)
(16:24:43) soul9: don't exactly remember how i tried though
(16:25:06) soul9: well, yes, can't put gcc on a wrt eh
(16:34:50) olegfink: I think the same youtube link that was sent to 9fans a few days ago should now appear on inferno-list.
(16:34:52) olegfink: oh, and hi.
(16:38:05) olegfink: by the way, those wrts are mips, right?
(16:38:46) anothy_x: yes
(16:38:52) C-Keen: olegfink: yes mipsel
(16:39:30) C-Keen: and with some soldering effort you will get a serial console as well
(16:42:35) C-Keen: hm, this LLVM stuff looks nice, maybe it can be persuaded to generate dis bytecode...
(16:44:15) anothy_x: i want more and more things to generate dis.
(16:44:43) anothy_x: my own summer project is going to be a pico2dis.
(16:45:33) olegfink: i've got a levelone wap-0006 which is rtl8186
(16:46:15) olegfink: maybe it would make sense running inferno on it.
(16:46:37) olegfink: C-Keen: by the way, you're the one with lemore, right?
(16:46:40) olegfink: er, lemote
(16:51:49) C-Keen: olegfink: yup
(16:55:02) aguent: if there's anyone being able to buy one of the lemote mips boxes (not the laptop ones) i would be happy to order one too
(16:56:14) C-Keen: aguent: order directly from lemote
(16:56:45) aguent: is this possible now? from what i heard last only chinese residents could buy them directly at the reseller in china
(16:57:05) C-Keen: nah, that's how I did it too, just send them an email
(16:57:21) aguent: oh that's good news :) thanks
(16:57:49) C-Keen: welcome :)
(17:19:33) olegfink: (sorry, was away for an excellent lunch)
(17:20:00) olegfink: C-Keen: there was a discussion about lemote port being a gsoc project; what's your opinion on this?
(17:24:48) C-Keen: olegfink: could be interesting but I am in no position to tell the amount of effort needed
(17:28:19) mennis [n=mennis@adsl-152-92-94.asm.bellsouth.net] entered the room.
(17:29:57) mennis left the room (quit: Client Quit).
(17:46:01) eno__ is now known as eno
(18:12:30) anothy_x: (not that anyone asked ;-) but...) i'm against. another oddball device with uncertain future and unclear availability?
(18:13:11) anothy_x: at a minimum, the student would have to already have one. that rather cuts your population.
(18:13:49) mennis [n=mennis@adsl-068-016-104-079.sip.asm.bellsouth.net] entered the room.
(18:14:19) soul9: heh, like the olpc port?
(18:14:23) soul9: does that work at all?
(18:14:55) soul9: anyone using that not counting anothy? ;-)
(18:15:25) anothy_x: anyone using that *counting* anothy? ;-)
(18:15:48) soul9: heh
(18:15:55) anothy_x: i'd love to, but it was sort of half-finished.
(18:16:05) soul9: i see
(18:32:33) eno: hi guys, i've just added inferno-minimal and inferno-devel ipk to optware arm/x86 feeds
(18:32:57) eno: so it (emu-g) runs on a lot of arm NAS devices
(18:33:47) eno: nslu2, linkstation, qnap, synology, wdc mybook world, ...
(18:33:57) eno: i even tested it on a sheevaplug
(18:34:29) eno: actually all debian arm lenny devices can use the optware/cs08q1armel ipkg feed
(18:35:03) eno: http://ipkg.nslu2-linux.org/feeds/optware/
(18:35:21) eno: chances are, if it's arm glibc linux, it runs emu-g just fine
(19:48:38) gdiaz_ios: off to home :)
(19:55:56) C-Keen: anothy_x: I guess at least the processor family will survive so that won't be lost effort
(19:56:21) C-Keen: anothy_x: but I agree there are more 'urgent' projects
(19:56:40) stu8ball: Processor family of what? XO or Lemote?
(19:56:55) olegfink: eno: cool.
