July 4, 2005 
< | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31 | >
July 4, 2005 [00:00:10] <@geist> ah[00:00:13] <@mmu_man> no wonder find_thread_info() was returning funky values for user_stack_base[00:00:15] <@geist> must not have been too often[00:00:15] <tic> you guys should just start using Linux... ;)[00:00:50] <@mmu_man> it makes some sense, but is really error prone[00:00:59] <tic> just port the VFS, ports, areas and you're set.[00:01:04] <@mmu_man> like the correct way of copying is to check how much to copy...[00:02:24] *** konrad77 has quit IRC[00:03:38] <@mmu_man> tic and get module dependancy hell ?[00:03:55] <@mmu_man> I'll never get back to having to use insmod[00:04:23] <@mmu_man> no reason to have the user know what the software should[00:04:46] <tic> of course you should![00:04:53] <tic> ic> i wuv lunix[00:05:09] <@mmu_man> sure I do :p[00:05:17] <@mmu_man> just doesn't fit the desktop[00:05:33] <tic> *nod*[00:05:37] <@mmu_man> and I still can't run a query on googlefs in linux :P[00:05:38] <tic> the module stuff really blows.[00:05:52] <tic> module/driver/whatever.[00:05:53] <@geist> it's designed around a different ideaology[00:05:58] <tic> no shit :)[00:06:10] <@mmu_man> it's *not* designed at all geist :)[00:06:16] <@geist> it's designed to let you not have to load all of the kernel if you dont need it, not to provide a binary driver interface[00:06:31] <@geist> it's assumed that you have the full source and can rebuild everything with every kernel version[00:06:36] <@mmu_man> it's using the elf loader to do linking, without the dependancy tracking even userland has[00:06:59] <@geist> most open source unixes have the same strategy[00:07:08] <@mmu_man> yes, which was the case 10 years ago for linux 1.1[00:07:09] <@geist> especially something so political as linux[00:07:39] <tic> mmu_man, right, nowadays most people don't really bother with that.[00:07:44] <@geist> I was figuring out how the OSX kernel module loader works[00:07:47] <tic> or rather, don't want to.[00:07:52] <@geist> it's pretty complex, but tries to solve all of those problems[00:07:58] *** WindowsUninstall has joined #haiku[00:08:05] <@geist> on top of all of that, sicne the IOKit is written in c++, they have to deal with old ABIs[00:08:06] <@mmu_man> yeah I read a little about that[00:08:33] <@mmu_man> looks much like what linux does, but with a description file of some sort to tell why and when to use the module so users do'nt have to know[00:09:04] <@geist> you declare what modules you depend on, and what versions are acceptable[00:09:15] <@mmu_man> linux tried to solve this with udev but it seems plain wrong[00:09:17] <tic> what's wrong with the beos driver loader thingy?[00:09:26] <@mmu_man> tic nothing :)[00:09:36] <@geist> nothing in particular. I dont know how well it'd scale up[00:09:42] <@geist> but it is reasonably elegant[00:09:47] <@mmu_man> well, except it forces you to use public interfaces in structs[00:09:52] <@geist> dependencies are entirely hard coded though[00:09:54] <@mmu_man> cause you can't directly link to other modules[00:10:03] <@mmu_man> which is good for cleaness[00:10:06] <@geist> you have to dynamically load the other modules, declare what versions you want, etc[00:10:18] <tic> does the pnp stuff help any?[00:10:21] <@geist> instead of the loader doing it automagically[00:10:37] <tic> yah.[00:10:42] <@mmu_man> there is nothing magic :)[00:10:47] <@mmu_man> there is no spoon (tm)[00:10:50] <@geist> not a terrible strategy, though it puts more load on the code[00:11:05] <@geist> yeah when I got to module loading in newos I could think of no easier way than the beos strategy[00:11:13] <@geist> once you have an elf loader, the beos strategy is super simple[00:11:30] <@geist> search the path, load the binary, hold a handle to all the opened ones. done[00:11:40] <@mmu_man> well the pnp stuff goes much further on the hierarchy stuff[00:11:54] <@geist> yeah I'm specifically talking about module loading[00:12:04] <@mmu_man> also classic modules just have to be declared in a list, they are taken care of[00:12:10] <tic> mmu_man, I don't like that you have to "install" pnp drivers :/[00:12:34] <@geist> yeha but that's kind of inevitable[00:12:49] <@mmu_man> well it doesn'(t feel as "fun" as the simple way we're used to[00:12:57] <tic> exactly. :|[00:12:59] <@geist> if you want to support buttloads of drivers, you have to at some point have an external entitity load only what you need[00:13:08] <@mmu_man> still wondering how it all works[00:13:15] <@mmu_man> yep[00:13:18] <@geist> this is definitely an area where haiku has gone far past newos[00:13:30] <tic> drivers/<vendorid>/<deviceid> and then symlinks to the deviceids supported?[00:13:34] <@mmu_man> the manager basically does what open firmware does with the hw tree[00:13:40] <@geist> the newos devfs is even simpler than beos. it simply iterates through every driver, loads them, gives them a chance to publish, and unloads if not present[00:13:49] <tic> yah.[00:13:55] <@mmu_man> tic the vendor/dev thing isn't always suitable[00:14:05] <tic> mmu_man, where?[00:14:09] <@geist> which frankly will work for quite a while[00:14:11] <@mmu_man> geist well R3 did that[00:14:18] <@mmu_man> or older[00:14:22] <@geist> oh the symlink thing was added then?[00:14:25] <@mmu_man> the symlink stuff appeared only at some point[00:14:33] <@mmu_man> because of scalability actually[00:14:37] <@geist> the strategy of scanning the drivers only when that part of the tree is walked is kind of interesting[00:14:38] <@mmu_man> as explained in a newsletter[00:14:45] <@geist> I dont think I've ever seen anything quite like it[00:14:52] <@geist> generally in other systems it comes from the other direction[00:15:08] <@geist> if the device exists, the node is published. the trick is whether or not you scan all drivers, or scan devices and try to match the driver[00:15:11] <tic> mmu_man, so when wouldn't it work to use vendor/devid?[00:15:27] <@geist> the beos strategy of trying to publish the device when the user "asks" for it is totally backwards[00:16:03] <tic> but it's nice. :)[00:16:10] <@mmu_man> tic just because even busses supporting vendor/dev don't use the same namespace[00:16:25] <tic> mmu_man, hrm, m'kay..[00:16:31] <@mmu_man> yeah it has some drawbacks like having to rescan everything from the usb bus manager ...[00:16:34] <@geist> I suppose. I can't think of any real advantage, except the logic of which to load is sort of automatically handled[00:16:51] <@mmu_man> at the time it made sense with fixed hw[00:17:03] <@mmu_man> with PnP it's more problematic[00:17:04] <@geist> and yeah it doesn't work well with dynamically published devices[00:17:09] <@geist> which became more of a problem later on[00:17:31] <@geist> the only real solution that works forever more is to have a database of device -> driver[00:17:40] <@geist> and load the driver when the device pops into existance[00:17:49] <@mmu_man> damn I do'nt have the fixed dosfs here[00:17:53] <@geist> which is lame[00:17:54] <tic> couldn't that database be built from the drivers though?[00:17:54] <@mmu_man> was wondering why cat didn't exit[00:18:07] <@geist> tic: sure, the details are up to the implementation[00:18:11] <@mmu_man> good old return-error-but-don't-set-numBytes[00:18:23] <@geist> ah[00:18:24] <@geist> gah I mean[00:19:24] <tic> Hrm.[00:19:40] <tic> I want speedstep.[00:20:01] <@geist> oh so your time can get off and audio starts skipping?[00:20:03] <@geist> have at it[00:20:12] <tic> Heh[00:20:16] <tic> Right, I want -working- speedstep.[00:20:29] <@Dr_Evil> that design sucks. when returning an error, requiring to set arguments to valid data is stupid[00:20:36] <PulkoMandy> +++[00:20:40] <@mmu_man> Dr_Evil no it doesn't[00:20:41] *** PulkoMandy has quit IRC[00:20:48] <@geist> why does he always do the +++ thing?[00:20:52] <@mmu_man> some BSD semantics actually set errno *while* returning 0[00:21:08] <tic> geist, isn't it some shorthand for ATH?[00:21:21] <@mmu_man> hmm though setting numBytes to 0 would return -1 from fsil but[00:21:21] <@geist> sure, it's an escape string to old style modems[00:21:25] <@geist> +++ATH[00:21:38] <@geist> guess it's just a cute macro of his[00:21:49] <tic> probably.. for some definition of "cute"[00:21:56] <@Dr_Evil> well, after +++ he would be unable to send the newline character[00:22:12] <@Dr_Evil> because the modem would enter command mode[00:22:14] *** Fanskapet has joined #haiku[00:22:18] <@geist> supposedly some modems had buggy firmware and you could *send* someone +++ATH and it would hang it up[00:22:30] <@geist> since the modem should only scan it's input from the terminal[00:22:32] <tic> cool.[00:23:00] <@geist> Dr_Evil: well, presumably he's not dialed directly into the IRC server[00:23:08] <@geist> in which case it obviously sent *something* after +++ :)[00:23:29] <@mmu_man> well in IRC you can just /quit somebody :)[00:23:51] <tic> weren't we discussing drivers?