Conversation with #inferno at Thu Nov 5 10:21:58 2009 on powerman-asdf@irc.freenode.net (irc) (11:05:52) npe [n=npe@195.207.5.2] entered the room. (13:47:20) eekee: good day mjl- good day everyone (15:44:03) anothy_x left the room (quit: Read error: 60 (Operation timed out)). (15:46:37) sswam1 [n=sswam@CPE-121-214-10-154.lnse3.win.bigpond.net.au] entered the room. (15:47:50) sswam left the room (quit: Read error: 60 (Operation timed out)). (16:09:49) raiz [n=raiz@p4FE88E71.dip0.t-ipconnect.de] entered the room. (17:07:26) mennis [n=mennis@adsl-068-016-104-079.sip.asm.bellsouth.net] entered the room. (17:26:31) anothy_x [n=a@adsl-99-155-110-247.dsl.bcvloh.sbcglobal.net] entered the room. (17:46:24) eno left the room (quit: Read error: 60 (Operation timed out)). (17:48:28) eno [n=eno@nslu2-linux/eno] entered the room. (18:43:39) mennis: Is there an obvious reason there is no boolean or single bit type? (18:43:49) mennis: (in limbo) (18:44:31) mjl-: not that i know (18:44:52) mjl-: i guess the typical reason is that machines don't have one bit units (18:45:00) mjl-: so you can easily use an int (18:45:06) mjl-: or byte may have been better (18:45:13) mennis: Fair enough. (18:45:49) mennis: just used to having a bool around. I'll take byte though. (18:46:02) mjl-: myeah, with a byte you'll have to cast to int i think? (18:46:05) mjl-: if(byte 1) ... (18:46:10) mjl-: i don't think that's allowed (18:46:25) mjl-: that's another reason for fewer types in limbo i think: you have to cast them explicitly (18:48:08) andguent: mennis: what do you think what compilers/interpreters do when you give them bool variables? (18:48:19) andguent: it's the exception they only really use one bit (18:48:39) andguent: e.g. in c structs. but that is really the exception than a rule (18:49:00) mennis: mjl-: Oh thanks that would have been irritating to discover. (18:49:24) andguent: in fact it often makes sense to have bool values the length of machine words (18:50:48) mjl-: andguent: yeah, arrays of bools could be done efficiently with real 1-bit language units. (18:51:07) mjl-: but might as well implement that in a library (unless you're really going to use your language like that a lot) (18:52:45) andguent: mjl-: on x86 (probably others too) you can do evil stuff with the help of the carry bit to facilitate bit array manipulations (18:53:25) andguent: like when serialiasing to a bitstream or vice versa (18:53:48) andguent: serialising? (18:53:52) andguent: serializing? (18:53:55) andguent: whatever... duh (18:57:15) mjl-: andguent: so i bet you can use that from a library too (18:57:31) andguent: mjl-: but that doesnt make sense in inferno then ;) (18:59:22) andguent: hmmm (18:59:25) andguent: what about gpus? (18:59:32) andguent: they are good at swizzling values (18:59:49) andguent: maybe that is interesting for certain bit array operations (18:59:56) andguent: but that is only a really vague thought (18:59:58) andguent: :D (19:00:17) mjl-: andguent: half true (19:00:22) mjl-: andguent: you can implement library in c (19:00:27) mjl-: and use them from limbo (19:00:44) npe left the room (quit: ). (19:00:48) andguent: mjl-: sure. just saying you loose the benefit from the carry tricks. i dont think that pays off (19:01:37) andguent: mjl-: because of all the calling overhead and stuff. the carry stuff really only pays off if you algorithm can just do the carry operation inside itself when you know what i mean (19:02:10) andguent: mjl-: the benefit here is you have a decission which bit to put..so you already get the branch for free. then you only set the carry in the appropriate way. shift and done (19:02:17) andguent: bypassing the stack and everything (19:02:21) andguent: 2 instructions. that's it (19:04:32) mjl-: ok (19:04:59) mjl-: there's still an option for such tricks when you implement operations that do more than just set 1 bit (19:05:09) mjl-: but i won't be doing it any time soon! :) (19:05:10) andguent: yeah (19:05:19) andguent: it all depends on the data structures you are working with (19:05:20) andguent: ;) (19:12:13) mjl-: so true! :) (19:12:51) mjl-: well, i must say, i'm using novt quite happily (19:12:53) mjl-: with ssh (19:13:01) mjl-: it's dog slow, that'll teach me (19:13:04) mjl-: but still (19:13:25) mjl-: i think i have to give native inferno on my laptop another shot. getting the vga working that is. (19:13:44) mjl-: i'm inspired by blstuarts remark that he ended up using some graphics card initialisation code from another project (19:13:58) mjl-: i know of svgalib & directfb that may have that code (19:14:04) mjl-: x11 driver was kind of big (19:14:20) mjl-: linux kernel may have framebuffer code in it for popular video cards? (19:14:23) andguent: okay people are killing me for that (19:14:24) andguent: but (19:14:32) mjl-: andguent: do you now more projects? (19:14:35) andguent: for framebuffer code: take a look at the haiku source (19:14:45) mjl-: ah, ok! (19:14:45) mjl-: thanks (19:14:47) andguent: they have the best vesa support i have ever seen (19:14:51) mjl-: andguent: what are people killing you fo? (19:14:52) andguent: and a lot of other cards too (19:14:55) mjl-: well, i don't want vesa :) (19:14:55) mjl-: ok (19:15:19) andguent: hm ...well haiku is quite contrary to a lot of p9 principles (19:15:31) andguent: and i know at least uriel would bash me for that :D (19:15:40) mjl-: hah, but i bet their graphics init code doesn't show too much of that (19:15:47) mjl-: haha, what are we now, a sect? (19:15:53) mjl-: that you cannot be interested in anything else :) (19:15:54) andguent: yes (19:16:01) andguent: didnt you follow that? :D (19:16:06) mjl-: darn :P (19:16:13) mjl-: i'm in knee deep already :P (19:16:16) mjl-: haha (19:16:25) andguent: let me see if i find that again (19:16:28) mjl-: just now that i've mentioned novt is working fine, it crashes :P (19:16:32) andguent: i am a bit sleepy. and my inet sucks for some reason (19:17:36) andguent: http://dev.haiku-os.org/browser/haiku/trunk/src/add-ons/accelerants (19:18:10) andguent: they are c++ files but iirc most of them are pretty much ansi c apart from the interface (19:19:17) mjl-: thanks, looks readable at first glance (19:19:20) mjl-: will look into (19:23:45) andguent: the good thing is that they only do basic framebuffer stuff so it's not too much code and it is actively developed (19:25:19) andguent: mjl-: for which card / laptop are you doing this btw.? (19:27:08) mjl-: ibm x40, intel 845gm (19:27:19) mjl-: which is in the i810, i830 family (19:27:25) mjl-: i don't know exactly which (19:27:34) mjl-: i know x.org has a "unified" intel driver (19:28:12) mjl-: which has lots of other stuff (funtionality, and related but different video card support), so is a bit annoying to read (19:32:04) andguent: hm i fear that stuff is not supported by the intel driver (19:32:24) andguent: " Support for i915 chipset and up based on the X driver," (19:32:44) andguent: what is your problem with vesa? (19:32:55) mjl-: ah, i had read over the "and up" part... (19:33:05) mjl-: doesn't vesa involve real-mode stuff? (19:33:11) mjl-: or other low-level arch hacks? (19:33:25) mjl-: i strongly prefer something with plain & simple registers (19:33:31) andguent: oh duh. i dont know (19:33:31) mjl-: so i can understand what's going on (19:33:37) andguent: you could have a look at the source (19:33:42) andguent: it's ridiculous small (19:33:47) mjl-: of course, if someone has vesa support for inferno i wouldn't hesitate to use it (19:34:02) andguent: and the vesa support is damned good in haiku (19:34:11) andguent: i had running the vesa thing on many machines (19:34:20) andguent: it works very well. and is pretty performant (19:34:42) andguent: i mean haiku has some other graphic requirements than plan9. but it just ran like butter with nearly no cpu impact (19:35:06) mjl-: i'm using vesa on plan 9 on the same machine too, works fine (19:35:37) andguent: oh then it probably makes more sense to port the plan9 one over? (19:36:47) mjl-: perhaps, but i would rather avoid the vesa complexity (19:36:54) mjl-: but perhaps it's not as bad as i have in mind (19:40:08) maht: BLS used a framebuffer on inferno in 640x400 mode :> (19:40:25) maht: or was it 320x280x4 , see iwp9 2009 papers (19:40:32) mjl-: yeah, i heard :) (19:40:34) mjl-: i want something bigger (19:40:38) mjl-: 1024x768 is the screen (19:40:45) mjl-: and i want at least 16bit (19:40:56) mjl-: i just need the magic register writes!! (19:41:05) andguent: i am getting an iMac in a few days. there's no chance i get p9 running on it? :D (19:41:13) maht: when I was a lad it was all memory mapped and you just wrote to a big array (19:41:55) mjl-: maht: isn't that still the case? :) (19:41:56) mjl-: oh! (19:42:02) mjl-: you're still a lad of course! (19:42:43) andguent: btw. once again i throw in that there is someone in the need of the powervr sgx register spec. any help would be much appreciated. discreet communication assured (19:45:49) maht: hehe tbh I have no idea how gfx is done in vga (19:54:34) eekee: andguent: not if it's PPC? :D (19:55:42) andguent: eekee: i am afraid that has nothing to do with ppc. or i am missing something (19:56:06) eekee: andguent: plan 9 port to ppc macs is not usable (19:56:22) eekee: oh.. yeah I'm on slightly the wrong track here ^^; (19:56:28) andguent: ah that direction you are shooting (19:56:33) eekee: ya (19:56:35) andguent: i thought you confused powervr somehow with ibm power (19:56:47) eekee: oh *tilts head* (19:57:04) andguent: hm no it's an intel imac (19:57:29) ***mennis thinks funtionality should be a word meaning something that sort of works. (19:57:37) andguent: without jtag or a serial line i really dont feel like debugging something like p9 on a mac (19:58:15) eekee: mennis: indeed (19:59:29) eekee: andguent: usb & firewire host controllers can be pretty standard. You could get a console on either. (19:59:48) eekee: I think... (20:00:03) andguent: eekee: well let's put it this way then: i am not up to that challenge (20:00:23) andguent: i am just a random stupid guy that knows nothing about computers. i am doomed to fail on such a project (20:00:31) eekee: k :) (20:00:46) eekee: I'm just a random stupd guy who thinks he knows a few things (20:18:51) mjl-: blegh, the annoying thing about reading other drivers is that they use the other OS'es entire infrastructure, calling functions that you don't know how they work etc. (20:19:01) mjl-: the directfb & linux fb code seem somewhat readable (20:19:04) mjl-: they are based on the xorg one (20:19:33) mjl-: but looking at them made me remember that i understood how it worked, just couldn't get the right registers set & memory accessed to display something... (20:19:42) mjl-: one bit wrong -> no video... (20:19:43) eekee: yeah -- oh that's neat. Linux drivers too often aren't :) (20:19:52) eekee: ya :/ (20:20:49) mjl-: it doesn't help i have to use the apg gart, for address translation. one more thing that can go wrong (20:20:59) mjl-: well, time to bake some pancakes now! (20:23:36) anothy_x: at the gsoc summit, a group of people got together for essentially a non-linux meeting. (20:24:17) anothy_x: the discussion was good, but the main practical upshot was trying to get a common driver