(19:57:02) stu8ball: I can vaguely recall something about AMD Geodes
(19:57:20) olegfink: geode is x86.
(19:57:22) C-Keen: stu8ball: lemote
(19:57:33) stu8ball: ah
(19:57:36) olegfink: lemotes (loongsons) are mips64
(19:57:58) stu8ball: olegfink, I know, with Geode I was referring to the XO.
(19:58:33) stu8ball: And I  can vaguely recall something hearing somethign bad about their future, althugh I've probably confused myself in some way about that.
(19:58:37) C-Keen: geodes are also found in other stuff
(19:58:40) stu8ball: yeah
(19:58:57) C-Keen: soekris boxes for example :)
(19:58:58) olegfink: linutop for one is pretty popular it seems.
(19:59:05) olegfink: yep, and alix
(20:01:21) ***olegfink just needs an ibm-quality arm or mips laptop
(20:02:10) C-Keen: you won't even get an ibm quality x86 laptop these days
(20:02:24) eno: how about a combination of x86 laptop + arm/mips NAS?
(20:03:04) olegfink: too hot on one part and too heavy/non-portable on the other.
(20:03:58) olegfink: speaking of which, has anyone seen those two-headed dells?
(20:04:19) C-Keen: two headed?
(20:05:39) olegfink: http://www.siftwire.com/dell-latitude-e4200-and-e4300-hybrid-notebook.html
(20:07:06) C-Keen: so is this more than a wired-per-button boot manager selection?
(20:08:18) olegfink: all I've seen are speculations, but it seems that those two systems run in parallel
(20:09:42) stu8ball: Surely running both at the one time would take away from the potential power savings of just booting straight into the ARM?
(20:10:20) olegfink: to my understanding, they were talking about hybernating the c2d
(20:10:24) stu8ball: Interesting enough concept anyway.
(20:10:31) stu8ball: I wonder if it physically contains a KVM switch.
(20:16:08) anothy_x: in terms of production quality, the thinkpads are about as good as they've ever been.
(20:16:17) anothy_x: and i also recall hearing that the geodes were end-of-lifed.
(20:16:49) anothy_x: sony's production values use to be almost as good as IBMs. apple's are on par.
(20:17:31) olegfink: I'm thinking of getting a used vaio c1. very nice little machine, but again 386.
(20:17:42) anothy_x: OLPC is "considering" making the next-gen XO (assuming such a thing ever happens) run on arm (which is what they should've done from the start)
(20:18:00) aguent: ooh thinkpad mips..now that would be awesome...
(20:19:00) anothy_x: my hope is that the most lasting effect of this netbook fad is that people start seeing that windows isn't needed on *every* device.
(20:19:20) anothy_x: i know totally non-technical with linux netbooks. they don't particularly care that it's linux, of course; it just does its job.
(20:19:51) aguent: people do. certainly not everyone..but i've seen plenty of only windows users switch to linux these day for the eeepc. 
(20:20:06) C-Keen: of course, because they don't buy computers because of operating systems
(20:20:26) anothy_x: right. but for so long the "industry wisdom" was that if the thing didn't run windows, nobody would buy it.
(20:20:27) C-Keen: the want certain tools
(20:20:44) anothy_x: today, "certain tools" more and more means "a web browser".
(20:20:49) C-Keen: now the point is that people think they can use only tools than need windows
(20:21:11) olegfink: i'm not that optimistic
(20:21:39) olegfink: s/now/2 years time/, s/windows/linux/
(20:21:42) olegfink: that's all
(20:21:44) C-Keen: oh, I am. One other killer is that in a lot of fields the only acceptable tool available runs on windows only
(20:22:01) anothy_x: olegfink: maybe. but even that would be an improvement.
(20:22:46) anothy_x: linux sucks less than windows, but the more important point is that it's easier to stick p9p and 9vx on it and pretend it isn't really linux at all.