[00:23:59] <tic> (and speedstep, for my laptop! :)[00:24:15] <tic> It's imperative that we stay focused![00:25:15] <@geist> sir yes sir[00:26:02] <tic> sprinting is a nice way to develop[00:26:59] <@mmu_man> http://www.osnews.com/comment.php?news_id=11069 that's for Axel :)[00:27:33] <@mmu_man> well speedstep should be possible... just need to recalibrate system_time on each freq change[00:27:49] <@geist> or dont use such a dumb system as relying on rdtsc for time keeping[00:27:50] <tic> will that be enough for audio and stuff?[00:27:58] <@mmu_man> APM BIOSes are (supposedly but some some are screwed up) kind enough to notify us[00:27:59] <@geist> I moved away from that in newos, but you guys didn't take it as far as I can tell[00:28:04] <@geist> rdtsc as a time base is a bad idea[00:28:22] <@mmu_man> what else do you use anyway ?[00:28:32] <@geist> monotonic timer + rdtsc as delta[00:29:00] <@geist> rdtsc came out of the fact that on ppc there's something similar[00:29:07] *** Leo44 has quit IRC[00:29:09] <@geist> but derived off bus speed, so shouldn't move around[00:29:23] <tic> bus speed changes as well.[00:29:41] <@geist> sure, not saying it's a good idea[00:29:45] <@geist> but that's where it came from[00:30:01] <@geist> the only way that really works is to use a monotonic timer derived off a clock that doesn't move around[00:30:05] <@mmu_man> hmm yeah the PearPC guys had been discussing the use of rdtsc...[00:30:07] <tic> yup.[00:30:27] <@mmu_man> guess the periodic timer used to resync and rdtsc in the middle...[00:30:49] <@geist> you have a periodic timer (maybe 1000Hz, or at least 100) that increments a master tick[00:31:09] <@geist> and save what rdtsc was at the time. if you need high res timing support, you take that master tick + delta rdtsc since last tick[00:31:24] <@geist> and if you dont need it (which most places dont) you just return the master tick. very fast[00:31:29] <tic> *yawn*[00:31:50] <@geist> and it works fine on < pentiums[00:32:13] <@geist> using rdtsc is one of those neat things that totally paints you into a corner later on[00:32:28] <@mmu_man> righto[00:32:46] <@geist> nowadays machines are fast enough that running the clock at 1000Hz is no big deal[00:33:53] <@Dr_Evil> how about implementing a nano_time() function that returns nano seconds?[00:34:01] <@mmu_man> well rdtsc isn't much more braindead than HZ in linux[00:34:09] <kma> I don't think rdtsc works on my machine :\[00:34:17] <@geist> sure, you can implement pico time if you want. it just might not be as accurate :)[00:34:21] <@mmu_man> when root has to fiddle with something in /proc to change the timer rate to get mplayer to work[00:34:40] <tic> yay linux[00:34:44] <@geist> yeah but the speed of the system timer really doesn't have a lot to do with that[00:34:50] <@geist> it's the architecture of the systema roun dit[00:34:56] <@geist> beos still only had a 10ms timer, IIRC[00:35:08] <@geist> and events were scheduled at that granularity[00:35:30] <@geist> but it worked well enough, sinc eit was probably reasonably real-time as far as response[00:35:37] <@mmu_man> it reprogrammed the PIT for each deadline[00:35:40] <@geist> it didn't suddenly take 100ms to respond to something[00:35:47] <@geist> you sure mmu_man?[00:35:54] <@mmu_man> that is not using it in fixed freq mode[00:35:59] <@geist> it didn't use the local apic on a smp system?[00:36:05] <@mmu_man> yes I read that thing quite recently[00:36:15] <@geist> ist> i never looked at that part of beos. does it have a single timer queue, or one per cpu?[00:36:21] <@mmu_man> hmm well it can use the apic timer, didn't read that one[00:36:39] <@mmu_man> had to fight around on a bloody laptop...[00:36:54] <@geist> newos does it similarly, it'll use the PIT if you have a single cpu, and local apic on smp system. each cpu has it's own queue[00:36:54] <tic> it has a broken dprintf though! :)[00:37:07] <tic> *Zzz*[00:37:16] <@mmu_man> I think the chipset just doesn't im)plement the PIT correctly[00:37:19] <@mmu_man> linux worked[00:37:21] <@mmu_man> windoze did[00:37:27] <@mmu_man> beos didn't boot at all...[00:37:45] <@geist> they might be using the RTC timer or local apic or ACPI timer[00:37:46] <@mmu_man> interrupts weren't even generated, had to change the bit in the edge-level control reg[00:37:50] <@mmu_man> god knows why[00:38:21] <@geist> something else I should look at too. in theory the RTC timer is reasonably accurate. I should probably program that to generate the system_time tick, and nothing else[00:38:22] <@mmu_man> I suppose the PIT works correctly only in fixed freq mode (mode 0 or 2 or I don't know)[00:38:34] <@geist> and then program the local apic/PIT to deal with timer events[00:38:52] <@geist> then I can go back to setting the timer event dynamically, whcih is what I used to do in newos early on[00:39:07] <@mmu_man> reading the RTC is slow...[00:39:08] <@geist> ist> i know that linux/bsd seem to use both sets of timers[00:39:29] <@geist> but you dont need to read it if you just program it to generate a bunch of interrupts[00:39:36] <@mmu_man> yep[00:40:08] <@geist> anyway, which timers to use where is an area where there are tons of ways to do the same thing[00:40:09] * mmu_man pets kernel: ixp_clock_workaround[00:40:22] <@mmu_man> at least the laptop boots now :)[00:40:30] <@geist> and hiding it behind an api that is fully portable is a bit of a challenge too[00:40:37] <@mmu_man> though I'll probably avoid ATI IXP based hw[00:40:52] <@mmu_man> and handling APIC/non-APIC also[00:40:55] <@mmu_man> ...[00:41:05] <@mmu_man> though not many don't have apic now[00:42:01] <@geist> PIT vs local apic is no big deal[00:42:34] <@geist> however, making the system aggressively use the local apic may be the best way[00:42:42] <@geist> instead of only using it in smp mode[00:42:52] <@geist> of course, in the near future we'll always be in smp mode[00:43:00] <@geist> so it's about to become somewhat of a moot point[00:43:14] *** WindowsUninstall has quit IRC[00:43:55] <@geist> anyway, gonna go hack on my arm emulator a bit[00:44:02] <@geist> then I think I'll write a simple microkernel for kicks[00:44:09] <@geist> never really did that, would be interesting to try[00:44:37] <clee> hahahaha[00:45:17] <@mmu_man> damn, will Apple let you work on Hurd ? :D[00:45:23] *** slaad has joined #haiku[00:49:22] <DaaT> hey slaad[00:49:28] <slaad> Hey DaaT.[00:49:37] <DaaT> how goes it[00:49:37] <DaaT> ?[00:52:37] <slaad> Not bad, just wopke up.[00:52:49] <DaaT> lazy bum[00:55:18] <slaad> It's only 8:30[00:55:52] <@Dr_Evil> Is the KERNAL sourcecode or disassembly available?[00:56:11] <@mmu_man> C64 ?[00:56:54] <@geist> mmu_man: well, as far as I can tell Apple is gonna be reasonably laid back with regards to that stuff[00:57:12] <@Dr_Evil> mmu_man yes[00:57:23] <@mmu_man> until they say it's their stuff :)[00:57:24] <@geist> seeing as most of what I'd work on is open source anyway[00:57:37] <@geist> and half of the team is already working on open source stuff[00:57:49] <@geist> the language of the contract is pretty reasonably[00:57:51] <@mmu_man> anyway[00:57:53] <@mmu_man> zzz[00:58:10] <@geist> since I'd be working on file systems stuff, I should probably not specifically go write a new fancy fs for linux or something[00:58:18] <@geist> but I dont have much of a desire to do that anyway[00:58:31] <slaad> Get a job at Apple, geist?[00:58:44] <@geist> and if I for example closed sourced newos and starting selling it a a desktop os, that'd be a problem[00:58:55] <@geist> slaad: yeah, just got the offer letter in the mail yesterday. will probably take it[00:59:15] <slaad> They snail mailed you? How quaint! :P[00:59:20] <@geist> fedex[00:59:30] <slaad> Good stuff though. Did Newos have any helping in the offer?[00:59:38] <DaaT> congrats geist[00:59:46] <@geist> I'm sure it did, I know there were looking at it. looked at the server logs :)[00:59:58] <@geist> though I think they were dissapointed that I hadn't written a native fs for it[01:00:02] <slaad> Haha[01:00:11] <@geist> since the apple job is specifically "local filesystem engineer"[01:00:18] <@geist> which means "work with dbg doing cool shit"[01:00:37] <slaad> Did they have any contact with you prior to this?[01:00:38] <@geist> it's just like good old times at Be.[01:00:42] <slaad> That'd be pretty cool I imagine.[01:00:57] *** Fanskapet has quit IRC[01:00:57] <slaad> Only now your work might be used by more than 3 people, geist ;)[01:01:04] <@geist> Be had just switched to x86, I had just gotten "into" using beos, went to work for dbg doing filesystem stuff[01:01:21] <@geist> apple is switching to x86, gotten into using OSX, goign to work with dbg on filesystem stuff[01:02:13] <DaaT> dejá vu...