| 00:00.03 | dc_ibook | ah cool. |
| 00:00.35 | PaxAnima | and waiting around for cworth to find the bug with xstroke and minipredict |
| 00:00.40 | moray | dc_ibook: I wasn't sure at the time what mallum was scooping up, but I'd forgotten about those sheets - if pb gets back I can check they're not still in the car I suppose, though I didn't notice them when I was taking my stuff |
| 00:01.37 | moray | dc_ibook: mallum said something about throwing away the maps we'd used before, but those are still in my bag, so I'm worried he had my keys instead :/ |
| 00:04.51 | dc_ibook | moray: mallum's just been looking through the bag of rubish and can't see them. |
| 00:05.16 | moray | ha, well it's good if they're not there :) |
| 00:05.52 | moray | though I'm disappointed pb hasn't appeared yet - is he *sleeping* or something? |
| 00:06.16 | PaxAnima | who needs sleep? |
| 00:06.22 | moray | quite |
| 00:06.31 | dc_ibook | heh |
| 00:06.43 | moray | although the answer would appear to be me - I'm stuck between needing to eat something and needing to get up early tomorrow morning |
| 00:07.33 | moray | dc_ibook: so did you work out what that programme was about? |
| 00:08.08 | moray | dc_ibook: and is mallum still having his hair eaten by pugs? |
| 00:10.21 | moray | hm, perhaps dc_ibook's been eaten by a pug now? |
| 00:11.30 | PaxAnima | most likely |
| 00:18.33 | *** join/#gpe cmarqu (colin@buug.mind.de) |
| 00:18.38 | *** join/#gpe walters (walters@verbum.org) |
| 00:28.56 | dc_ibook | ! |
| 00:29.12 | dc_ibook | cmarqu: hey |
| 00:29.59 | moray | hi cmarqu, walters |
| 00:39.55 | dc_ibook | sleep time now |
| 00:45.52 | PaxAnima | spung: got the magic line to convert oo.org dictionaries |
| 00:46.09 | spung | cool |
| 00:46.26 | PaxAnima | perl -ne ' s,/.*,,; next if ( /[.1-9]/ && /.{0,1}/ ); print;' nb_NO.dic | sort | gzip > dict.gz |
| 00:46.31 | PaxAnima | something like that |
| 00:46.43 | spung | why gzip? |
| 00:46.57 | PaxAnima | save space? |
| 00:47.09 | spung | jffs compresses already |
| 00:47.24 | PaxAnima | true |
| 00:47.43 | PaxAnima | anyway, the dictionary should be compressed with dict_comp, a util I made for libdictionary... |
| 00:54.36 | PaxAnima | what's the default locale when using gpe? |
| 00:55.20 | PaxAnima | spung: where are you from? |
| 00:55.38 | spung | Portugal |
| 00:55.55 | PaxAnima | threre is a portugese dictionary for openoffice.org |
| 00:56.51 | PaxAnima | http://ftp.cyberfenix.net/pub/openoffice/contrib/dictionaries/download_dictionary.html |
| 00:58.41 | PaxAnima | and it's much smaller than the norwegian one |
| 00:59.57 | spung | that could be a bad sign |
| 01:00.09 | spung | i hope it's not brazilian portuguese |
| 01:00.12 | PaxAnima | but the norwegian one was veeeery big |
| 01:00.21 | PaxAnima | nope |
| 01:00.36 | PaxAnima | there are two portugese ones; one for brazil and on for portugal |
| 01:00.50 | spung | everyone spells it "portugese" i wonder why :) it's portuguese :) |
| 01:01.32 | PaxAnima | ;) |
| 01:01.45 | PaxAnima | true |
| 01:03.53 | PaxAnima | spung: if you want to try minipredict with the portugUese one, you must overwrite /usr/share/dict/38k_dict.cd |
| 01:04.16 | spung | :D |
| 01:04.25 | spung | ok, i'll try it out |
| 01:05.22 | spung | my battery's max is 84% any idea why? :(((((((( |
| 01:05.35 | PaxAnima | what model do you have? |
| 01:09.11 | spung | 3760 |
| 01:09.19 | PaxAnima | hm.. dunno |
| 01:09.42 | spung | i heard about bateries dying once |
| 01:10.21 | spung | don't remember which model |
| 01:19.39 | PaxAnima | spung: are you trying minipredict now? |
| 01:22.12 | spung | yes |
| 01:22.16 | spung | how does it work? |
| 01:22.35 | PaxAnima | have you started it? |
| 01:22.59 | spung | oh nevermind |
| 01:23.34 | PaxAnima | after you've started (add the mushroom to the mbdock), it catches all input from xbkd/xstroke |
| 01:24.00 | spung | it doesn't work with |
| 01:24.07 | spung | rxvt + xstroke |
| 01:24.31 | PaxAnima | there us a bug with xstroke.. |