(20:22:55) C-Keen: hehe
(20:23:36) olegfink: indeed (especially the 'sucks less' part)
(20:24:00) olegfink: I've just finished yet another iteration of trying to convince myself NT actually sucks less
(20:24:15) aguent: that won't work
(20:24:21) aguent: it's metaphysically impossible
(20:24:30) anothy_x: less than what?
(20:25:02) aguent: certainly it can't suck less than it does suck... and apart from that...
(20:25:35) olegfink: less than linux
(20:25:46) olegfink: actually, I still think that holds true for NT
(20:25:59) aguent: 4.0?
(20:26:00) anothy_x: oh. erm, no. you're wrong. sorry. ;-)
(20:26:01) olegfink: but not for the win32 userland built on top of it.
(20:26:21) olegfink: aguent: every selling windows version nowadays is based on NT
(20:26:34) anothy_x: oh, you mean just the kernel?
(20:26:40) aguent: olegfink: yes. sorry. 
(20:27:14) anothy_x: theoretically possible, i guess. but unlikely. linux started from an image of unix and then botched it, which is likely better than the second-system effect NT got from copying VMS.
(20:27:41) anothy_x: fun game: y'know HAL, from the movie 2001?
(20:27:42) C-Keen: and os/2
(20:27:58) anothy_x: (C-Keen: true)
(20:28:01) aguent: book>film @ 2001
(20:28:03) olegfink: hm, I thought (apart from being a successor to os/2) that nt was made more or less from scratch
(20:28:27) stu8ball: Linux as just a kernel supports far more stuff than NT as just a kernel does, is the sheer amount of crap you support means anything. Like l33t virtualisation stuff, Linux wins there.
(20:28:28) olegfink: nt shares as much with os/2 as it does with 9x, afaik
(20:28:29) anothy_x: there is/was a rumor that "HAL" was derived by removing 1 from each letter in IBM.
(20:28:40) olegfink: ...and WNT is VMS+1
(20:28:41) stu8ball: Heuristic ALgorith,
(20:29:17) stu8ball: I read the book long ago, and have been meaning to read at least a couple of its sequels.
(20:29:22) stu8ball: Haven't done so yet. :(
(20:29:26) aguent: anothy_x: afaik kubrik spent some time in the ibm labs while producing the film
(20:29:45) aguent: stu8ball: 2010 is okay...forget about the rest
(20:29:45) anothy_x: MS claims NT was made from scratch (except for the file system, which i think even they admit was a fork). it sees unlikely, though.
(20:30:16) olegfink: at least both os/2 and win32 applications run in an emulated environment on NT
(20:30:21) olegfink: (and dos, and posix)
(20:30:49) anothy_x: that doesn't say much, though.
(20:31:35) C-Keen: well in terms of handling network and filesystems I consider NT as botched as linux
(20:31:44) olegfink: if they were copying something they probably wouldn't abstract from the api/abi they've been copying?
(20:32:39) anothy_x: who knows what's going on underneath? all the "emulation" tells you is that the api/abi changed. the underlying bit could be 90% the same.
(20:32:48) anothy_x: (or 2% the same. we have no idea)
(20:33:43) olegfink: (actually, nt4 sources are (illegally) available and win2003 sources take part in an academic program)
(20:34:23) anothy_x: okay, theoretically, sure. but i've certainly not read them, and i'm betting for the "we" in this room, my statement holds.
(20:35:17) ***aguent still wonders why the nt code never leaked completely
(20:35:37) aguent: should be pretty interesting for a whole bunch of people
(20:36:07) olegfink: aguent: it did, for nt4
(20:36:07) anothy_x: it'd also contaminate a whole bunch of things.
(20:36:32) anothy_x: when the leak happened, the samba guys quickly, loudly, and publicly made clear that anyone who read it wouldn't be working on samba.