[01:02:16] <@geist> it's nice working on something you can really get excited about[01:02:22] <DaaT> something's been changed in the matrix[01:02:29] <@geist> I'm firmly a OS X user now, so it's pretty exciting[01:03:15] <@geist> only bummer is the lack of any really good editors on OS X[01:03:23] <slaad> Of the code variety?[01:03:23] <@geist> there are a few *fantastic* editors on windows[01:03:28] <DaaT> will you have to move geist?[01:03:29] <@geist> yeah, code editors[01:03:36] <@geist> physically move? no[01:03:43] <DaaT> nice[01:04:01] <@geist> the apple campus is just ac ouple miles farther away than my current job[01:04:34] <DaaT> cool[01:04:35] <@geist> though when I use a mac, I tend to use vim and/or Eddie[01:04:37] <slaad> I've never used a fancy code editor. In Be I use Eddie + makefiles. In Windows I only do web monkey coding, so I just use notepad.[01:04:43] <@geist> pavel has a version for OS X[01:04:55] <slaad> Has it advanced much?[01:05:01] <@geist> though it's pretty buggy[01:05:07] <@geist> the shell stuff doesn't work, for example[01:05:14] <@geist> maybe I can get him to give me the source and fix it[01:05:21] <slaad> Ah... that's one of the best things about Eddie.[01:05:53] <slaad> Code, command-/, command-z, enter, command-d for each error. Rinse and repeat.[01:06:09] <DaaT> that could be your first task at apple :)[01:06:20] <@geist> yeah I'll send an email to him right now[01:06:37] * mmu_man pets XEmacs[01:07:00] <slaad> mmu_man man is a Loonix user in disguise.[01:07:05] * DaaT smacks mmu_man[01:07:17] <DaaT> oops, sorry :P[01:09:10] <@mmu_man> well native XEmacs is better than in *nix :p[01:09:26] <@mmu_man> and I can guess it's probably better in osx too[01:09:32] <@mmu_man> since X11 ... hmm well[01:10:45] <@geist> okay, just sent him an email[01:10:49] <@geist> need to eat[01:11:05] <slaad> I guess you'd get special privileges with him, eh?[01:11:58] <@mmu_man> zzz[01:12:11] *** mmu_man has quit IRC[01:13:17] <@geist> maybe I dunno[01:13:27] <@geist> haven't talked to pavel in years. dont know if he even remembers me[01:14:18] <slaad> Ah.[01:15:32] <DaaT> send a pic of you as an attachment then :)[01:15:41] <@geist> hah[01:15:51] <@geist> I try to keep the number of pics of me down to a minimum[01:16:18] <DaaT> i'm sure everyone appreciates that[01:16:20] <DaaT> ;)[01:24:46] *** fyysik has joined #haiku[01:32:50] <slaad> Anyone here good with algorithms and want to give me a hand?[01:41:40] * DaaT hears a cricket...[01:53:24] *** Koki has joined #haiku[02:09:50] *** DaaT has quit IRC[02:27:44] *** vision has joined #haiku[02:27:47] <vision> hi[02:27:53] *** vision is now known as kl1[02:28:11] <kl1> is posible to do ssh whith BeOS?[02:28:22] <slaad> Yes.[02:28:38] <kl1> cause y try and it say error[02:32:23] <slaad> What version of BeOS are you using?[02:37:32] *** kl1 has quit IRC[02:39:24] *** sektor has joined #haiku[02:40:16] <brennanOS> slaad: still want algorithm help?[02:42:39] *** sektor has left #haiku[02:43:39] *** petterhj has joined #haiku[02:48:05] <slaad> Nope... I just this second figured it out :)[02:48:11] <brennanOS> slaad: damn you[02:48:15] * slaad grins.[02:48:23] <slaad> I can still give it to you if you want.[02:48:24] <brennanOS> something interesting?[02:48:25] <brennanOS> sure[02:48:31] <brennanOS> my brain needs some work[02:48:57] <slaad> It's a 1D set difference.[02:49:06] <slaad> ie. Given (1 -> 6), (9 -> 11), (14 -> 18), (20 -> 24) and (0 -> 3), (4 -> 7), (8 -> 12), (15 -> 16), (17 -> 21), (23 -> 25) it needs to produce (3 - > 4), (14 -> 15), (16, 17), (21, 23)[02:49:40] <slaad> If you draw those sets on a 1D plane it probably makes a lot more sense.[02:50:28] <brennanOS> hm.. never seen that notation... what does (x -> y) indicate?[02:50:32] <brennanOS> is this a graph?[02:50:55] <slaad> Yeah. But only in one D.[02:51:28] <slaad> It's actually x1->x2. So 1 -> 6 would be [1, 6]. (ie. 1, 2, 3, 4, 5, 6)[02:53:00] <slaad> It's actually for a Constructive Solid Geometry in a ray tracer.[02:53:57] <slaad> So you might have a sphere centred on the origin of radius 5 and another one of radius 4 and you could make a hollow sphere with the difference. [-5, 5] difference [4, 4] gives you [-5, -4], [4, 5][02:53:58] *** fyysik has quit IRC[02:54:08] <brennanOS> oh[02:54:30] <brennanOS> so its a set of points on a cartesian plane[02:54:38] <slaad> Yeah. But only for the one dimension.[02:55:31] <brennanOS> hmm[02:55:35] <brennanOS> one dimension[02:56:01] <brennanOS> how can you describe a location in one dimesion with two data points?[02:56:26] <slaad> Because it's a range.[02:56:31] <brennanOS> ahhh[02:56:31] <slaad> I'm probably not explaining it very well.[02:56:39] <brennanOS> that would explain the notation[02:56:46] <brennanOS> that makes sense[02:57:09] <brennanOS> youve tricked me into thinking about a math problem[02:57:17] <brennanOS> I dont like you anymore[02:57:26] <slaad> Mwaha! :)[02:57:49] <brennanOS> although its kind of interesting... I just recently finished a linear algebra course[02:58:29] <brennanOS> it was pretty cool[02:58:42] <slaad> I'm not sure I've done linear algebra...[02:58:59] <brennanOS> vectors, matrices, and n-dimensional spaces[02:59:12] <brennanOS> properties of n-dimensional spaces[02:59:31] <brennanOS> properties of sets of point within a space, and so on[02:59:36] <slaad> http://beos.bong.com.au/temp/difference.png by the way[03:00:01] <slaad> Ah... I think I did that sort of thing in first year.[03:00:13] <brennanOS> its possible[03:00:21] <brennanOS> its covered to some degree in Calculus as well[03:00:35] <slaad> Yeah... I do engineering though, so we have difference course names and stuff.[03:00:48] <brennanOS> in US university, its a 3rd/4th year class[03:01:01] <brennanOS> I guess if you are in engineering it could be first year[03:01:26] <slaad> We did matrices and vectors and stuff in first semester. I know we did some space related stuff as well. But I think I've forgotten it :)[03:01:35] <slaad> Most of my course is signal processing stuff.[03:01:37] <brennanOS> it satisfied a math requirement for CS[03:01:51] <brennanOS> I have one damn class left, and no desire to take it[03:02:06] <slaad> What is it?[03:02:25] <brennanOS> sequential and parallel algorithms[03:02:47] <brennanOS> I have the book already, I went through nearly half of it last sememster before dropping due to time constraints[03:02:53] <brennanOS> I work full time and go to school at night[03:03:02] <slaad> Ah.[03:03:14] <brennanOS> It sucked, I was getting an A[03:03:17] <slaad> I do part time work, part time Uni and part time coding :)[03:03:20] <slaad> Why'd you drop it then?[03:03:32] <brennanOS> but I was starting to fall behind, due to the math class being pretty hard[03:03:39] <brennanOS> so I stuck with the math[03:03:42] <slaad> Ah[03:04:02] *** thies_ has quit IRC[03:04:08] <brennanOS> I was doing about 15 hours a week in both at the time[03:04:32] <brennanOS> the hardest two classes ive taken[03:04:42] <brennanOS> hardest = most challenging[03:05:08] <slaad> Ouch.[03:05:21] <brennanOS> heh[03:05:22] <brennanOS> yep[03:05:35] *** thies_ has joined #haiku[03:05:39] <brennanOS> that algorithm class is essentially another senior level math[03:06:43] <brennanOS> but I already have a job in the industry so Im not in a huge hurry[03:06:50] <@geist> a 3rd level math class I took involved a lot of matrices[03:06:54] <@geist> linear algebra IIRC[03:07:07] <brennanOS> of course its kind of a lame job[03:07:14] <@geist> a lot of space conversion stuff[03:07:16] <brennanOS> geist: did you interview w/Apple?[03:07:21] <brennanOS> geist yeah[03:07:23] <@geist> yeah, got an offer in the mail too[03:07:29] <brennanOS> geist: to do what?[03:07:35] <@geist> work at Apple[03:07:52] <brennanOS> geist: no shit ;) What area?[03:07:58] <@geist> Core OS, filesystems[03:08:03] <@geist> just like at Be. working with dbg[03:08:11] <brennanOS> cool[03:08:16] <brennanOS> I want a job like that[03:08:32] <brennanOS> I guess I should write a filesystem or something then, eh?[03:08:59] <@geist> you should learn a lot of everything, and dive in and in general just be an expert[03:09:03] <@geist> if you want the good jobs[03:09:14] <@geist> I'll be much more junior here, which is perfectly fine with me[03:09:18] <@geist> as long as I'm being pushed[03:09:33] <@geist> I'm feeling stagnant now, like I haven't learned much in the last few years[03:09:34] <@geist> I dont like that[03:09:53] <clee> it's a shitty feeling, that[03:10:17] <clee> geist: do you have any numbers on how many ex-Be engineers work at Apple these days?[03:10:55] <@geist> ist> i know of maybe 5 or 6[03:11:05] <slaad> Hrm... maybe I didn't solve this. Damnit.[03:11:16] <clee> didn't they hire the guys who worked at Gobe on Productive, too?