| 01:24.44 | PaxAnima | you must start minipredict _after_ your app to use it with xstrkoe |
| 01:25.20 | spung | not really |
| 01:25.24 | spung | works fine with gpe-edit |
| 01:25.28 | spung | just not rxvt |
| 01:25.48 | PaxAnima | hmm.. ok: for _some_ apps, apparently... |
| 01:26.05 | PaxAnima | anyway, it's there (and it's annoying me) |
| 01:26.30 | spung | ahh |
| 01:26.33 | spung | different bugs |
| 01:26.42 | spung | i see the one you mention now |
| 01:26.49 | spung | the rxvt seems a bit different |
| 01:27.22 | PaxAnima | hopefully cworth finds out why |
| 01:27.54 | PaxAnima | are you using the english dictionary, or the one from oo.org? |
| 01:28.57 | spung | english |
| 01:29.03 | PaxAnima | ok |
| 01:42.30 | PaxAnima | oh well, time to get some sleep |
| 01:42.58 | PaxAnima | spung: if you find any bugs, or have suggestions, please contact me here on IRC or mail me at paxanima@handhelds.org |
| 01:43.11 | spung | ok |
| 01:43.27 | PaxAnima | good night |
| 02:52.20 | *** join/#gpe walters (walters@verbum.org) |
| 03:15.05 | srd | how do I get rid of the awful grey titlebar in gpe? |
| 03:15.19 | srd | to something better looking, like the kde-like one in the screenshots on the gpe homepage |
| 03:27.58 | *** join/#gpe walters (walters@verbum.org) |
| 07:51.35 | *** join/#gpe florian (~fuchs@dialin-145-254-095-111.arcor-ip.net) |
| 08:13.21 | *** join/#gpe PaxAnima (~petter@v058b.studby.ntnu.no) |
| 08:13.42 | PaxAnima | morning |
| 08:14.32 | pigeon | PaxAnima: Gee another day for you already? :) |
| 08:14.59 | PaxAnima | couldn't sleep, so I've slept 2,5 hours tonight :/ |
| 08:15.20 | PaxAnima | almost couldn't sleep, that is |
| 08:16.10 | PaxAnima | and I don't see why... I had maybe 2 cups of coffee, and that's _nothing_ compared to my normal dose ;) |
| 08:53.41 | *** join/#gpe dc_ibook (~dctanner@pc-80-193-218-21-hw.blueyonder.co.uk) |
| 10:00.55 | *** join/#gpe drkness (~sharath@nwkea-http-2.sun.com) |
| 10:27.08 | *** join/#gpe eah (~seeah@pro.proact.se) |
| 10:39.12 | *** join/#gpe pierre (~pierre@AClermont-Ferrand-201-1-4-141.abo.wanadoo.fr) |
| 10:46.04 | *** join/#gpe drkness (~sharath@nwkea-http-1.sun.com) |
| 11:28.55 | *** join/#gpe colin_werk (www@vws1.kens.com) |
| 11:30.33 | colin_werk | Hi. |
| 11:30.43 | PaxAnima | hi |
| 11:31.06 | cmarqu_werk | PaxAnima: You were missed at Fosdem. |
| 11:31.21 | PaxAnima | too much work and travel to get to belgium |
| 11:31.37 | cmarqu_werk | Yeah :/ |
| 11:32.04 | PaxAnima | maybe next time? |
| 11:32.19 | cmarqu_werk | Hope so. It was fun. |
| 11:32.38 | cmarqu_werk | Are there cheap flights to England from your place? |
| 11:33.08 | PaxAnima | hmm.. not too cheap |
| 11:33.33 | cmarqu_werk | :/ |
| 11:34.08 | PaxAnima | but I plan to travel to england some time this year, I hope |
| 11:34.18 | cmarqu_werk | There are flights from Germany to England for as low as EUR 0.99. Plus some fees. |
| 11:35.07 | PaxAnima | wow |
| 11:35.32 | cmarqu_werk | We had the idea of a GPE meeting in Cambridge. Still need to plan that of course. |
| 11:35.48 | PaxAnima | ah |
| 11:36.36 | PaxAnima | would be cool |
| 11:39.49 | cmarqu_werk | Indeed. |
| 13:34.23 | *** join/#gpe mallum (~mallum@pc-80-193-218-21-hw.blueyonder.co.uk) |
| 13:35.02 | mallum | hey |
| 13:53.12 | cmarqu_werk | Hi mallum. Had a good trip home? |
| 14:00.40 | PaxAnima | hey, mallum |
| 14:03.40 | moray | hello mallum, manage to get dc shipped off? |
| 14:09.27 | mallum | hi, sorry AFK |
| 14:10.02 | mallum | cmarqu_werk: yes got home fine thanks to pb's driving and moray navigation skills. Thanks btw for driving too ! |
| 14:10.17 | mallum | moray: yes, I put him in a cab early this morning |
| 14:10.31 | mallum | moray: did you find your key thing ? |
| 14:11.06 | moray | mallum: I think pb's going to check in his car some time - if it's not there I probably dropped it, which would be irritating but hardly a disaster |