(20:36:38) aguent: olegfink: oh i thought it was just a bit of the network subsystem but not everythink
(20:36:54) aguent: heh
(20:36:56) olegfink: no, afaik it even includes the whole userspace
(20:37:14) olegfink: anothy_x: yeah, it's a direct consequence of the current copyright law I think.
(20:37:28) olegfink: that's why I have no intention to read it either.
(20:38:14) olegfink: but well, eh, linux sucks less as a whole.
(20:38:16) aguent: i didnt say that i care much either..still a surprising thing given there's so much opportunity for a wide variety of people
(20:38:24) olegfink: at least its video driver doesn't hang the kernel/
(20:38:35) anothy_x: it'd also contaminate your brain, which is why i have no intention of reading it.
(20:38:48) C-Keen: olegfink: you haven't used a nvidia kernel module for a long time eh
(20:39:03) olegfink: I never had an nvidia gpu
(20:39:42) C-Keen: the situation gets the same as on windows. If there are enough people working on it that are no good at programming the system will suffer
(20:40:04) aguent: also the development model at M$ is disastrous
(20:40:20) aguent: there are some blogs from ex-ms-employees describing their experiences
(20:41:12) aguent: it's very hierachical and when you want to commit even a tiny user space detail it amounts to about 50 people reviewing each with veto rights
(20:41:34) C-Keen: well enforced code review is not bad at all per se
(20:41:38) olegfink: aguent: by the way, on thinkpads topic, I have a crazy dream of getting a butterfly's case and putting some arm/mips board and at least 1024x768 10" flat panel inside.
(20:42:14) aguent: C-Keen: it were mostly management type of guys...so not code review but actually feature review
(20:42:29) stu8ball: would kill for one of those fold-out keyvoards they used to have
(20:42:51) aguent: olegfink: maybe you would be able to use a Belco Alpha-400 board 
(20:44:02) olegfink: there's also beagleboard, but both don't have a fpga, and the design would need quite a lot of peripherals like ps/2 for keyboard/trackpoint, lvds for the panel etc.
(20:44:17) anothy_x: when you've got as many people as they do using their stuff, it's entirely sensible to be very conservative about changes.
(20:45:05) aguent: anothy_x: agreed, still the progress in the windows version is laughable at best. 
(20:45:46) aguent: technically even windows 7 is not so much advanced im comparison to let's say xp or even 4.0
(20:46:27) aguent: also internet explorer is still the same disguisting piece of crap
(20:46:28) uriel: that is the reason plan9 and inferno are so 'conservative' about releasing changes, wouldn't want to upset the millions of users out there..
(20:46:38) aguent: heh
(20:54:49) gualteri left the room (quit: "leaving").
(21:19:52) coriolan [n=wolf@a88-114-221-198.elisa-laajakaista.fi] entered the room.
(22:04:15) mjl-: meehee
(22:04:18) mjl-: http://code.google.com/p/inferno-os/source/detail?r=393
(22:04:22) mjl-: anyone seen that yet? :)
(22:05:18) mjl-: code duplication strikes again :P
(22:08:27) anothy_x: anyone know what "forse" is?
(22:08:41) mjl-: not me
(22:09:25) mjl-: the ssh seems to be a server too
(22:14:52) mjl-: eno: having emu on many arm nas devices is pretty good
(22:14:57) mjl-: now i only have to pick & buy one :)
(22:15:48) tombohannon left the room (quit: leguin.freenode.net irc.freenode.net).
(22:15:48) C-Keen left the room (quit: leguin.freenode.net irc.freenode.net).
(22:15:48) te left the room (quit: leguin.freenode.net irc.freenode.net).
(22:15:48) maht left the room (quit: leguin.freenode.net irc.freenode.net).
(22:16:15) tombohannon [n=tombohan@h117.255.18.98.static.ip.windstream.net] entered the room.
(22:16:15) maht [n=maht__@85.189.31.174.proweb.managedbroadband.co.uk] entered the room.
