--- Log opened Mon Jun 06 00:00:45 2005
00:57 -!- xhrl [~ThomasW@24.80.39.250] has joined #fink
01:37 -!- eno-away [~eno-away@adsl-216-100-134-151.dsl.snfc21.pacbell.net] has quit [Read error: 110 (Connection timed out)]
01:55 -!- eno-away [~eno-away@adsl-216-100-134-134.dsl.snfc21.pacbell.net] has joined #fink
02:00 < cirdan> !trout pogma 
02:00  * Melian takes the trout and shoves it up pogma's right nostril.
02:00 < cirdan> :-)
02:00  * cirdan sleeps
02:23 -!- broeken [~chatzilla@fswfirewall.fss.uu.nl] has joined #fink
02:49 -!- Omnifarious is now known as Omni|Sleep
02:54 -!- broeken [~chatzilla@fswfirewall.fss.uu.nl] has quit ["Chatzilla 0.9.68a [Firefox 1.0.4/20050511]"]
03:12 -!- ringerc [~craig@dsl-202-72-144-62.wa.westnet.com.au] has joined #fink
03:13 -!- bakshi [~bakshi@S0106001217ba40c7.cg.shawcable.net] has quit []
03:19 -!- broeken [~chatzilla@fswfirewall.fss.uu.nl] has joined #fink
04:33 -!- broeken [~chatzilla@fswfirewall.fss.uu.nl] has quit [Read error: 54 (Connection reset by peer)]
04:38 -!- broeken [~chatzilla@fswfirewall.fss.uu.nl] has joined #fink
05:19 -!- vasi [~vasi@modemcable214.145-70-69.mc.videotron.ca] has joined #fink
05:48 -!- Feanor [~astrange@feanor.developer.opendarwin] has quit [Read error: 110 (Connection timed out)]
06:18 -!- Netsplit niven.freenode.net <-> irc.freenode.net quits: htodd, nkuttler, rand-irc, broeken, cirdan, das_, Murr
06:19 -!- _mcp [~hightower@wolk-project.de] has joined #fink
06:21 -!- Netsplit niven.freenode.net <-> irc.freenode.net quits: gregj, pnorman, gzl, gecko2, Clef, Omni|Sleep, dbast, mcp, jack-
06:21 -!- _mcp is now known as mcp
06:23 -!- Netsplit over, joins: pnorman
06:24 -!- Netsplit over, joins: gregj, dbast
06:24 -!- Netsplit over, joins: nkuttler
06:24 -!- broeken [~chatzilla@131.211.90.5] has joined #fink
06:25 -!- Netsplit over, joins: gzl
06:25 -!- das_ [~das@128.104.18.150] has joined #fink
06:26 -!- cirdan [~chris@68.45.134.216] has joined #fink
06:26 -!- htodd [htodd@207.173.201.47] has joined #fink
06:26 -!- rand-irc [~rand@209.89.24.184] has joined #fink
06:26 -!- Netsplit niven.freenode.net <-> irc.freenode.net quits: htodd, nkuttler, rand-irc, cirdan, das_, broeken
06:28 -!- Netsplit over, joins: cirdan, das_, rand-irc, htodd, nkuttler, broeken
06:28 -!- Omni|Sleep [~hopper@Omnifarious.sustaining.supporter.pdpc] has joined #fink
06:28 -!- Clef [~Clef@69-167-25-3.stmnca.adelphia.net] has joined #fink
06:28 -!- jack- [jack@life.hater.be] has joined #fink
06:28 -!- Omni|Sleep [~hopper@Omnifarious.sustaining.supporter.pdpc] has quit [Connection reset by peer]
06:29 -!- jack- [jack@life.hater.be] has quit [Killed by kornbluth.freenode.net (kornbluth.freenode.net (kornbluth.freenode.net ghosted jack-))]
06:29 -!- Clef [~Clef@69-167-25-3.stmnca.adelphia.net] has quit [Killed by kornbluth.freenode.net (kornbluth.freenode.net (kornbluth.freenode.net ghosted Clef))]
06:29 -!- jack- [jack@life.hater.be] has joined #fink
06:29 -!- gecko2 [gecko@gecko2.user] has joined #fink
06:30 -!- Clef [~Clef@69.167.25.3] has joined #fink
06:31 -!- Netsplit niven.freenode.net <-> irc.freenode.net quits: htodd, nkuttler, rand-irc, Clef, broeken, das_, cirdan
06:32 -!- Netsplit over, joins: Clef, cirdan, das_, rand-irc, htodd, nkuttler, broeken
06:32 -!- Omni|Sleep [~hopper@2002:d8fe:1010:1:204:e2ff:feb2:7b2d] has joined #fink
06:36 -!- xerxes1358 [~xerxes__@145.116.2.111] has joined #fink
06:59 -!- Feanor [~astrange@mp1-248-180.dialup.emory.edu] has joined #fink
07:08 -!- xerxes1358 [~xerxes__@145.116.2.111] has left #fink ["Leaving"]
07:30 -!- regeya [~shane@adsl-sp3-cdale176.micgi.com] has quit [Client Quit]
07:32 -!- regeya [~shane@adsl-sp3-cdale176.micgi.com] has joined #fink
07:47 -!- macubergeek [~Uber-l33t@pcp04373021pcs.nrockv01.md.comcast.net] has joined #fink
07:47 < macubergeek> hey is anyone attending the wwdc this morning?