framework. (20:25:00) anothy_x: i've gotten at least theoretical agreement that the group wants to solve for the non-unix case as well as the non-linux case. (20:25:44) anothy_x: it'd be nice to be able to drop in drivers from some common system, although i'm a bit skeptical of that for us. (20:26:11) anothy_x: but even so: having a defined framework will make reading drivers written to it much easier. (20:28:20) andguent: it had to be ken c... (21:21:26) j123m [n=quassel@unas-revda.interra.ru] entered the room. (21:24:13) j123m left the room (quit: Remote closed the connection). (21:24:31) j123m [n=quassel@unas-revda.interra.ru] entered the room. (21:25:00) j123m left the room ("http://quassel-irc.org - Разговаривать удобно. Везде."). (21:28:18) j123m [n=quassel@unas-revda.interra.ru] entered the room. (21:28:59) j123m left the room (quit: Client Quit). (21:29:14) j123m [n=quassel@unas-revda.interra.ru] entered the room. (21:31:38) eekee left the room (quit: "FWOOMP"). (22:02:45) mjl-: anothy_x: i guess for some subset of devices you can write common drivers (22:03:07) mjl-: but i bet for the newer stuff it won't always match (22:03:56) mjl-: just quality of code helps too of course :) (22:04:05) mjl-: such as reasonable names for constants (22:06:03) mjl-: wow, what's with the whining about "fatelf" being ignored (22:06:20) mjl-: i saw it on hackernews/reddit, now slashdot (22:06:51) mjl-: as if the idea is obviously great and everyone has a constitutional right to get loads of attention for their new pet idea (22:30:05) sswam1 left the room (quit: "leaving"). (22:32:26) mennis: I hope they are going to use some smart compression ala Franz's slim binary paper. (22:34:19) mjl-: do you have a link for that? (22:34:31) mjl-: i think one of the reasons people didn't like it is that binaries just become huge (22:34:41) mjl-: plus the benefit isn't that big on open systems (22:34:52) mjl-: where you have package tools that take care of all that (22:35:00) mennis: mmm it's on acm if you have an account. (22:35:13) mjl-: nopes unfortunately... ☹ (22:35:20) mjl-: well, i'll read it in a few decades :P (22:35:51) mjl-: i don't even dare carry a linux/bsd binary (with shared libs especially) between two machines of the same architecture (22:36:39) mjl-: a month or so ago i very briefly looked at tools to compress binaries (unpacking before running), to get a small emu (22:36:55) mjl-: but didn't find something obviously easy & good quick enough (22:37:01) mjl-: plus stripped emu isn't all that huge.. (22:37:24) mjl-: mennis: what was the smartness in the compression? (22:37:26) mennis: His method wasn't by any means simple but it was intelligent. (22:38:28) andguent: mjl-: do you pay for your acm account? (22:39:04) ***mennis does. (22:39:07) mjl-: hah, that would be a lousy deal! paying and not getting the access :P (22:39:08) mjl-: ;0 (22:39:29) andguent: acm is a big cancer (22:39:34) andguent: i hate it. wish a passion (22:39:34) mjl-: :D (22:39:44) andguent: s/wish/with/ (22:40:34) andguent: i mean seriously. it hurts us all (22:40:36) mennis: In short lzw like. But it was done on the source in an intelligent fashion, not on symbol streams. (22:41:04) mjl-: mennis: source as instructions? (22:41:37) mjl-: at least that's what binary executable diff'ers do (22:42:12) mjl-: e.g. when instructions are the same, but offsets change. the shift of the offset is encoded effiently then (22:43:29) mjl-: but on binary compression: i was somewhat impressed by μtorrent, small binary, good functionality, no bloat. (22:44:15) mennis: No it was more like figure out what kind of "operations" are being done then create a translation table for those operations and then have one "binary" which follows different paths based on