(22:16:15) te [i=tao@gateway/shell/blinkenshell.org/x-6c35d7f46b3ae2df] entered the room.
(22:16:15) C-Keen [i=ckeen@pestilenz.org] entered the room.
(22:25:52) npe left the room (quit: ).
(22:31:31) mjl-: heh
(22:31:35) mjl-: fmtinstall in limbo code?
(22:31:39) mjl-: # fmtinstall('B', mpfmt);
(22:31:42) mjl-: in sshserver.b
(22:31:49) mjl-: commented out, but still
(22:32:18) mjl-: i'm really wondering what kind of code there is :)
(22:35:16) mjl-: i'm also expecting some changes to the crypto libs.  keyring
(22:35:28) mjl-: since it was commented out by accident some time ago
(22:36:40) mjl-: the server says it's ssh-1.5.  i think that's ssh v1
(22:36:51) mjl-: v2 is 1.99 & >=2.0
(22:50:52) eno: mjl-: go SheevaPlug
(22:51:15) eno: http://www.marvell.com/products/embedded_processors/developer/kirkwood/sheevaplug.jsp
(22:51:54) eno: i also suggested in http://permalink.gmane.org/gmane.os.inferno.general/4399
(22:52:20) eno: to coordinate the mips/powerpc and/or uclibc cross compilation effort
(22:52:26) mjl-: i'm reading about them now
(22:52:30) mjl-: they look pretty fine
(22:52:44) mjl-: and there seem to be docs :)
(22:52:56) mjl-: not sure if they specify all of the hardware...
(22:53:45) eno: there's one linkstation recently announced, it's based on the same board
(22:53:46) mjl-: eno: ah yes
(22:54:04) eno: qnap 119/219 as well
(22:54:07) mjl-: that's a buffalo product google tells em?
(22:54:23) eno: these are with internal SATA
(22:54:57) eno: if you like VGA, there's also openrd-base and openrd-client
(22:55:57) eno: yeah, buffalo linkstaion pro LS-XHL
(22:56:08) eno: http://www.smallnetbuilder.com/index.php?option=com_content&task=view&id=30760&Itemid=75
(22:56:41) eno: openrd: http://www.marvell.com/products/embedded_processors/developer/kirkwood/openrd.jsp
(23:01:43) mjl-: i wonder why that buffalo storage product needs as >=1ghz processor
(23:03:15) stu8ball: DRMing your filez
(23:03:34) stu8ball: Well, probably not
(23:04:41) eno: economy of scale
(23:14:54) mjl-: eno: btw, what did you have in mind when you said the think about changes going upstream?
(23:15:17) mjl-: that we provide recipies to build an emu, and then make an ipkg, for use on those mipsel machines?
(23:15:55) eno: i don't see mkfiles/mkfile-Linux-mips for example
(23:16:48) mjl-: no, those aren't in inferno-os
(23:16:55) mjl-: perhaps they can be made to
(23:17:04) mjl-: they are at the link given by rmiller
(23:17:12) eno: yeah, that's what i mean
(23:17:34) mjl-: yeah ok
(23:18:27) mjl-: not sure how vitanuova integrates such stuff into inferno-os.  with dependencies on e.g. openwrt sdk's
(23:20:09) anothy_olpc [n=anothy@adsl-99-155-111-119.dsl.bcvloh.sbcglobal.net] entered the room.
(23:22:03) eno: they can probably add basic support, and leave the toolchain specific things to cross compilation recipes
(23:24:19) mjl-: yeah, i guess adding rmiller's wrt54g-src.tar.gz would go a long way
(23:27:44) coriolan left the room (quit: Remote closed the connection).
(23:28:35) mjl-: eno: do you have a sheevaplug?
(23:29:13) aguent left the room (quit: "This computer has gone to sleep").
(23:29:18) eno: i got one recently
(23:29:32) mjl-: and, are you liking it?