[03:11:23] <@geist> they didn't work at Be[03:11:34] <clee> well, right, but I mean, they worked with BeOS[03:11:42] <@geist> I only know Be people[03:11:46] * clee nods[03:11:47] <clee> fair enough[03:11:55] <@geist> but yeah, the claris works guys may be there too[03:19:38] <brennanOS> sorry, had to hold my brand spanking new daughter there[03:20:04] <brennanOS> now shes laying in a basket looking just cute as hell[03:21:58] <slaad> How old are you, brennanOS?[03:22:03] <brennanOS> 31[03:22:25] <brennanOS> why?[03:22:28] <brennanOS> school?[03:22:41] <slaad> Yeah, school + kids.[03:22:53] <brennanOS> well, the kid is only 3 days old[03:23:06] <brennanOS> and I already have a B.S. degree in Psychology[03:23:10] <slaad> Woah. That's new.[03:23:14] <brennanOS> the CS degree will be degree 2[03:23:18] <brennanOS> yeah[03:23:25] <brennanOS> shes way new[03:23:36] <slaad> Still under warranty incase something breaks, eh? ;)[03:23:39] <brennanOS> yeah[03:23:51] <brennanOS> we got the 2 year maintenance contract[03:23:55] <slaad> Are you playing Mr Mum?[03:23:56] <slaad> Hehe.[03:24:18] <brennanOS> yeah, my wife is taking a bath... doesnt take much to look at a sleeping child, though[03:24:24] <brennanOS> they dont do much at this age[03:24:28] <brennanOS> just look cute[03:24:47] <slaad> I had a brain fart and thought it'd be 11:30am in brennanOSville like it is here.[03:25:06] <brennanOS> oh[03:25:15] <brennanOS> nope, 8:30 pm[03:25:52] <brennanOS> geist: well, congrats on the offer... you going to take it?[03:27:16] <brennanOS> brennanOSville, I like that[03:28:49] <slaad> Are you planning to work? Or be a Mr Mum full time?[03:29:18] <@geist> brennanOS: yeah[03:29:26] <brennanOS> slaad: I alreay work[03:29:47] <brennanOS> slaad: I am a "Java developer"[03:30:21] <brennanOS> geist: neat-o... do you have to move to Cupertino?[03:30:37] <brennanOS> slaad: we are working on the new point of sale system for Best Buy[03:31:05] <slaad> I meant with the new arrival.[03:31:10] <slaad> Ah, okay.[03:32:10] *** fyysik has joined #haiku[03:32:31] <@geist> brennanOS: no, I live in the bay area already[03:32:42] <@geist> though if gas prices keep going up I may eventually do so[03:33:09] <brennanOS> do you take the BART?[03:33:18] <@geist> doesn't go to cupertino[03:33:26] <brennanOS> heh[03:33:46] <@geist> the bart is mostly there to get flks from east bay and upper peninsula into and out of san francisco[03:33:53] <brennanOS> yeah[03:33:58] * brennanOS pictures a BART map[03:34:14] <brennanOS> my sister used to live in Concord[03:34:19] <brennanOS> and BART to work[03:35:16] <@geist> I live in SF, so it's mostly a matter of driving down 280 to cupertino[03:35:31] <@geist> not really any longer than I currently commute, but it's reasonably long. ~40 miles[03:35:42] <brennanOS> hm[03:35:47] <brennanOS> not too bad, how long does that take?[03:35:54] <@geist> about 45 minutes[03:36:02] <brennanOS> that not bad at all[03:36:09] <brennanOS> in fact thats good for 40 miles[03:36:20] <@geist> yeah that's mostly because 280 is like 75mph[03:36:48] <brennanOS> around here 40 miles would take an hour or more[03:36:50] <brennanOS> likely[03:37:10] <@geist> and this is all relying on not having to commute during rush hour[03:37:16] <brennanOS> oh[03:37:19] <@geist> I usually work 10-7 or so[03:37:19] <brennanOS> well there you go[03:37:38] <brennanOS> off hours are nice[03:37:48] <brennanOS> its a whole different world[03:38:11] <slaad> Those are long hours.[03:38:29] <brennanOS> that looks normal to me[03:38:34] <brennanOS> I work 8 - 5[03:38:36] <slaad> I'd prefer to go in early and leave early though.[03:38:45] <@geist> you live in europe?[03:38:54] <slaad> Ah... I do 8:30 - 4/5.[03:39:01] <slaad> Nope... .au[03:39:11] <brennanOS> Austria?[03:39:14] <brennanOS> hehe[03:39:17] * slaad whaps.[03:39:19] <slaad> Australia.[03:39:27] <@geist> yeah, good ol austria[03:39:35] <@geist> we have one of your mans here. he's running our state[03:39:40] <brennanOS> produces governors left and right[03:39:50] <slaad> Hehe.[03:39:57] <brennanOS> and horrible dictators[03:40:13] * geist has a headache[03:40:22] <@geist> caffiene isn't making it better[03:40:29] * brennanOS hands geist a beer[03:40:32] * slaad hugs his coffee[03:40:36] <@geist> generally if I have a headache is either because I haven't had enough caffiene or water[03:40:43] <@geist> trouble is they're sort of mutually exclusive[03:40:50] <brennanOS> I cant drink caffiene anymore[03:40:51] <@geist> since caffiene makes you pee[03:40:54] <slaad> Watery coffee?[03:40:54] <brennanOS> get panic attacks[03:41:10] <@geist> so I'll finish this coffee and drink a bunch of water[03:43:43] <brennanOS> I dont know if this kid of mine could possibly be any cuter[03:44:55] <slaad> Just wait until Kid 1.0 turns into Teenager 1.0.[03:45:19] <brennanOS> Ill just sit and look at the baby pictures[03:45:38] <slaad> Haha.[03:47:21] <brennanOS> geist: you work for YellowTab right now?[03:47:37] <@geist> no[03:47:54] <brennanOS> ok[03:47:59] <brennanOS> where do you work now?[03:48:15] <@geist> Danger, Inc.[03:48:47] *** petterhj has quit IRC[03:50:26] <brennanOS> what do they do[03:51:07] <@geist> www.danger.com[03:51:25] <brennanOS> but Im trying to annoy you with questions[03:51:32] <slaad> Heh[03:51:34] <@geist> just busy[03:51:46] <@geist> and I'm not real happy with the company (hence why I'm leaving)[03:51:53] <@geist> and thus I'm probably not going to say nice things[03:52:05] <@geist> and it's the weekend, I dont really want to talk and/or think about it[03:52:18] <@geist> since I'm really dreading giving notice on tuesday[03:52:31] <brennanOS> Is it run by a bunch of dipshits?[03:52:46] <@geist> again, see above[03:52:53] <slaad> Hehe.[04:04:33] *** Koki has quit IRC[04:28:00] *** durin42 has quit IRC[04:31:22] *** durin42 has joined #haiku[04:32:08] *** thrain42_ is now known as thrain42[04:41:25] *** SiCuTDeUx has quit IRC[04:44:04] *** ilorvi has quit IRC[04:45:40] *** fyysik has quit IRC[04:46:44] *** ilorvi has joined #haiku[05:17:21] *** dj_long has quit IRC[05:54:40] *** Teknomancer has joined #haiku[05:56:30] *** Betabel has quit IRC[06:08:21] *** Teknomancer has quit IRC[06:33:15] *** brennanOS has quit IRC[06:48:57] *** miqlas has joined #haiku[06:49:02] <miqlas> Re![06:51:51] *** miqlas has quit IRC[07:31:57] *** khorben has joined #haiku[07:32:44] *** Methe has quit IRC[08:02:17] *** dos4gw has joined #haiku[08:22:01] *** Dr_Evil has quit IRC[08:33:08] *** JBurton has joined #haiku[08:33:08] *** ChanServ sets mode: +o JBurton[08:41:15] <@JBurton> hi all[08:41:21] <muco> hi :)[08:44:25] <tic> heya[08:56:31] *** mazon is now known as Mazon[09:07:38] <@JBurton> yo tic[09:07:43] <tic> 'sup?[09:07:48] <[Beta]> moaning.[09:07:57] <JBurton> hey [Beta][09:08:07] <@JBurton> tic I'm sleepy[09:08:11] <[Beta]> hey JBurton, sup?[09:08:13] <tic> This is nice: http://dirtsimple.org/2004/12/python-is-not-java.html[09:08:38] <[Beta]> nice news about geist.[09:09:55] <JBurton> where, [Beta] ?[09:11:02] <[Beta]> in here (I dont think its public knowledge yet)[09:13:00] <tic> apple.[09:13:43] *** Kernel86 has quit IRC[09:13:44] *** muqo has joined #haiku[09:14:57] *** muco has quit IRC[09:19:40] <@JBurton> ooh nice[09:34:00] <tic> In R5, messages are sometimes stored as RAWT. How do I retrieve that as a regular BMessage?[09:34:03] <tic> is it perhaps flattened?[09:35:32] *** PulkoMandy has joined #haiku[09:40:19] <sys2> what about geist?[09:40:29] <sys2> anything new hapened in haiku btw?[09:42:27] <@JBurton> some small stuff[09:51:14] <@JBurton> tic "stored" where ? on disk ?[09:51:56] *** emwe has joined #haiku[09:52:02] <@JBurton> it could be flattened, yes, tic[09:58:33] <tic> JBurton, when sent through drag'n'drop and whatnot.[09:59:10] <tic> JBurton, Previous() doesn't work, so I have to get _previous_. On Zeta, it's a B_MESSAGE_TYPE as they're flattened anyway, and on R5 they're B_RAW_TYPE. So I have to add a hack. :)[09:59:16] <@JBurton> flattened, IIRC[09:59:24] <@JBurton> oh[10:00:55] *** Teknomancer has joined #haiku[10:02:57] <tic> cool, works in R5. Hope it works in Zeta too. :P[10:07:27] <tic> argh, I hate having to type in a ;[10:08:21] *** Mandaryne has joined #haiku[10:09:51] <@JBurton> a ";" ?[10:10:12] <@JBurton> well you have to write it after every other line in C/C++, tic I'm sorry :P[10:15:13] <tic> yeah, it just sucks[10:15:23] <Teknomancer> huh :)[10:15:23] <Teknomancer> C++ rocks.[10:15:32] <Teknomancer> i'm typing ; in python now :^)[10:15:44] <tic> No, C++ sucks. But that wasn't what I was talking about. I belive that having to explicitly tell the compiler that there's a \n coming sucks.[10:16:03] <@JBurton> hmmm what do you mean ?[10:16:21] <tic> Often you write one statement per line.[10:16:34] <@JBurton> yes[10:16:35] <tic> therefore, the common case is for lines to ends with ";\n"[10:16:51] <@JBurton> hmmm no, why the ";" ?