| 14:11.47 | moray | mallum: would just mean that the people who were there don't get to have quite as many l33t signatures on their gpg keys... ;) |
| 14:18.48 | mallum | moray: doh! |
| 14:20.59 | moray | mallum: ... while, presumably, I look just as "l33t" as I would have done, getting all those people's signatures nonetheless -- although I hear there's some cryptographic thing you can use this stuff for too? ;) |
| 14:24.17 | mallum | hehe |
| 14:31.24 | cmarqu_werk | Argh, what happened to my other self? |
| 14:39.12 | cmarqu_werk | moray: Ah, so I'll be the only one with your signature, or did you lose my single slip too? |
| 14:43.14 | moray | cmarqu_werk: I've signed yours and re-uploaded; I can email it to you if you like |
| 14:43.27 | moray | cmarqu_werk: I had slips from a few other people too, but basically yes |
| 14:45.12 | moray | cmarqu_werk: if they don't turn up, I'll still have signed several people's keys who were there, so should only be two steps from anyone assuming those people sign everyone's keys |
| 14:49.28 | cmarqu_werk | moray: Okay, cool. |
| 14:50.15 | cmarqu_werk | What's uncool is that I didn't verify Russell Cokers ID. A direct Australia connection would have been nice... |
| 14:50.43 | moray | cmarqu_werk: hm. I'd already signed his key before |
| 14:54.14 | *** join/#gpe dc_ (~dc@modem-43-58-60-62.vip.uk.com) |
| 14:54.20 | dc_ | mallum: hey |
| 14:54.25 | dc_ | mallum: I made it back at alst |
| 14:54.29 | dc_ | last. |
| 14:55.32 | moray | dc_: any problem, or it's just slow to get back? |
| 14:55.45 | dc_ | moray: just slow trasins |
| 15:00.26 | cmarqu_werk | Hi dc_ |
| 15:00.34 | dc_ | cmarqu_werk: hey |
| 15:00.41 | dc_ | cmarqu_werk: oh, btw. |
| 15:00.54 | dc_ | cmarqu_werk: I get the progressive zooming working :) |
| 15:00.59 | dc_ | cmarqu_werk: it looks really neat |
| 15:01.05 | dc_ | cmarqu_werk: I'll commit it later. |
| 15:01.14 | cmarqu_werk | dc_: Cool. |
| 15:02.00 | cmarqu_werk | BBL |
| 15:02.04 | *** part/#gpe cmarqu_werk (www@vws1.kens.com) |
| 15:04.33 | mallum | dc_: cool :-) |
| 15:04.39 | mallum | dc_: no school ? |
| 15:05.49 | dc_ | yep :) |
| 15:07.02 | mallum | hehe |
| 15:07.05 | dc_ | authough I must write this essay for tomorrow :\ |
| 15:07.16 | mallum | dc_: Im cleaning up the theme engine at the moment |
| 15:07.21 | dc_ | mallum: good :D |
| 15:07.54 | mallum | dc_: it should be pretty easy to port alot of the metacity city themes then ;-) |
| 15:08.07 | dc_ | great. |
| 15:08.22 | mallum | dc_: http://art.gnome.org/show_theme.php?themeID=220&category=metacity could be quite easy ( and nice-ish ) |
| 15:09.02 | dc_ | ah, good old micro gui |
| 15:09.03 | *** join/#gpe mccarthy (~mccarthy@granular.che.pitt.edu) |
| 15:09.15 | mccarthy | is pb back from FOSDEM now? |
| 15:09.25 | mccarthy | ibot seen pb |
| 15:09.25 | | mccarthy: i haven't seen 'pb' |
| 15:09.30 | mccarthy | ibot seen pb_ or pb__ |
| 15:09.31 | | i haven't seen 'pb_ or pb__', mccarthy |
| 15:09.32 | dc_ | mccarthy: yes.. |
| 15:09.36 | mccarthy | ibot seen pb_ |
| 15:09.36 | | pb_ <~pb@cte13.ulb.ac.be> was last seen on IRC in channel #handhelds.org, 1d 2h 59m 3s ago, saying: 'hi zecke_'. |
| 15:09.44 | dc_ | mccarthy: probly just sleeping |
| 15:09.44 | mccarthy | ibot seen pb__ |
| 15:09.44 | | pb__ <~pb@2002:5160:45ef:0:20a:95ff:fe76:cd6c> was last seen on IRC in channel #gpe, 15d 14h 13m 52s ago, saying: 'yes, that's true.'. |
| 15:09.49 | mccarthy | dc_: ok |
| 15:10.33 | mccarthy | dc_: you any good at C coding nowadays? (IIRC you learned just as you were getting started with the GPE project, which means you are probably better than me now ;) |
| 15:10.47 | dc_ | mccarthy: hehe. |
| 15:11.01 | dc_ | mccarthy: why do you ask? |
| 15:11.15 | dc_ | mccarthy: but yes, you are correct. |
| 15:11.38 | dc_ | mccarthy: beware though, I am an uber leet c coder now I have a c pocket reference. |
| 15:11.42 | mccarthy | dc_: I am working on doing something new in some work related code, but I am not good at it.... |
| 15:12.05 | dc_ | mccarthy: what kind of stuff? |