(23:29:43) mjl-: it seems i can get one (incl. shipping) for 100 eur
(23:29:51) mjl-: but i don't want US power sockets i guess
(23:30:00) eno: yeah, it's very nice.
(23:30:29) eno: i only tried the tftp/nfs boot with the rootfs in dev kit
(23:31:01) eno: there might still be some problem regarding SDIO
(23:31:16) eno: official linux kernel support starts from 2.6.30
(23:32:12) mjl-: pretty good
(23:32:26) eno: http://www.cyrius.com/debian/kirkwood/sheevaplug/gallery.html
(23:32:33) mjl-: did you happen to live in a place with the right power sockets?
(23:32:39) mjl-: i guess a conversion thing shouldn't be a problem
(23:32:47) mjl-: would make it twice the size though probably :)
(23:32:48) eno: you can easily remove the US power plug
(23:33:28) eno: "http://www.cyrius.com/debian/kirkwood/sheevaplug/images/img_0007.jpg
(23:34:29) mjl-: ah, perfect
(23:34:30) mjl-: thanks
(23:35:47) eno: is wrt54g-src.tar.gz only on /n/sources ?
(23:36:08) mjl-: i think so
(23:36:15) mjl-: is sources unreachable?
(23:36:18) mjl-: is can put up a copy
(23:36:51) mjl-: ah, it works
(23:36:54) eno: i'm at a location that don't have easy access to /n/sources
(23:37:08) mjl-: no web interface either?
(23:37:09) mjl-: http://plan9.bell-labs.com/sources/contrib/miller/inferno/linksys/
(23:37:35) eno: excellent, thank you
(23:38:34) eno: i might integrate into the optware build recipe
(23:50:53) mjl-: sounds good
(23:50:58) mjl-: did you see rmillers update a moment ago?
(23:56:54) eno: just saw the email, i actually got the updated tar.gz already
(00:07:47) mjl-: a sheevaplug would be perfect for a plan 9 cpu server...
(00:08:05) mjl-: i can't believe they are this cheap :)
(00:43:57) hotaru2k3 [n=hotaru@cpe-24-29-193-226.neo.res.rr.com] entered the room.
(00:56:39) mjl-: ok, two sheevaplugs on their way here
(00:56:44) mjl-: (only one for me)
(00:59:10) ooooo [i=none@201.80.224.34] entered the room.
(01:19:21) mennis left the room (quit: Client Quit).
(01:21:30) caerwyn [n=chatzill@c-76-119-5-71.hsd1.ma.comcast.net] entered the room.
(03:40:38) hotaru [n=hotaru@cpe-24-29-193-226.neo.res.rr.com] entered the room.
(03:41:17) hotaru2k3_ [n=hotaru@cpe-24-29-193-226.neo.res.rr.com] entered the room.
(03:41:36) hotaru2k3 left the room (quit: "ChatZilla 0.9.84 [Firefox 3.1b3/20090305152042]").
(03:41:43) hotaru left the room (quit: Client Quit).
(03:41:48) hotaru2k3_ left the room (quit: Client Quit).
(03:42:40) hotaru2k3 [n=hotaru@cpe-24-29-193-226.neo.res.rr.com] entered the room.
(03:56:27) npe [i=npe@naist-wavenet126-102.naist.jp] entered the room.
(04:03:33) hotaru2k3 left the room (quit: "Leaving").
(04:03:52) hotaru2k3 [n=hotaru@cpe-24-29-193-226.neo.res.rr.com] entered the room.
(04:58:54) ajray [n=ajray@Brag-16-244.rh.ncsu.edu] entered the room.
(05:10:41) caerwyn left the room (quit: Read error: 104 (Connection reset by peer)).
(05:11:40) caerwyn [n=chatzill@c-76-119-5-71.hsd1.ma.comcast.net] entered the room.
(05:16:53) tombohannon left the room (quit: ).
(05:19:17) mennis [n=mennis@adsl-152-92-94.asm.bellsouth.net] entered the room.