[10:16:59] <tic> so, why should I have to tell the compiler that? Instead, there should be a \ construction to break the lines if I need to.[10:17:33] <@JBurton> you just do printf("blablabla\nblablabla\n");[10:17:35] <tic> but the compiler should just assume that if the line ends, then it's a complete statement.[10:17:37] <tic> bah.[10:17:43] <tic> you don't understand what I mean.[10:17:47] <tic> I don't want to type in ;.[10:17:56] <muqo> tic: go use python :)[10:18:00] <tic> muqo, I do.[10:18:11] <tic> muqo, but sometimes I have to use the C-family of languages.[10:18:15] <@JBurton> sorry, tic :([10:18:18] <muqo> too bad :)[10:18:24] <muqo> i'm sorry for you :)[10:18:45] <tic> JBurton, when writing C/C++, how often do your statements span >1 line?[10:18:55] <muqo> often[10:18:56] <@JBurton> sometimes[10:19:04] <muqo> with c++ really often to me[10:19:09] <@JBurton> but I never write such stuff, tic[10:19:13] <muqo> templates really requires more than 1 line :)[10:19:15] <Teknomancer> python sucks[10:19:16] <@JBurton> I just hit enter[10:19:17] <tic> JBurton, never write what stuff?[10:19:19] <Teknomancer> compared to C++ that is[10:19:24] <tic> JBurton, and then what happens?[10:19:32] <tic> Teknomancer, hahaha. Right.[10:19:33] <@JBurton> er... what do you think it happens ?[10:19:49] <tic> JBurton, the editor inserts a ; for you?[10:19:52] <Teknomancer> yeah haiku written in python would rock indeed.[10:19:52] <Teknomancer> bah[10:19:56] <@JBurton> the ";" is what indicates the end of a statement[10:20:02] <@JBurton> so you can write[10:20:03] <@JBurton> printf([10:20:11] <@JBurton> "a"[10:20:11] <@JBurton> )[10:20:11] <tic> JBurton, right, I understand that.[10:20:11] <Teknomancer> even your official python guide says a simple file operation can be done upto 1000 times (yes 1000) times faster in C++[10:20:11] <@JBurton> ;[10:20:12] <@JBurton> and it'll work[10:20:26] <tic> Teknomancer, I know that. But speed isn't everything...[10:20:30] <tic> JBurton, I _KNOW_ that.[10:20:32] <Teknomancer> then use VB[10:20:43] <@JBurton> hrmmmmmmmm[10:20:44] <Teknomancer> ;-P[10:20:44] <tic> Teknomancer, so VB is a crippled language, and cannot be compared to Python.[10:20:54] <Teknomancer> python is crippled compared to c++[10:20:56] <tic> Teknomancer, just keep quiet if you don't have anything useful to add, please.[10:21:03] <Teknomancer> yes sir.[10:21:17] <tic> JBurton, look. Often, when writing code, you write something like[10:21:21] <Teknomancer> you have the right to say c++ sucks, while i have the right to shut up.[10:21:26] <tic> printf("here I am!\n");[10:21:28] <tic> ic> i = 2;[10:21:29] <tic> etc[10:21:42] <@JBurton> tic ok[10:21:49] <tic> JBurton, in those cases, why should I tell the parser that there's an end of line coming?[10:22:04] <tic> JBurton, and if I _do_ want to break statements across multiple lines, I should just be able to put a \ at the end of the line[10:22:06] <tic> like:[10:22:16] <tic> printf("something very long\[10:22:22] <tic> and here it goes")[10:22:27] <@JBurton> you can do that[10:22:28] <tic> (okay, so that works anyway.. ;)[10:22:31] <tic> bad example.[10:22:35] <tic> anyway.[10:22:44] <@JBurton> printf("something very long"" and there it goes");[10:22:47] <tic> Do you understand what I mean?[10:22:50] <@JBurton> on> on the other line[10:22:54] <tic> yes yes, I know.[10:22:55] <@JBurton> printf("something very long"" and there it goes");[10:22:57] <@JBurton> ouch[10:23:08] <tic> So, do you see my point?[10:23:26] <@JBurton> no, as you can still split the strings without any construction[10:23:30] <@JBurton> printf("something very long"[10:23:30] <@JBurton> " and there it goes");[10:23:42] <tic> I'm not talking about strings![10:23:45] <tic> I'm talking about statements in general.[10:23:47] <tic> argh.[10:23:50] <muqo> template<typename T1_, typename T2_>[10:23:50] <muqo> INLINE[10:23:50] <muqo> const usize_t fetchMinimumSize([10:23:50] <muqo> const T1_& v1[10:23:50] <muqo> , const T2_& v2[10:23:50] <muqo> , const utils::bool2type<false>&[10:23:51] <@JBurton> :([10:23:52] <muqo> , const utils::bool2type<false>&)[10:23:54] <muqo> { return 0; }[10:23:56] <muqo> yeah that would be MUCH better in 1 line[10:23:57] <Teknomancer> ,[10:24:13] *** PulkoMandy has quit IRC[10:24:13] <@JBurton> maybe you can paste something here, tic ? http://rafb.net/paste/[10:24:20] <tic> yes, was going to.[10:25:07] <@JBurton> muqo :P[10:25:17] <Teknomancer> muqo hehe[10:25:18] <@JBurton> templates are evil, anyway, muqo :P[10:25:27] *** xeD has joined #haiku[10:25:30] *** Korli has joined #haiku[10:25:30] *** ChanServ sets mode: +o Korli[10:25:37] <@Korli> hey JBurton[10:25:39] <muqo> JBurton: depending on your attitude, they can be evil[10:25:48] <@JBurton> hi Korli[10:26:01] <tic> JBurton, http://rafb.net/paste/results/MRci6Y60.html[10:26:59] <muqo> yeah you can omit () if you are not interested in the result[10:27:07] <muqo> like in old vb[10:27:17] <muqo> that would mean Unlock[10:27:22] <muqo> instead of Unlock()[10:27:26] <muqo> muuuuuch better[10:27:30] <tic> Huh?[10:27:37] <muqo> wouldn't it be better?[10:27:38] <tic> just shut up, please?[10:27:45] <tic> No. I'm only talking about ;s[10:27:49] <tic> JBurton, got it?[10:27:54] <muqo> why not () too![10:28:03] <tic> Crap, Vision doesn't do ignore. :([10:28:05] <muqo> they are not really that needed[10:28:39] <Teknomancer> yeah i've would have done it a long time ago[10:28:41] <tic> no, that's ambigous. ref vs call.[10:28:42] <@JBurton> tic wait a sec[10:28:50] <tic> but ; is totally unambigous.[10:28:56] *** Konrad has joined #haiku[10:29:00] <muqo> tic: c++ is L2 already so nothing new[10:29:09] <@JBurton> tic you mean this line ? BString *pred = gGui->BuildPredicate( \[10:29:09] <@JBurton> &values)[10:29:09] <tic> muqo, L2?[10:29:32] <tic> JBurton, no, just generally how you'd write C/C++ using my proposal: skip all ;, and use \ to break lines where needed.[10:29:36] <muqo> tic: context dependant semantics over the same syntax[10:29:48] <@JBurton> aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaah[10:29:49] <@JBurton> tic[10:29:52] <@JBurton> okay I understood[10:30:15] <tic> JBurton, so, I just think that's something that shouldn't be there at all.[10:30:19] <@JBurton> tic I also was annoyed by the use of ";" when I passed from basic to pascal[10:30:29] <@JBurton> tic but now I got used to it, and I think it's nice[10:30:36] <Teknomancer> it is :^)[10:30:38] <tic> JBurton, why is it nice?[10:30:47] <tic> Sorry, I don't understand in what way it would help you.[10:32:28] <@JBurton> well so that is clear where is an statement end[10:32:37] <@JBurton> and you can use the indentation you prefer[10:32:39] <@JBurton> and style[10:32:40] <@JBurton> and whatnot[10:32:49] <Teknomancer> true..[10:32:57] <Teknomancer> i don't like being forced to indent the way python forces.[10:33:05] <@JBurton> or ruby[10:33:08] <Teknomancer> especially since [tab] is invisible[10:33:10] <muqo> Teknomancer: actually i like python too[10:33:14] <Teknomancer> so do i[10:33:16] <muqo> it's just another point of view[10:33:17] <Teknomancer> but C++ doesn't suck :)[10:33:28] <muqo> Teknomancer: it does in some ways :)[10:33:38] <Teknomancer> muqo that's subjective[10:33:41] <muqo> it's just not the ; the point[10:33:53] <Konrad> Good morning[10:33:58] <muqo> the point is more related to the fact that there is no 100% std compiler out there[10:34:01] <Konrad> JBurton hi[10:34:05] <@JBurton> hi Konrad[10:34:10] <Teknomancer> hi Konrad[10:34:14] <Konrad> Hi Teknomancer[10:34:25] <muqo> that the std comitee is too slow to accept new things[10:34:37] <muqo> like moveables[10:34:53] <Teknomancer> what moveables?[10:35:26] <muqo> i'll google a link, give me a sec[10:35:32] <Teknomancer> 1[10:35:33] <Teknomancer> 2[10:35:36] <Teknomancer> :^)[10:35:54] <muqo> i can't find it[10:36:01] <muqo> sigh[10:36:14] <Teknomancer> np[10:36:40] <muqo> it would be a way to avoid temporaries[10:37:09] <muqo> whan the reference to const would not cut it :)[10:38:31] <muqo> well just think to the std lib[10:38:45] <muqo> it's really too small to be usefull[10:38:54] <muqo> compare it with python[10:39:00] <muqo> or smalltalk :D[10:39:29] <Teknomancer> you mean stl ..[10:39:33] <muqo> yup[10:39:38] <muqo> stl was the pre standard name[10:39:49] <muqo> now it's just standard library :)[10:39:52] <tic> stl is just to make up for C++ inherent suckyness. ;)[10:40:16] <muqo> tic: i suppose you don't know boost, do you? :)[10:40:33] <tic> boost.python. :)[10:40:39] <Teknomancer> ugh[10:40:43] <muqo> yeah you know a reflective thing :)[10:42:10] <Teknomancer> JBurton are you coding in python as well ?