| 15:12.24 | mccarthy | dc_: this kind "ev->flags |= FLAG_ALARM;" |
| 15:12.57 | mccarthy | I am wondering if the FLAG_ALARM can be any old integer, and how *large* of an integer, if so |
| 15:13.42 | mccarthy | what do you call that stuff.... bit operators? |
| 15:14.42 | PaxAnima | yup |
| 15:14.58 | dc_ | hrm |
| 15:15.47 | dc_ | does most stuff work on the 39xx or is that a pb_ question? |
| 15:18.14 | moray | dc_: I'm fairly sure 'most stuff works', but pb would know what doesn't yet |
| 15:18.26 | dc_ | quite |
| 15:18.51 | moray | http://www.handhelds.org/projects/h3900.html was last updated a couple of weeks ago |
| 15:26.10 | *** join/#gpe asdls (~tima@vp164217.hk.uac1.hknet.com) |
| 15:28.01 | *** join/#gpe cmarqu (colin@buug.mind.de) |
| 15:32.15 | *** join/#gpe prpplague (~JoeBob1@12.148.134.9) |
| 15:38.04 | mccarthy | PaxAnima: any good refs on bi operators that would answer my question? |
| 15:38.12 | mccarthy | s/bi/bit |
| 15:38.21 | PaxAnima | hmm.. what's the problem? |
| 15:39.33 | mccarthy | Without getting into details, I need to know if I do "flags |= NUM" what *type* NUM should be and how *big* it can be |
| 15:39.50 | PaxAnima | flags and NUM should be of the ame type |
| 15:40.31 | PaxAnima | a |= b is just a bitwise OR of a and b |
| 15:40.37 | cmarqu | mccarthy: FWIW, I found a cool set of bit op macros on snippets.org. |
| 15:41.03 | cmarqu | This makes for "speaking" operations. |
| 15:46.23 | mccarthy | PaxAnima: Let me explain details quickly... (then you tell me if this will work, please ;) |
| 15:46.59 | mccarthy | In a simulation, I have a bunch of (say, N) Particles.... |
| 15:47.18 | mccarthy | I need to *remember* from 1 time step to the next which particles are touching each other... |
| 15:48.53 | mccarthy | I was wondering if I could do something like "particles_touching[i] |= j;" to set that they are touching and then later "particles_touching[i] & j;" to *check* if they are touching... |
| 15:49.12 | mccarthy | If so, is there a limit on how big j is? and is it OK, if it is simply an int? |
| 15:49.59 | PaxAnima | not sure what thte specs say, but a long is 4bytes, IIRC |
| 15:50.11 | PaxAnima | so if N < 32, yould try that |
| 15:50.36 | mccarthy | ouch... N might be 30,000! |
| 15:51.09 | PaxAnima | oh |
| 15:51.55 | mccarthy | Oh well, I was hoping that I could find an easier way then what I am currently doing.... |
| 15:52.07 | moray | mccarthy: if you don't mind using say (only) 8 times more space you could just have a 2 by 2 matrix with shorts |
| 15:52.40 | moray | mccarthy: you could pack the stuff into bitfields, but unless you use a support library that's likely to cause more trouble than it's worth |
| 15:52.52 | *** join/#gpe pb_ (~pb_@dsl-62-3-66-201.zen.co.uk) |
| 15:53.03 | moray | hi pb_ |
| 15:53.05 | mccarthy | moray: how would your method work? |
| 15:53.09 | mccarthy | hi pb_ |
| 15:53.32 | pb_ | hi moray |
| 15:53.34 | pb_ | hi mccarthy |
| 15:53.44 | moray | mccarthy: well, you could just have particles_touching[i][j] as false/true |
| 15:54.23 | dc_ | pb_: hey |
| 15:54.32 | mccarthy | moray: but even if N = 1000, that means that I have 10^6 particles_touching elements (doesnt it?) |
| 15:54.44 | cmarqu | pb_: Hi |
| 15:54.47 | pb_ | hi cmarqu |
| 15:55.09 | pb_ | dc_: yo |
| 15:55.24 | moray | mccarthy: using bits rather than bytes only saves 8 times the space, which isn't much compared to 1000x1000 |
| 15:55.46 | moray | mccarthy: again, you could half the space by doing something so you only needed to store [i[ |
| 15:56.09 | dc_ | we need gpe badges :D |
| 15:56.09 | moray | [i][j] (effectively) and not also [j][i], but again that doesn't affect that much what will work or not |
| 15:56.38 | pb_ | moray: I have no idea what it is you're discussing, but 10^6 bits is only ~120kB, which isn't an outrageous amount of memory in this day and age. |
| 15:56.47 | moray | pb_: indeed |
| 15:56.52 | pb_ | dc_: indeed, that's a good idea. Someone should look into that. |