architecture. I'm a bit fuzzy but here's how I remember it: (22:46:32) j123m left the room (quit: Read error: 104 (Connection reset by peer)). (22:47:41) mennis: If there was a meaning to a text in literature that could be perfectly expressed in each natural language then the combination of the meaning and the parts of each language used to express those parts would be the equivalent to slim binaries. But it built in reverse. (22:48:02) mennis: Is that retarded? (22:48:09) mennis: I think I butchered it. (22:49:01) mjl-: i think i see where that is going (22:49:03) mjl-: sounds complicated :) (22:49:30) mjl-: does it operate on the source code? (22:49:54) mjl-: or some intermediary format, between source & machine code (22:50:43) mjl-: seems hard to extract meaning from machine code, especially if that's to be transformed again to code for other machines. that would almost be utopic :) (22:50:44) mennis: ooh: http://snipurl.com/t2v2d [docs_google_com] (22:51:24) mjl-: ah, thanks (22:52:15) mennis: Well it was oberon. :) (22:53:18) mjl-: ah, i have to look into oberon sometime! (22:53:38) mjl-: architecture-neutral code (22:53:46) mjl-: i start to see resemblence to llvm code :) (22:53:57) npe [n=npe@94-224-251-223.access.telenet.be] entered the room. (22:54:28) mennis: It has it's influences in Limbo and acme. (22:55:27) mennis: Coraid's CTO is a bit of an Oberon buff. (23:00:54) mjl-: still sounds reasonable. i don't see why they try hard to compress stuff. they have a graph offseting cpu progress to disk progress. but that's probably speed, not capacity (so no need to have small binaries i would say) (23:01:02) mjl-: i've been wondering for inferno (23:01:17) mjl-: whether the vm could do any inlining of other modules at jit-time (23:08:07) mjl-: heh, i'm not sure how i happy i would be to jit firefox at runtime ;) (23:39:47) j123m [n=quassel@unas-revda.interra.ru] entered the room. (00:19:38) mennis: Anyone familiar with the windows build error mk: don't know how to make 'dev' (00:19:42) mennis: ? (00:22:53) mennis: I saw an unresolved thread on the mailing list. (00:37:57) mennis left the room (quit: Read error: 60 (Operation timed out)). (01:18:17) sswam [n=sswam@CPE-121-214-10-154.lnse3.win.bigpond.net.au] entered the room. (01:18:27) raiz left the room (quit: Client Quit). (01:21:29) npe left the room (quit: ). (01:29:56) sswam left the room (quit: "leaving"). (01:41:45) sswam [n=sswam@CPE-121-214-10-154.lnse3.win.bigpond.net.au] entered the room. (01:58:31) crink [n=crink@unaffiliated/crink] entered the room. (02:14:21) crink left the room. (04:15:22) sswam left the room (quit: "leaving"). (05:46:37) j123m left the room (quit: Read error: 104 (Connection reset by peer)). (05:52:14) j123m [n=quassel@unas-revda.interra.ru] entered the room. (06:45:49) me__ [n=venkates@c-68-55-179-48.hsd1.md.comcast.net] entered the room. (10:05:05) j123m left the room (quit: Remote closed the connection). (10:05:56) j123m [n=quassel@unas-revda.interra.ru] entered the room. (10:53:45) j123m left the room (quit: Remote closed the connection). (11:00:31) j123m [n=quassel@unas-revda.interra.ru] entered the room. (11:04:15) mjl-: good day all (12:37:45) mjl-: the tls bug (12:37:54) mjl-: a good moment to update inferno's ssl/tls code :P (12:40:57) mjl-: from what i understood there are two related problems: 1. key renegotiation doesn't bind the negotiation to the existing keys/ciphers. so an attacker (mitm) can create an encrypted connection, exchange some data with server, and then let the client continue the connection legitimately. (12:41:59) mjl-: and 2. some servers that use tls (e.g. http servers) read a request, determine they need more authentication (e.g. client certs), so ask the tls connection to get