[10:42:29] <@JBurton> Teknomancer no, but I'd like to learn it[10:42:37] <@JBurton> and I'd like to learn ruby too[10:42:37] <Teknomancer> ok[10:42:42] <@JBurton> I have both on my hd actually[10:43:03] <muqo> JBurton: just remember 'pass', that would solve your first problem :)[10:43:30] <muqo> JBurton: if you are interested in patterns and OOP smalltalk would be better to play with[10:43:35] <Teknomancer> i've started on python myself[10:43:42] <@JBurton> muqo :)))[10:44:06] <tic> python is yum[10:44:09] <tic> pure eyecandy[10:44:19] <muqo> ah yeah and of course python is NOT THAT OO[10:44:24] <muqo> :D[10:45:11] <tic> of course it's not OO, not everything is an object! Or wait, ...[10:45:18] <tic> ;)[10:45:19] <@JBurton> muqo watch out!!! you will be flamed now :)[10:45:26] <Teknomancer> lol[10:45:27] <Teknomancer> true[10:45:31] <muqo> so[10:45:35] <tic> mm, duck typing..[10:45:36] <muqo> everything is an object?[10:45:36] <muqo> sure?[10:45:41] <tic> yeah.[10:45:43] <Teknomancer> nope[10:45:44] <tic> no, not really[10:45:45] <Teknomancer> i don't think[10:45:49] <tic> not literal numbers.[10:45:56] <tic> and keywords[10:46:05] <Teknomancer> and whitespaces[10:46:05] <tic> but everything else is.[10:46:14] <tic> yes. and whitespaces.[10:46:17] <muqo> is smalltalk i have a Boolean, that has two subclasses, True and False. both are singletons[10:46:17] <muqo> in stalk there is no if[10:46:17] <tic> whitespace[10:46:22] <muqo> (condition) ifTrue: [ doSomething][10:46:32] <muqo> ifTrue is a method of Boolean[10:46:38] <muqo> that is OO[10:47:20] <@JBurton> oh[10:47:32] <muqo> what about a for?[10:47:51] <tic> strlen (x:xs) = 1 + (strlen xs)[10:47:53] <tic> strlen [] = 0[10:47:54] <muqo> 1 to: 100 do: [i| do something with i][10:47:54] <tic> ! :)[10:48:34] <muqo> tic: that looks more as a rip from haskell to me :)[10:48:40] <tic> muqo, that -is- haskell. :)[10:48:52] <muqo> tic: gh, that would explain a lot :)[10:49:01] <tic> a lot?[10:49:13] <muqo> a lot of things in the fact that it looks like haskell :)[10:49:39] <tic> aha. well, yes. thought you meant before the pasting..[10:49:42] <muqo> that si pure functional[10:50:04] <muqo> c++ metaprogramming layer stole a lot from haskell[10:51:13] <tic> yap[10:51:20] <tic> lazy evaluaton is a cool thing.[10:51:43] <muqo> the Cool Thing out there[10:52:00] <muqo> i miss folding with std containers[10:52:22] <muqo> (not related to lazy eval)[10:52:35] <tic> PyPy have this thunk() object which is rather cool.[10:52:38] <tic> forces lazy evaluation.[10:52:52] <tic> but I don't know how well it'd work for having lazy lists[11:03:55] <@JBurton> brb coffee[11:05:15] <muqo> have a nice coffee :)[11:15:44] <kma> bah.. suck up :)[11:18:24] <@JBurton> re[11:34:59] <@Korli> http://www.bedoper.com/bedoper/2005/32.htm ahah[11:35:12] *** Teknomancer has quit IRC[11:35:48] *** DaaT has joined #haiku[11:46:54] <@JBurton> Korli eheh[11:59:14] *** Sil2100 has joined #haiku[12:17:25] <CIA-6> jackburton * r13418 /haiku/trunk/src/kits/interface/Menu.cpp: the menuitem wasn't deselected when you moved the mouse away from the menu window. Thanks to Axel for reporting.[12:21:18] *** slaad has quit IRC[12:29:25] *** kr1stof has joined #haiku[12:30:47] <CIA-6> bonefish * r13419 /haiku/trunk/src/kits/app/ (Application.cpp Looper.cpp):[12:30:47] <CIA-6> Removed the handling of B_QUIT_REQUESTED in BApplication::DispatchMessage().[12:30:47] <CIA-6> It's already done in BLooper. Now we additionally send a reply to it, when[12:30:47] <CIA-6> the message comes from the registrar on shutdown.[12:41:22] *** DaaT has quit IRC[12:54:59] *** Koki has joined #haiku[13:14:04] *** BGA has joined #haiku[13:14:04] *** ChanServ sets mode: +o BGA[13:14:37] <CIA-6> bonefish * r13420 /haiku/trunk/Jamrules: BuildPlatformMain now uses the original DEFINES. Otherwise we e.g. get __HAIKU__ although compiling under R5.[13:16:10] <CIA-6> bonefish * r13421 /haiku/trunk/src/tools/rc/ (Jamfile R5Compatibility.h rc.cpp): Fixed the rc build. Sorry for the inconveniences.[13:20:04] *** WindowsUninstall has joined #haiku[13:24:25] *** Dr3w|Work has joined #haiku[13:27:55] *** Kernel86 has joined #haiku[13:29:55] *** Shawn_ has joined #haiku[13:30:02] *** Shawn_ has quit IRC[13:32:42] *** Leo42 has joined #haiku[13:37:46] <Dr3w|Work> Hey ppl.[13:39:42] <Sil2100> Hi[13:44:29] *** BGA has quit IRC[14:06:24] *** lizdeika has joined #haiku[14:24:50] *** Lebuzzer has joined #haiku[14:24:59] <Lebuzzer> tic[14:25:02] *** DaaT has joined #haiku[14:25:07] <Lebuzzer> hiya DaaT[14:25:19] <DaaT> hey Lebuzzer :)[14:25:25] <Lebuzzer> sup?[14:25:39] <Lebuzzer> holiday here today... Independence Day[14:25:46] <DaaT> yeah, 4th of july :)[14:25:48] <Lebuzzer> Lotsa fireworks[14:25:56] <DaaT> well, getting psyched about the trip tomorrow :)[14:26:01] <Lebuzzer> to where?[14:26:46] <DaaT> amsterdam[14:26:52] *** lizdeika has quit IRC[14:26:54] <DaaT> bunking with ithamar :P[14:27:48] <Lebuzzer> Wow, quite a trip![14:27:50] *** lizdeika has joined #haiku[14:27:59] <Lebuzzer> Don't know about that "bunking with Ithamar" part tho.. :-P[14:28:07] <Lebuzzer> Hope it's double-bunks[14:28:13] <DaaT> lol[14:28:20] <DaaT> you're just jealous[14:28:26] <DaaT> and Frank's going there as well[14:28:31] <Lebuzzer> Of the trip, yes...[14:28:43] <Lebuzzer> When you see Ithamar, tell him we have high hopes for USB under Zeta. :-)[14:29:28] <DaaT> you have probs with usb?[14:29:41] <tic> Lebuzzer was just thinking of you.[14:29:54] <Lebuzzer> tic See? Your ears were burnin'[14:29:57] <tic> :)[14:30:02] <tic> have a new binary for you..[14:30:06] <Lebuzzer> DaaT Yes[14:30:09] <Lebuzzer> tic yay![14:30:16] * Lebuzzer watches his mail[14:30:17] <DaaT> lunch, bbiab[14:30:24] <Lebuzzer> DaaT l8r[14:33:45] <tic> mail sent.[14:34:24] <Lebuzzer> unzipping![14:35:31] <Lebuzzer> woo hoo[14:35:34] <Lebuzzer> testing begins...[14:38:51] *** WindowsUninstall has quit IRC[14:40:31] *** WindowsUninstall has joined #haiku[14:42:11] <CIA-6> jackburton * r13422 /haiku/trunk/src/kits/app/MessageFilter.cpp: The default version of BMessageFilter::Filter() doesn't do anything. Small style changes, removed a todo.[15:01:21] <DaaT> Lebuzzer, you were talking about fireworks? http://www.nasa.gov/mission_pages/deepimpact/main/index.html?skipIntro=1[15:01:35] <Lebuzzer> I know...very cool...was reading about that today![15:01:39] <DaaT> better here: http://www.nasa.gov/mission_pages/deepimpact/multimedia/pia02121.html[15:03:19] <Lebuzzer> thanks[15:03:23] <DaaT> np[15:03:25] <DaaT> awesome pic[15:04:16] <DaaT> and who said those NASA guys didn't have a sense of humour? "This mission is truly a smashing success," said Andy Dantzler, director of NASA's Solar System Division[15:04:21] <DaaT> AH![15:05:15] * Zaranthos straps DaaT to a rocket and shoots him into the sky![15:05:23] <Zaranthos> *pop*[15:05:31] <DaaT> weeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeI CAN SEE MY HOUSE FROM HEREEEEEE[15:06:23] <Zaranthos> :-)[15:06:56] <DaaT> and you said i'd never get anywhere in life...[15:07:02] <Zaranthos> hehe[15:09:07] * DaaT sings "I believe I can flyyyyyyyyyy... i believe i can touch the skyyyyyyy"[15:10:11] * lizdeika gives DaaT a cookie just shut him up :)[15:10:23] <DaaT> :P[15:15:55] *** Lebuzzer has quit IRC[15:19:03] *** petterhj has joined #haiku[15:19:52] <DaaT> hey petterhj[15:23:35] <petterhj> hey DaaT :)[15:24:47] <DaaT> how's it going?[15:25:58] <petterhj> fine fine thank you :) just woke up[15:26:00] <petterhj> with you?[15:26:24] <DaaT> doing good thx[15:27:06] <petterhj> :)[15:30:17] *** Koki2 has joined #haiku[15:40:42] *** Koki has quit IRC[15:43:44] <CIA-6> phoudoin * r13423 /haiku/trunk/src/tests/kits/net/new_stack/interfaces/ether_drivers/ether_drivers.c:[15:43:44] <CIA-6> Added simple output support to Ethernet device(s) layer.[15:43:44] <CIA-6> Forgot to commit this long ago...[15:46:50] *** kr1stof has quit IRC[15:47:10] *** WindowsUninstall has quit IRC[15:47:13] *** Mandaryne is now known as PulkoMandy[15:49:45] *** xeD has quit IRC[15:56:20] *** lizdeika_ has joined #haiku[16:01:52] *** Koki2 has quit IRC[16:14:38] *** lizdeika has quit IRC[16:21:29] *** Lt_Henry has joined #haiku[16:22:17] *** xeD has joined #haiku[16:37:03] *** Konrad has quit IRC[16:37:12] *** Konrad77 has joined #haiku[16:39:31] *** emwe_ has joined #haiku[16:44:41] *** emwe has quit IRC[16:44:42] *** emwe_ is now known as emwe[16:45:05] *** DaaT has quit IRC[16:46:12] *** illissius- has joined #haiku[16:46:54] *** mahlzeit has joined #haiku[16:46:54] *** ChanServ sets mode: +o mahlzeit[16:47:17] <@JBurton> hey mahlzeit[16:47:21] <@mahlzeit> hi[16:51:37] <agentmumu> hey, ho[16:51:49] <@mahlzeit> who you calling a ho? :-)[16:51:56] <Dr3w|Work> hey![16:52:19] <agentmumu> mahlzeit: hehe[16:52:27] <agentmumu> always the one who asks[16:52:49] <@mahlzeit> or maybe you were doing the ramones?[16:54:31] *** illissius has quit IRC[16:56:25] <@mahlzeit> brb[16:58:42] <@JBurton> hey agentmumu[16:59:07] <agentmumu> huhu JBurton[16:59:52] <lizdeika_> agentmumu: cleanup your haiku.image repository :)[17:00:53] <Dr3w|Work> Grrr....