| 15:57.03 | moray | [15:47:16] <mccarthy> I need to *remember* from 1 time step to the next which particles are touching each other... |
| 15:57.03 | moray | [15:48:51] <mccarthy> I was wondering if I could do something like "particles_touching[i] |= j;" to set that they are touching and then later "particles_touching[i] & j;" to *check* if they are touching... |
| 15:57.03 | moray | [15:49:10] <mccarthy> If so, is there a limit on how big j is? and is it OK, if it is simply an int? |
| 15:57.06 | moray | (for pb_) |
| 15:57.09 | pb_ | dc_: you made it home to Sussex without incident then? |
| 15:57.18 | dc_ | pb_: yeh, wasn't too bad. |
| 15:57.37 | dc_ | ooooh. |
| 15:57.46 | mccarthy | moray: you got all but the time right ;) |
| 15:57.57 | dc_ | http://www.business-gifts-co.com/your_logo.htm :} |
| 15:58.16 | dc_ | pb_: it would seem we can order as few as 30 from there. |
| 15:58.24 | pb_ | dc_: yes indeed. |
| 15:58.27 | dc_ | pb_: they'll go down well at the next event. |
| 15:58.37 | pb_ | dc_: I have a quote somewhere for some GPE coffee mugs as well. I should dig that out and actually order some. |
| 15:58.49 | dc_ | pb_: oh wow, yes. |
| 15:59.03 | dc_ | pb_: I'd be willing do buy a few :) |
| 15:59.04 | cmarqu | Indeed. And nils said that he might have connections to get T-Shirts done |
| 15:59.14 | dc_ | :D |
| 15:59.16 | pb_ | dc_: cool |
| 15:59.19 | moray | mccarthy: if you really want to use bytes, you want to use something along the lines of (probably not right:) array[a][(int) b/8] |= 1 << (b % 8); |
| 15:59.29 | pb_ | cmarqu: ah great. |
| 15:59.41 | dc_ | well, I'll investigate further into the badges. |
| 15:59.48 | moray | mccarthy: if you really want to use bits, I meant |
| 16:00.02 | mccarthy | my brother-in-law owns a silk screening company (T-shirts) |
| 16:00.27 | mccarthy | moray: I am not wedded to bits, I am simply looking for the easiest/fastest way to do what I explained.... |
| 16:00.39 | dc_ | mccarthy: that sounds good. |
| 16:01.19 | moray | mccarthy: what's fastest depends on the processor etc., but using (say) bytes wastefully may well be as fast / faster than decoding from bitfields |
| 16:01.33 | mccarthy | currently, since I have to actually *remember* stuff about the old contact, I made a struct that was about 24 long (where 24 was my guess as to the largest number of touching particles) |
| 16:01.38 | mccarthy | then I would loop through them |
| 16:02.21 | mccarthy | I was at least a little smart about it in that they are sorted and "end-terminated" so that if I reach a k>j or k=-1, I would break out of the loop |
| 16:02.50 | mccarthy | (where, of course, I am looping for i through k=1-24) |
| 16:03.06 | *** join/#gpe _paulj (~paulj@itg-gw.cr008.cwt.esat.net) |
| 16:03.29 | moray | mccarthy: do you want a list of touching particles, or to be able to look up some pair of particles' status (or both)? |
| 16:04.47 | mccarthy | moray: mostly just a list of all particles (j) that i is touching (so kind of both, but I wont traverse my list randomly) |
| 16:05.37 | pb_ | mccarthy: how many particles is each one likely to be touching? |
| 16:05.40 | moray | mccarthy: well, if there are likely to be few touching particles compared to the total number you might be best to e.g. have a linked list for each particle |
| 16:05.53 | mccarthy | pb_ typicallt < 24 |
| 16:06.25 | mccarthy | moray: would using a g_list be fast? |
| 16:06.36 | mccarthy | of is glib stuff slow? |
| 16:06.48 | mccarthy | s/of/or |
| 16:06.48 | pb_ | mccarthy: g_list is pretty fast for most operations, though g_slist is faster. |
| 16:07.05 | mccarthy | pb_: I think g_slist would be fine for me... |
| 16:07.10 | dc_ | t-shirts with the nice gpe boot splash would be amazing |
| 16:07.29 | pb_ | mccarthy: I think they both have the property that it is much faster to prepend to the beginning than append to the end, because the latter involves chasing pointers all the way through the list. |