more authinfo, and after that has succeeded, they interpret the original request as if under the new authinfo (13:17:20) sswam [n=sswam@CPE-121-214-10-154.lnse3.win.bigpond.net.au] entered the room. (13:33:13) maht left the room (quit: zelazny.freenode.net irc.freenode.net). (13:36:51) maht [n=maht__@85-189-31-174.proweb.managedbroadband.co.uk] entered the room. (14:17:09) j123m left the room (quit: Remote closed the connection). (14:17:20) j123m [n=quassel@unas-revda.interra.ru] entered the room. (14:50:49) andguent: mjl-: just curious: did you have a look at st while writing novt? (15:03:01) mjl-: andguent: i think i had the first version of novt before looking at st. then i only looked at it for a few mins (15:03:17) mjl-: so i don't know how that code works really (15:03:21) mjl-: but it looked quite small :) (15:03:24) andguent: ;0 (15:03:25) andguent: ;) (15:03:50) mjl-: but around that time i realised i would just look at ecma-48 and implement a subset of the control sequences defined there (15:04:07) andguent: mjl-: iwp9 in germany? what do you think. feel free to drop a positive comment :D (15:04:23) mjl-: and circumvent the whole "what should i claim to be compatible with" (vt100, xterm, something else...) (15:04:30) mjl-: iwp9 in germany sounds like a good plan :) (15:04:45) mjl-: especially since it would be cheap for me to go :) (15:04:58) mjl-: i'm wondering if there was a iwp9v5 chat at the last iwp9 (15:05:01) andguent: it's near the swiss border too...so if done at appropriate time it could be easily connected to a skiing holiday before or afterwards (15:05:13) andguent: i missed that in the mail :-/ (15:05:33) mjl-: hah, i combine it with a visit to my parents in .ch (15:06:16) mjl-: ah, i see you sent a mail to 9fans! (15:06:52) andguent: yeah i thought i get this rolling asap (15:07:22) andguent: for one german people are usually not that flexible and for me it would be quite some effort to pull it off. so better start early (15:08:41) andguent: i thought a university hosted iwp9 would maybe attract some new guys (15:08:54) andguent: maybe one could do a real "workshop" for those too (15:08:59) andguent: depending on how much interest there is (15:09:17) andguent: would be nice pr for plan9...especially on a campus (15:09:22) andguent: and better yet. on mine :D (15:13:52) mjl-: yes, having a university (or company) backing it is very good (15:14:13) mjl-: otherwise it will quickly start costing (lots of) money to get a place & infrastructure i think (15:53:20) npe [n=npe@195.207.5.2] entered the room. (16:19:23) qed left the room (quit: "Lost terminal"). (17:00:03) mennis [n=mennis@adsl-068-016-104-079.sip.asm.bellsouth.net] entered the room. (17:15:25) mennis left the room (quit: Remote closed the connection). (17:17:18) mennis [n=mennis@adsl-068-016-104-079.sip.asm.bellsouth.net] entered the room. (17:55:57) anothy_x: andguent: at iwp9v4 there was talk of getting sape to host v5. (17:56:27) anothy_x: he's in... antwerp? amsterdam? some a-city in northwest europe. (17:57:03) anothy_x: i forget who said they were going to talk to him about it; geoff of ron, i think. (17:57:33) anothy_x: if you want to be more proactive, that's the best list of people to talk to, i'd say. (18:00:36) mjl-: sape is in antwerp (which is in belgium) (18:01:06) mjl-: and that too is a good option (18:02:29) andguent: anothy_x: thanks. well if people already decided where it's going to be next. i won't stand in the way of course (18:03:44) anothy_x: there was no decision, just (geoff|ron) said "i'm going to poke sape about hosting it" (18:04:40) andguent: anothy_x: okay thanks. if they don't reply to the list. i try to contact them offlist about those things (18:04:43) andguent: thanks for the