[17:01:17] * Dr3w|Work is frustrated that he isn't good enough a programmer to really give a PowerPC port a kick in the ass...[17:01:18] <agentmumu> lizdeika_: you mean the old images? i prefer to keep them, maybe someone wants to try an old version :)[17:01:27] <@JBurton> :P[17:01:32] <Dr3w|Work> anyone here want an iMac mini for free?[17:01:35] <@JBurton> agentmumu erm... what about new images ? :P[17:01:57] <agentmumu> JBurton: i just booted beos :)[17:02:02] <agentmumu> i've[17:02:03] <@JBurton> :=)[17:02:26] * Dr3w|Work will buy someone an iMac mini...[17:02:42] <lizdeika_> choose me[17:02:53] *** Konrad77 has quit IRC[17:02:56] *** lizdeika_ is now known as lizdeika[17:03:49] <agentmumu> but I have to do a new checkout, I ran rm -rf Haiku instead of rm -rf /Haiku (Haiku = haiku svn repository and /Haiku = Haiku partition)[17:03:55] <Dr3w|Work> lizdeika: could you get Haiku booted to app server and MiniTerminal in 10 weeks on it? :)[17:04:06] <agentmumu> so it will take some time[17:04:07] <lizdeika> not at all[17:04:49] <Dr3w|Work> :)[17:05:03] <agentmumu> JBurton: axeld should show up here this night, right?[17:06:03] <@JBurton> agentmumu apparently :P[17:06:20] <@JBurton> agentmumu ouch, that hurts :)[17:06:38] <agentmumu> JBurton: beside a not working libroot.so i have haiku compiling under linux[17:06:45] <agentmumu> i guess it's a linking problem[17:06:45] <@JBurton> I read the news item[17:06:49] <@JBurton> that's great :)[17:07:01] *** Konrad has joined #haiku[17:07:04] <@JBurton> how did you do for bash ?[17:07:20] <@JBurton> agentmumu I think it could be more a libc issue[17:07:38] <@JBurton> if it was a linking problem, it shouldn't link[17:07:40] <@JBurton> no ? :)[17:07:53] <agentmumu> oh, bash is still missing, i've been at my girlfriends place over the weekind, but bash shouldn't be a big problem[17:08:03] <agentmumu> JBurton: maybe it links to some linux lib[17:08:17] <agentmumu> which is then missing on haiku[17:08:25] <@JBurton> agentmumu could be[17:08:45] <@JBurton> agentmumu but I mean: how did you manage to run the Terminal without bash ? :P[17:08:56] <agentmumu> this is what makes me wonder:[17:08:57] <agentmumu> schmidp@telomer ~/work/openresearch/haiku/haiku_svn/objects/x86.R1/system $ ldd kernel.so[17:09:01] <agentmumu> linux-gate.so.1 => (0xffffe000)[17:09:03] <agentmumu> objects/x86.R1/system/kernel/linkhack.so => not found[17:09:09] <agentmumu> JBurton: i copied over bash from an old image[17:09:12] <@JBurton> ah okay[17:09:20] <agentmumu> i also had to copy over libroot.so[17:09:21] *** ablyss has left #haiku[17:09:30] <@JBurton> hmmm[17:09:32] <Dr3w|Work> agentmumu: is this the normal BeOS terminal running on Haiku?[17:09:44] <agentmumu> Dr3w|Work: MiniTerminal? no[17:09:59] <@JBurton> linkhack.c is a file which is compiled, actually[17:10:04] <@JBurton> but it's empty[17:10:04] <Dr3w|Work> no, I mean are you trying to get Terminal running on BeOS/[17:10:49] <Dr3w|Work> crap. I mean, are you trying to get "Terminal" running, rather than MiniTerminal[17:10:51] <agentmumu> JBurton: maybe there are just some "if os!= beos" rules in some libroot.so jamfiles which are not working yet[17:10:59] <@JBurton> yeah it could be[17:11:41] <Dr3w|Work> agentmumu: you say you have Haiku compiling under Linux? Does it work?[17:11:45] <Dr3w|Work> I am trying the same now.[17:11:51] <Dr3w|Work> I get *massive* problems :)[17:11:57] <@JBurton> Dr3w|Work you should read agentmumu's blog, really :)[17:12:08] <Dr3w|Work> URL?[17:12:19] <agentmumu> Dr3w|Work: www.schmidp.com/blog[17:12:21] <@JBurton> www.agentmumusblog.com[17:12:22] <@JBurton> :P[17:12:24] <agentmumu> hehe[17:12:47] <agentmumu> i'll write a howto after i have a complete working image[17:12:57] <Dr3w|Work> agentmumu: hey![17:13:02] <Dr3w|Work> we work on a similar goal then[17:13:07] <Dr3w|Work> I have a read me somewhere...[17:13:28] <@JBurton> agentmumu you used oliver's compiler, right ?[17:13:33] <Dr3w|Work> http://www.h2o.demon.co.uk/haiku/HaikuPPCBuildReadMe.txt[17:13:33] <agentmumu> JBurton: yes[17:13:37] <@JBurton> ok[17:14:06] <Dr3w|Work> agentmumu: what I did was aiming at getting a PPC build up, but its on Linux.[17:14:11] <agentmumu> Dr3w|Work: i know your readme, but you should have a look at olivers readme, on how to setup a a cross compiler[17:14:43] <agentmumu> oh, you've updated it[17:14:46] <Dr3w|Work> Umm.. I wanted to to compile PPC binaries though.[17:14:47] <agentmumu> mom[17:15:14] <agentmumu> Dr3w|Work: well, you still want to compile your binaries for beos and not for linux[17:15:59] <Dr3w|Work> oh see, so I was compiling Haiku code on Linux and as it wasn't cross-compiled Linux/PPC->Haiku/PPC then it was still compiling Linux binaries?[17:16:18] <agentmumu> yes[17:16:19] <Dr3w|Work> I didn't realise that at all.[17:16:26] <Dr3w|Work> where is olivers doc?[17:16:56] <agentmumu> Dr3w|Work: in svn buildtools[17:17:01] <agentmumu> INSTALL-as-cross-compiler-on-LINUX[17:17:14] <CIA-6> rudolfc * r13424 /haiku/trunk/src/add-ons/accelerants/nvidia/engine/nv_acc_dma.c: added some comments about the NV18 speedfix and resetting registers on NV28.[17:19:03] <Dr3w|Work> can you point the path? /haiku/trunk/ ...?[17:19:56] <agentmumu> buildtools/trunk/[17:20:07] <agentmumu> not haiku[17:22:23] *** Begasus has joined #haiku[17:24:00] <Dr3w|Work> ok - got it![17:24:03] <Dr3w|Work> this is great![17:24:43] <@JBurton> bye all[17:24:58] <Dr3w|Work> agentmumu: great day to point this all out to me - Monday and Wednseday nights are my BeOS nights away from my gf :)[17:25:02] *** JBurton has quit IRC[17:25:11] <Dr3w|Work> I am going to update my doc tonight so its right.[17:25:21] <Dr3w|Work> oh hold on.[17:28:20] <Dr3w|Work> can I just create a --target=ppc-pc-beos?[17:29:21] <agentmumu> i don't know, i think haikus build system still depends on some beos headers[17:29:38] <agentmumu> i would use olivers howto[17:29:55] <agentmumu> Dr3w|Work: or maybe i got you wrong[17:30:03] <Dr3w|Work> but I don't want to use --taret=i586-pc-beos on a PowerPC system though?[17:30:15] <agentmumu> no of course not[17:30:38] <agentmumu> i hope there is a target like ppc-pc-beos[17:31:01] <Dr3w|Work> I have a feeling its going to use PEF if there is though...[17:31:08] <Dr3w|Work> not elf.[17:31:41] <agentmumu> wel[17:32:01] <agentmumu> hm, i don't know anything about beos on ppc, so i don't know :)[17:32:29] <Dr3w|Work> :)[17:33:36] <Ghostride> agentmumu: Do you have a rss feed of your blog?[17:36:46] *** mmu_man has joined #haiku[17:36:47] *** ChanServ sets mode: +o mmu_man[17:43:32] <agentmumu> Ghostride: yes, http://www.schmidp.com/blog/index.php?/feeds/index.rss2[17:46:06] <Ghostride> *subscribes*[17:46:11] <agentmumu> does anyone know if there is a reason for libhack.so?[17:49:38] *** fyysik has joined #haiku[17:49:52] <Dr3w|Work> agentmumu: just posted a message to binutils mailing list asking about their --target option.[17:50:02] <Dr3w|Work> I wonder if you can jsut put anything in there...[17:50:51] *** xeD has quit IRC[17:51:50] <Dr3w|Work> answer=no.[17:52:30] *** ChrisK077 has joined #haiku[17:57:03] <Dr3w|Work> think I have to create some files in binutils first.[17:57:10] <Dr3w|Work> agentmumu: are you around tonight?[18:00:58] *** Mazon has quit IRC[18:01:36] *** Mazon has joined #haiku[18:01:47] *** Dr3w|Work has quit IRC[18:30:02] *** Methe has joined #haiku[18:31:45] *** Methe has quit IRC[18:31:57] *** Methe has joined #haiku[18:33:40] *** SiCuTDeUx has joined #haiku[19:03:09] *** DarthVader has joined #haiku[19:03:53] <tic> 'elo[19:04:01] <tic> jean-louis?[19:04:49] *** TLF has joined #haiku[19:05:00] <agentmumu> DarthVader: yes[19:05:18] <DarthVader> tic: hello[19:05:27] <agentmumu> [Dr3w|Linux]: yes[19:06:34] <tic> moo[19:08:21] *** aGNUstic has joined #haiku[19:09:07] *** mmu_man has quit IRC[19:09:57] <aGNUstic> Hello guys[19:10:04] <@mahlzeit> hi aGNUstic[19:10:35] *** mmu_man has joined #haiku[19:10:37] *** ChanServ sets mode: +o mmu_man[19:15:48] *** crash|_ has joined #haiku[19:16:31] <aGNUstic> mahlzeit, mmu_man, Begasus, Karina, Procton, and Zaranthos - I recognize those names[19:16:41] <@mahlzeit> that's because we are famous![19:16:58] <tic> I thought I'd be recognized too :([19:17:01] <@mmu_man> "f*ck me I'm famous!" ?[19:17:09] <Sil2100> ^^[19:17:10] <tic> haha[19:17:19] <@mahlzeit> mmu_man: we have a shrine for you over here[19:17:21] <tic> Now let's see if I can get Subversion up and running...[19:17:39] <@Korli> mahlzeit you were famous :)[19:17:46] <@mmu_man> I think it's the name of a song[19:17:57] <@mmu_man> from a stupi french dj[19:19:10] <@mahlzeit> ah, so you were quoting? (hence the quotes, i suppose)[19:19:31] <@Korli> mahlzeit :p[19:20:08] <aGNUstic> So how is the Haiku project going?