| 16:07.36 | moray | mccarthy: certainly if you only need to look at the members of the list in order, you can just iterate over list->next much more quickly than random lookups |
| 16:08.11 | moray | dc_: certainly a good idea for the next event anyone goes to, if not before |
| 16:08.25 | dc_ | quite |
| 16:09.24 | mccarthy | pb_: I *think* prepending would be fine... |
| 16:10.05 | mccarthy | hmmmm, I wonder if cleaning my code will actually make it any faster, or just make it prettier... |
| 16:10.08 | moray | mccarthy: although it's faster to build a sorted list as you add items than to sort the list afterwards |
| 16:11.13 | mccarthy | moray: I think it is naturally "sorted" where by sorted I simply mean that i only remembers about particles with index lower than itself (or vice versa) |
| 16:11.21 | mccarthy | I dont need any further sorting |
| 16:12.30 | mccarthy | is doing "g_boolean touching[N][N]" the same as moray's earlier suggestion? |
| 16:12.53 | moray | mccarthy: yes, I didn't know then you were using glib |
| 16:13.18 | moray | mccarthy: (plain C doesn't provide a boolean type) |
| 16:13.29 | mccarthy | moray: I am not, yet, but I might since the linked list stuff might be easier that way (but would it be faster to write it by hand?) |
| 16:14.09 | mccarthy | this is one of those codes that every single operation needs to be as *fast* as possible (it runs for *months* sometimes!) |
| 16:14.11 | moray | it's not going to be faster to write it by hand - even if you had some wonderfully optimised routine you could just do that bit manually - the e.g. GList structure is hardly complex |
| 16:15.05 | pb_ | mccarthy: if speed is your primary object, the g_boolean touching[N][M] is as good as it is going to get. |
| 16:15.06 | moray | mccarthy: with only 1000 particles you really shouldn't be having memory problems however many structures you make for each combination, within reason |
| 16:15.29 | mccarthy | well, it is more likely 10,000 actually |
| 16:16.13 | pb_ | If you're running on a big computer, a 100MB array is probably not the end of the world. |
| 16:16.50 | moray | indeed, buying an extra 100 MB where you didn't have it ought to be counted as very cheap compared to an N month type delay |
| 16:16.51 | pb_ | or, using bits you could cut that by a factor of eight with probably negligible speed impact. |
| 16:17.26 | moray | (I'm assuming this calculation's being done on a 'normal' platform, not an ipaq or something ;) |
| 16:17.54 | mccarthy | moray: ;) (dualie athlons mostly) |
| 16:19.05 | mccarthy | I wonder if *anyone* that does these types of simulations ever actually got good programmers involved (I have *tried* to get an open-source project started to make this code better, but we have gotten nowhere so far) |
| 16:20.17 | pb_ | mccarthy: I guess that unless you happen to have good programmers within your department already, it's probably going to be cheaper to write bad code and buy a faster computer than to hire someone to write good code. |
| 16:21.12 | mccarthy | pb_: sometimes we get graduate students that are good coders, but everyone else here is worse than me ;-0 |
| 16:22.12 | pb_ | mccarthy: sounds like you should go make friends with people in the computing faculty :-) |
| 16:23.10 | mccarthy | pb_: yes, but they are ina different school. engineering only has computer *engineers* (so I guess I could get someone to make hardware that runs my code ;) |
| 16:23.42 | pb_ | hehe. |
| 16:24.40 | mccarthy | The GPE project has managed to teach me how to put a gtk gui on the front of my code (if I ever get around to it)! |
| 16:24.52 | mccarthy | (thats how I justify it to myself anyway ;) |
| 16:54.32 | *** join/#gpe dc__ (~dc@modem-251-57-60-62.vip.uk.com) |
| 17:17.00 | cmarqu | http://cmarqu.dyndns.org/fosdem2003/ -- I'll protect that with a password in a few days after you looked at it |
| 17:19.09 | mallum | http://cmarqu.dyndns.org/fosdem2003/html/bg_P2080300_img.html the baby mat ! |
| 17:19.45 | cmarqu | :) |