infos! (18:09:19) ***uriel is more and more convinced that 'conferences' are only for morons (18:10:18) uriel: fuck that, partying and hacking is all that is needed (18:13:04) anothy_x: also, those grapes are probably sour, anyway. (18:13:47) uriel: *sour* is quite a huge understatement (18:16:32) uriel: but that will teach me once more to stop counting on certain people to act with decency and honesty (not that I will ever actually *learn* but) (18:18:29) sswam1 [n=sswam@CPE-121-214-10-154.lnse3.win.bigpond.net.au] entered the room. (18:22:32) sswam left the room (quit: Read error: 60 (Operation timed out)). (18:30:46) npe left the room (quit: ). (20:34:17) j123m left the room (quit: Remote closed the connection). (20:45:33) sswam1 left the room (quit: "leaving"). (21:30:33) s_justin [n=justin@68.16.104.79] entered the room. (21:39:24) npe [n=npe@94-224-251-223.access.telenet.be] entered the room. (21:56:44) te [i=tao@gateway/shell/blinkenshell.org/session] entered the room. (21:56:56) te is now known as qed (22:26:44) npe left the room (quit: Read error: 104 (Connection reset by peer)). (22:27:16) npe [n=npe@94-224-251-223.access.telenet.be] entered the room. (23:01:17) mennis left the room (quit: ). (00:31:02) sswam [n=sswam@CPE-121-214-10-154.lnse3.win.bigpond.net.au] entered the room. (00:31:33) sswam left the room (quit: Client Quit). (00:37:15) s_justin left the room (quit: ). (00:53:23) me__ left the room (quit: "leaving"). (01:04:56) anothy_x left the room (quit: Read error: 54 (Connection reset by peer)). (01:14:10) anothy_x [n=a@adsl-99-155-110-247.dsl.bcvloh.sbcglobal.net] entered the room. (01:29:24) npe left the room (quit: ). (02:26:22) me__ [n=venkates@c-68-55-179-48.hsd1.md.comcast.net] entered the room. (03:05:10) me__ left the room (quit: "leaving"). (09:17:00) rapidfx [n=host666@vl-cen-ce1.avtlg.ru] entered the room. (10:18:23) raiz [n=raiz@p4FE8956F.dip0.t-ipconnect.de] entered the room. (10:36:18) npe [n=npe@94-224-251-223.access.telenet.be] entered the room. (11:44:18) j123m [n=quassel@unas-revda.interra.ru] entered the room. (13:48:59) eekee [n=ethan@sourcemage/guru/eekee] entered the room. (14:14:59) andguent: when i run acme on top of wm/wm in acme-sac. right-clicking on files opens them in a new window. so not in the old acme. any ideas whats happening ther? (14:16:05) andguent: -- it is not always happening...so i dont know whats going on (14:17:41) andguent: i can reproduce it this way: start wm/wm acme ... open a new acme. close that new one. and then try right clicking in the Navigator helper (14:58:55) raiz left the room (quit: Client Quit). (15:23:14) sswam [n=sswam@CPE-121-214-10-154.lnse3.win.bigpond.net.au] entered the room. (15:31:35) j123m left the room (quit: "No Ping reply in 180 seconds."). (15:31:35) j123m [n=quassel@94.190.65.228] entered the room. (16:23:39) j123m left the room (quit: zelazny.freenode.net irc.freenode.net). (16:23:39) anothy_x left the room (quit: zelazny.freenode.net irc.freenode.net). (16:27:38) j123m [n=quassel@94.190.65.228] entered the room. (16:27:38) anothy_x [n=a@adsl-99-155-110-247.dsl.bcvloh.sbcglobal.net] entered the room. (16:36:32) j123m left the room (quit: zelazny.freenode.net irc.freenode.net). (16:36:32) anothy_x left the room (quit: zelazny.freenode.net irc.freenode.net). (16:41:06) j123m [n=quassel@94.190.65.228] entered the room. (16:41:06) anothy_x [n=a@adsl-99-155-110-247.dsl.bcvloh.sbcglobal.net] entered the room. (16:41:23) j123m_ [n=quassel@unas-revda.interra.ru] entered the room. (16:44:11) j123m left the room (quit: "No Ping reply in 180 seconds."). (18:12:13) npe left the room (quit: ). (18:28:11) sswam left the room (quit: "leaving"). (19:34:24) anothy_x left the room (quit: Read error: 110 (Connection timed out)).