[19:20:20] <aGNUstic> Legitimate question.[19:21:23] <@mahlzeit> it's not finished yet[19:21:28] <@mahlzeit> that's all i know[19:21:38] <aGNUstic> Darn.[19:22:07] <@mahlzeit> but progress is steady[19:22:40] <aGNUstic> I can handle that. I loved BeOS.[19:22:57] *** oco has joined #haiku[19:25:19] *** emwe has quit IRC[19:26:23] <aGNUstic> ...[19:27:06] <aGNUstic> bomb[19:28:14] <tic> Let's see if I can figure out svn+ssh://...[19:32:17] <aGNUstic> Later guys. It was nice to see name I know.[19:32:32] *** aGNUstic has left #haiku[19:32:58] *** Korli has quit IRC[19:34:39] <@mahlzeit> who was that?[19:36:04] *** xeD has joined #haiku[19:37:53] <Begasus> asus> a long time BeOS user .. ;)[19:38:32] <@mahlzeit> yeah apparently[19:52:36] *** DaaT has joined #haiku[19:55:23] *** Dr_Evil has joined #haiku[20:00:23] *** Konrad has quit IRC[20:02:52] *** ChrisK077 has quit IRC[20:11:11] *** petterhj has quit IRC[20:12:04] *** xeD has quit IRC[20:13:46] *** dos4gw has quit IRC[20:39:17] *** Korli has joined #haiku[20:39:17] *** ChanServ sets mode: +o Korli[20:39:36] *** WindowsUninstall has joined #haiku[20:56:33] <[Dr3w|Linux]> hey folks.[20:57:07] <@mahlzeit> hey[21:04:53] *** xeD has joined #haiku[21:08:32] *** ChrisK077 has joined #haiku[21:09:47] *** mmu_man has quit IRC[21:36:00] *** Pietmey has joined #haiku[21:37:52] *** xeD has quit IRC[21:46:38] *** ChrisK077 has quit IRC[21:48:11] <PulkoMandy> +++[21:48:20] *** PulkoMandy has quit IRC[21:53:12] *** Pascal has joined #haiku[21:54:02] *** Leo44 has joined #haiku[21:54:34] *** Leo42 has quit IRC[21:55:55] *** Pascal has left #haiku[21:58:26] *** petterhj has joined #haiku[22:00:53] <[Dr3w|Linux]> geist: are you around?[22:00:59] <@geist> yo[22:02:10] <[Dr3w|Linux]> do you know - if I wanted to compile binutils to target powerpc-pc-haiku or powerpc-pc-beos as elf, do I create a BFD for it?[22:02:26] <@geist> I do not know[22:02:30] <[Dr3w|Linux]> ah OK.[22:02:31] <@geist> I do not work on haiku[22:02:57] <@geist> thoguht I was thinking about trying to build a linux toolchain and see if they made the build system work on linux yet[22:03:35] <[Dr3w|Linux]> Yeah, thats basically what I am doing, agentmmu is doing the same, but I want to see on Linux/PPC.[22:03:36] <@mahlzeit> the resource compiler doesn't work on linux[22:04:06] <agentmumu> mahlzeit: it does now :)[22:04:10] <@mahlzeit> yay![22:04:16] <agentmumu> but![22:04:16] <[Dr3w|Linux]> hey agentmumu :)[22:04:18] <@mahlzeit> how did you do it?[22:04:28] <agentmumu> mahlzeit: i used cosmoe :)[22:04:32] * scanty wonders about getting haiku going on SPARC[22:04:41] <@mahlzeit> heh nice[22:04:59] <[Dr3w|Linux]> agentmumu: did you see my q. to geist re. BFD?[22:05:03] <agentmumu> but entry_ref doesn't work with cosmoe, so i had to hack it a bit[22:05:16] <agentmumu> [Dr3w|Linux]: i don't know what bfd means?[22:05:33] <[Dr3w|Linux]> BDF seems to be the way bintuils targets different platforms.[22:07:33] <scanty> bfd is binary file descriptor library[22:07:45] *** Korli has quit IRC[22:07:51] <[Dr3w|Linux]> geist: it used to build on Linux about a year ago.[22:08:08] <[Dr3w|Linux]> no, about 2 years ago.[22:08:14] <[Dr3w|Linux]> when david reid was around.[22:10:11] <scanty> (or big fucking deal, as the story goes, IIRC)[22:12:30] *** TLF has quit IRC[22:13:12] *** Pietmey has quit IRC[22:15:58] <agentmumu> i hope axeld can help me getting libroot linking correctly under linux[22:16:30] <[Dr3w|Linux]> scanty: what does that mean?[22:17:20] <scanty> what, bfd?[22:18:27] <[Dr3w|Linux]> no (big fu.ing deal)[22:19:20] <scanty> oh...[22:19:29] <scanty> it was back in the day when RMS and that other dude were planning out BFD...[22:19:51] <scanty> and I think the other guy was complaining that it would be very difficult to do.[22:19:58] <scanty> so RMS says "big fucking deal"[22:20:03] <scanty> and that was that :)[22:20:10] <[Dr3w|Linux]> ah I see :)[22:22:31] <scanty> leave it to GNU software for weird names/stories ;)[22:24:44] <scanty> [Dr3w|Linux], are you the same Dr3w I spoke with on BeShare some months ago WRT to haiku and SPARC?[22:25:09] <[Dr3w|Linux]> Hmmm... can I just change configure.tgt to use elf instead of coff...[22:31:36] *** muqo has quit IRC[22:39:46] <CIA-6> bonefish * r13425 /haiku/trunk/src/bin/Jamfile: Forget to check in this one. Sorry. shutdown should build again now.[22:44:16] <[Dr3w|Linux]> agentmumu: are you around?[22:46:01] <agentmumu> yes[22:47:24] <[Dr3w|Linux]> I love you x.[22:47:35] <[Dr3w|Linux]> :) joke - its OK, I worked out what I was going to ask :)[22:47:59] <agentmumu> hehe, ok :)[22:49:37] <[Dr3w|Linux]> I think my binutils src dir was messed up from a previous configure. I deleted it and started again and it seems to be working now.[22:50:42] <agentmumu> :)[22:59:17] <[Dr3w|Linux]> Hmmm.... do I hack powerpc-pc-beos from coff to elf, or do I add powerpc-pc-haiku as elf?[23:01:58] <Sil2100> See you later everyone[23:02:07] *** Fanskapet has joined #haiku[23:02:07] *** Sil2100 has quit IRC[23:04:10] *** ilorvi is now known as as[23:04:46] *** as is now known as ilorvi[23:04:48] *** mahlzeit has quit IRC[23:06:01] <CIA-6> jackburton * r13426 /haiku/trunk/src/kits/app/ (Handler.cpp Looper.cpp): If a filter was added to a handler which belonged to a looper, the filter's looper was never initialized. BHandler::SetLooper() now also sets the looper for every filter (moved from BLooper::AddHandler)[23:17:46] *** Begasus has quit IRC[23:23:47] *** miqlas has joined #haiku[23:26:19] <miqlas> Re![23:31:08] <CIA-6> jackburton * r13427 /haiku/trunk/src/kits/interface/Menu.cpp: Don't assume we have an owner. This fixes a crash in MenuWorld, but needs more investigation[23:38:32] <Dr_Evil> is anyone here using haiku media kit right now?[23:39:38] <miqlas> MediaKit-Beta3 ?[23:40:03] <Dr_Evil> I don't think we have a beta 3[23:41:47] <CIA-6> axeld * r13428 /haiku/trunk/src/kits/app/ (Message.cpp app.src dano_message.cpp dano_message.h):[23:41:47] <CIA-6> *Very* basic read-only Dano/Zeta message support. Only briefly tested.[23:41:47] <CIA-6> Not really suited for message sending/retrieval (because target stuff is[23:41:47] <CIA-6> missing), but should work more or less with disk based messages.[23:41:57] <miqlas> ftp.beos.spb.ru/Incoming/.....[23:43:14] <miqlas> ftp://ftp.beos.spb.ru/incoming/MediaKit(beta3)Fixed.pkg[23:43:19] <miqlas> whats this Kit?[23:43:40] *** WindowsUninstall has quit IRC[23:43:52] *** axeld has joined #haiku[23:43:52] *** ChanServ sets mode: +o axeld[23:45:37] *** JBurton has joined #haiku[23:45:37] *** ChanServ sets mode: +o JBurton[23:45:46] <@JBurton> yo[23:46:06] <@axeld> Hi JBurton[23:46:41] *** fps-away has joined #haiku[23:48:14] <@JBurton> hi axeld[23:49:04] <Dr_Evil> hi axeld[23:49:14] <Dr_Evil> miqlas I don't know[23:49:19] <@axeld> Hi Dr_Evil[23:49:34] <@JBurton> axeld the MenuWorld sample code application is working (more or less) in haiku... should I put it in the tree or it doesn't fit there ?[23:50:06] <@JBurton> it still crashes in a BButton method, though[23:50:10] <miqlas> Dr_Evil, thx.[23:50:35] <@axeld> JBurton: dunno, but it's a test app, or isn't it? :-)[23:50:51] <@JBurton> yeah[23:50:58] <@JBurton> it is[23:50:59] <[Dr3w|Linux]> axeld: hey![23:51:27] <@axeld> Hi [Dr3w|Linux][23:52:50] <[Dr3w|Linux]> axeld: agentmumu pointed out that when I am trying to build Haiku on powerpc, I am not setting my binutils and gcc to target BeOS, I am just compiling the source tree on Linux as linux binaries. As Haiku on PPC will use elf, I need to patch binutils etc. Should I just hack the existing stuff from coff to elf, or should I write a patch so there is a powerpc-pc-haiku that is elf?[23:53:27] <[Dr3w|Linux]> what/how did you do it?[23:53:37] <@axeld> [Dr3w|Linux]: Well, I build on Linux/PPC :-)[23:53:44] <[Dr3w|Linux]> Yes, I am on linux PPC.[23:53:56] <@axeld> And Linux uses ELF[23:54:01] <@axeld> not COFF[23:55:16] <[Dr3w|Linux]> agentmumu: are you still around?[23:55:27] <agentmumu> [Dr3w|Linux]: yes[23:55:39] <[Dr3w|Linux]> yes, but the files that I am producing are just Linux binaries that I can run in Linux.[23:55:41] <miqlas> Zeta activation : http://techriots.com/zeta/screen17.jpg :>[23:56:48] <@axeld> [Dr3w|Linux]: Right now, only the kernel and the boot loader can be built under Linux correctly, AFAICT[23:57:28] <agentmumu> axeld: i managed to build nearly the wohle, tree, only libroot.so doesn't work and bash doesn't compile yet (but this not a big problem)[23:57:32] <agentmumu> -,[23:57:34] <Dr_Evil> miqlas just use the provided tools, like objdump and DiskProbe[23:57:55] <[Dr3w|Linux]> agentmumu: you know my problem right?[23:59:45] *** [Dr3w|Linux] is now known as McCall[23:59:52] <agentmumu> [Dr3w|Linux]: maybe I was wrong that you need a cross compiler, because as I found out haiku on ppc doesn't want to be binary compatible