| 17:20.11 | pb_ | cmarqu: wow, you took plenty of pictures then :-) |
| 17:20.54 | cmarqu | pb_: Yup. And I already threw some away. I'll post some to the fosdem guys as well, exluciding the GPe ones if you want. |
| 17:21.04 | cmarqu | excluding even. |
| 17:21.36 | cmarqu | My DSL is getting /.'ed! :) |
| 17:21.40 | pb_ | hehe |
| 17:22.43 | mccarthy | looks like dc was hard at work the whole time :-) |
| 17:23.05 | cmarqu | Yeah :) |
| 17:23.10 | mallum | mccarthy: yeah every moment dc would get his laptop out |
| 17:23.27 | mallum | mccarthy: and then start hassling people for crossover cables etc |
| 17:23.35 | cmarqu | Almost nearly as often as mallum was going out to smoke. |
| 17:25.20 | mallum | http://cmarqu.dyndns.org/fosdem2003/html/cc_P2090324_img.html sums up the wireless networking |
| 17:26.44 | mallum | cmarqu: ;-) |
| 17:36.30 | pb_ | dc__: oh, did your airport arrive while you were away? |
| 18:06.01 | *** join/#gpe flypiper (~netwinder@12-222-141-210.client.insightBB.com) |
| 18:06.24 | pb_ | hi flypiper |
| 18:06.41 | flypiper | pb_: hello :) |
| 18:25.12 | *** join/#gpe walters (walters@verbum.org) |
| 19:04.25 | *** join/#gpe pb_ (~pb@pc2-cmbg4-3-cust239.cmbg.cable.ntl.com) |
| 19:05.41 | *** join/#gpe jg (~jg@dhcp-r17.crl.dec.com) |
| 19:05.52 | pb_ | jg: hi |
| 19:25.22 | jg | pb_: what is the bug number for the release blockers? |
| 19:25.37 | pb_ | jg: 99 |
| 19:25.45 | pb_ | jg: (see topic) |
| 19:29.13 | jg | how do I get the depends on? |
| 19:30.24 | pb_ | jg: click the "dependency tree" link |
| 19:30.38 | pb_ | I think the URL in the topic should take you straight there. |
| 19:33.00 | jg | You're right: thanks. |
| 19:56.28 | *** join/#gpe moray (~moray@moray-ascendit.tsps1.freenet6.net) |
| 20:22.04 | *** part/#gpe mccarthy (~mccarthy@granular.che.pitt.edu) |
| 20:35.58 | *** join/#gpe Bullet-Man-2 (~vadim@bgp01031502bgs.sothfd01.mi.comcast.net) |
| 21:01.00 | *** join/#gpe florian_ (~fuchs@pD955BBDA.dip.t-dialin.net) |
| 21:01.04 | florian_ | hi all! |
| 21:01.07 | pb_ | hi florian_! |
| 21:02.11 | florian_ | pb_: ah, back home again? ;-) |
| 21:02.24 | pb_ | florian_: yep :-) |
| 21:03.57 | pb_ | cmarqu: thanks for the gpg key, but I think you may have sent me the wrong one :-} |
| 21:04.36 | cmarqu | pb_: Unh. |
| 21:04.40 | cmarqu | Let me check. |
| 21:05.10 | pb_ | cmarqu: looks like you sent me key 9AD7E3DB "Philippe Teuwen <philippe.teuwen@ael.be>" . Mine is EF424F7D. |
| 21:05.18 | cmarqu | Argh. |
| 21:07.13 | pb_ | florian_: we got slightly lost on the way out of Brussels (!), but managed to arrive at the tunnel with about three minutes to spare. |
| 21:08.18 | cmarqu | Hi florian_ |
| 21:12.39 | florian_ | pb_: we found our way out using gps and some intuition... |
| 21:12.47 | florian_ | cmarqu: hi! |
| 21:13.03 | pb_ | cmarqu: that worked :-) |
| 21:13.05 | moray | pb_: I think we'd technically left Brussels before we got lost... |
| 21:13.18 | pb_ | moray: I guess that's true. |
| 21:13.54 | florian_ | cmarqu: when did you arrive? |
| 21:14.10 | cmarqu | pb_: Even the encrypted part? |
| 21:14.14 | pb_ | cmarqu: Yes. |
| 21:14.46 | cmarqu | florian_: I slept for almost 4 hours shortly after Aachen, and then it took me till 3 in the morning. |
| 21:16.54 | florian_ | cmarqu: oh cool... |
| 21:18.01 | cmarqu | Yeah, it was indeed cool after a while, not to say cold :) |
| 21:18.22 | cmarqu | Just my nose though, the rest was in my sleeping bag. |
| 21:31.00 | *** join/#gpe TheMasterMind1 (foobar@h-68-164-97-230.MCLNVA23.covad.net) |
| 22:06.57 | *** join/#gpe pierre (~pierre@AClermont-Ferrand-201-1-4-167.abo.wanadoo.fr) |
| 22:07.13 | pierre | Hi all |
| 22:07.39 | pierre | Hope pb_ is not away.. |
| 22:07.47 | pierre | 2 problems.. |
| 22:08.24 | pierre | 1) I need your bugzilla mail to reassign to you the http://europa.armlinux.org/cgi-bin/bugzilla/show_bug.cgi?id=161 |
| 22:08.57 | pierre | ------- Additional Comments From Kevin Lewin 2003-02-10 10:56 ------- |
| 22:08.57 | pierre | pb asked to open this bug. In Bugzilla I can not select "kernel" as a |
| 22:08.57 | pierre | component. Please leave this open for pb to have a look at. |
| 22:09.44 | pierre | 2) I would like to change the ownership of gpe-conf to florian. I dont know how to do this in bugzilla |
| 22:11.26 | cmarqu | pierre: Hi. I could do the 2nd thing if florian_ agrees. |
| 22:11.40 | pierre | I asked him before.. |
| 22:12.21 | cmarqu | Yeahm he also told me, but I'm not sure it was a definite "yes". |
| 22:12.59 | pierre | ah ok. I thoght it was.. |
| 22:13.26 | cmarqu | Could be. |
| 22:13.43 | pierre | It seams that he is more than me the unofficial maintainer.. |
| 22:17.05 | florian_ | cmarqu: yep, is ok - i mailed pierre to do this |
| 22:17.29 | florian_ | ;-) |
| 22:18.00 | pierre | have I missed somthing ? |
| 22:19.08 | cmarqu | Okay, cool. |
| 22:19.42 | florian_ | no, i'm somewhat slow |
| 22:20.51 | pierre | Hi pb_ |
| 22:20.58 | pb_ | hi pierre |
| 22:21.13 | cmarqu | florian_: BTW, did you see the FOSDEM photo link yet? http://cmarqu.dyndns.org/fosdem2003/ |
| 22:21.13 | pierre | do you have the question 5 lines before? |
| 22:21.30 | pb_ | pierre: about bug 161? |
| 22:21.33 | pierre | yes |
| 22:21.40 | pb_ | yeah, I'm looking at it now. |
| 22:21.51 | pierre | I tried to reassign to pb@gnu.org |
| 22:22.05 | pierre | no, to philb@gnu.org |
| 22:22.14 | pb_ | ok, I'll assign it to myself. |
| 22:22.17 | pierre | but its not your bugzilla account |
| 22:22.25 | pierre | nice. |
| 22:22.54 | florian_ | cmarqu: not yet... |
| 22:22.55 | pb_ | yeah, my bugzilla account is pb@nexus.co.uk it seems. |
| 22:22.57 | cmarqu | florian_: What is your mail address for GPE bugzilla? |
| 22:24.39 | florian_ | cmarqu: florian.boor@kernelconcepts.de |
| 22:24.46 | florian_ | i hope ;-) |
| 22:25.07 | cmarqu | Yup, works. |
| 22:25.16 | cmarqu | Congrats, it's yours now :) |
| 22:28.07 | florian_ | cmarqu: may i? wget -m http://cmarqu.dyndns.org/fosdem2003 |
| 22:28.30 | pierre | nice photos.. |
| 22:29.02 | pierre | cool to add some head to the nicks.. |
| 22:29.30 | florian_ | yes, very nice :-) |
| 22:29.44 | cmarqu | florian_: Yup. Or I could make a CD and send it. |
| 22:29.45 | pierre | I'm particulary impressed with dc.. |
| 22:30.04 | pierre | he look so young.. |
| 22:30.10 | pb_ | pierre: he's 15 |
| 22:30.12 | florian_ | cmarqu: how mucg is it? |
| 22:30.16 | cmarqu | florian_: A full mirroring is likely very slow. |
| 22:30.22 | florian_ | pierre: :-) |
| 22:31.21 | florian_ | cmarqu: dsl? |
| 22:31.41 | cmarqu | florian_: 25 MB, but thats not even quite the original size. |
| 22:31.45 | cmarqu | florian_: Yup, DSL. |
| 22:32.23 | cmarqu | 85MB with the web site photos plus the originals. |
| 22:32.41 | *** join/#gpe eah_ (~eah@pro.proact.se) |
| 22:33.31 | pigeon | Gee... |
| 22:33.40 | pigeon | A lot of gpe people. |
| 22:33.44 | pb_ | pigeon: :-) |
| 22:34.01 | florian_ | cmarqu: i'll mirrior the web site - i think here are some more interested in it :-) |
| 22:34.09 | cmarqu | Ok. |
| 22:35.01 | pigeon | pb_: Wish I were there. |
| 22:39.18 | eah_ | Hi, Has anyone wrote some sync-program to some linux calendrar for GPE ? |
| 22:41.04 | florian_ | eah_: not yet :-( But someone will do... soon. |
| 22:41.34 | eah_ | florian_: ok, thanks! |
| 22:41.40 | *** join/#gpe paul ([Snm8WiMCa@hibernia.jakma.org) |
| 22:42.14 | florian_ | eah_: but copying the database files to a pc should work |
| 22:44.54 | eah_ | florian_: yes, but's it's pretty neat to edit the cal on your desktop when you have it around :-) |
| 22:46.03 | eah_ | florian_: Do you know if there's any perl-bindings to micro-sql ? |
| 22:52.03 | florian_ | eah_: you can use sqlite direct on these files - perl bindings for sqlite shoulh exist. Maybe check www.sqlite.org or sqlite wiki |
| 23:07.08 | florian_ | good night! |
| 23:40.07 | *** join/#gpe walters (walters@verbum.org) |
| 23:48.54 | *** join/#gpe mallum (~mallum@pc-80-193-218-21-hw.blueyonder.co.uk) |