07:49 < macubergeek> anybody awake yet? ;-)
07:50 < vasi> yep
07:50 < vasi> awake, not wwdc
07:50 < macubergeek> well howdy!
07:50 < macubergeek> I'm waiting to hear what Jobs sez
07:50 < macubergeek> I suppose Slashdot will report it
07:51 < vasi> i'm just here cuz i wanted ask someone on #macdev about weak linking
07:51 < vasi> totally forgot it was WWDC :-)
07:51 < macubergeek> I dont' think the rumors of apple switching to intel are true
07:51 < vasi> i doubt they're going to go x86 for Macs...anything else is possible
07:51 < macubergeek> I do suspect they might announce macosx for intel tho
07:52 < macubergeek> they talked to intel to get license to intel optimized compiler I expect
07:52 < vasi> eg: Intel making PPC chips, or Intel making chips for a media center or other non-Mac device
07:52 < macubergeek> yeh I can see that
07:52 < vasi> but why do OS X for x86? every app would need a recompile
07:52 < vasi> and they'd lose many hardware sales
07:52 < macubergeek> yeh I can see that
07:53 < macubergeek> I doubt they'd abandon ppc
07:53 < macubergeek> yeh why buy their stuff when I can run macosx on cheap intel hardware
07:56 < macubergeek> have you had probs with fink on tiger?
07:57 -!- drm [~drm@65.210.36.227] has joined #fink
07:58 < vasi> macubergeek, we've managed the transition pretty well
07:58 < vasi> some problems, but not huge ones
07:58 < vasi> and they're getting fixed fast :-)
07:59 < macubergeek> I was having problems with nessus-ssl but I decided to wait it out a bit
07:59 < vasi> try searching the FAQ and the mailing lists, and if you don't see your problem there, email the maintainer
08:00 -!- drm [~drm@65.210.36.227] has quit [Client Quit]
08:00 < macubergeek> yeh I did that. The maintainer said he didn't have Tiger yet so couldn't finish testing
08:00 < macubergeek> it had a problem with gcc 4
08:00 < vasi> lemme check the build report, one sec
08:01 < macubergeek> k thanks
08:01 < vasi> does the bottom of this page: http://fink.opendarwin.org/build/2005-05-08/logs/libnessus-ssl.log  look like what you saw?
08:03 < macubergeek> yeh something like that plus conflicting versions of libnet
08:03 < macubergeek> with another package
08:03 < vasi> ooh
08:03 < vasi> well i can't really check it out carefully right now, i'm working on my own packages
08:03 < macubergeek> np thanks anyway
08:04 < macubergeek> I'll just wait it out a bit and try it again later this month
08:04 < vasi> 44 down, 7 to go
08:04 < vasi> :-)
08:04 < macubergeek> packages?
08:04 < vasi> you can try emailing fink-users or fink-devel, maybe somebody else has seen your problem and knows a fix
08:04 < macubergeek> k thanks
08:04 < vasi> well not quite packages, those numbers are .info files
08:04 < vasi> which may each represent multiple related packages
08:04 < macubergeek> oh
08:04 < vasi> (eg the -shlibs and -dev bits)
08:05 < macubergeek> k
08:05 < macubergeek> well don't work too hard
08:05 < macubergeek> l8tr
08:05 -!- macubergeek [~Uber-l33t@pcp04373021pcs.nrockv01.md.comcast.net] has quit ["leaving"]
08:10 -!- Fang [~Fang@2002:53cd:a59f:1:0:0:0:1] has joined #fink
08:16 -!- ^baba^ [~baba@YahooBB220041001026.bbtec.net] has joined #fink
08:16 -!- ^baba^ is now known as baba
08:29 < vasi> oh crap, perl 5.8.7 is out
08:30 < vasi> time for more perl variant fun!
08:32 -!- Feanor [~astrange@feanor.developer.opendarwin] has quit [Read error: 110 (Connection timed out)]
08:32 -!- Feanor [~astrange@mp1-248-195.dialup.emory.edu] has joined #fink
08:36 < baba> variants suck
08:39 < pogma> no they don't
08:41 < vasi> bab, why do variants suck?
08:41 < vasi> bab
08:41 < vasi> a
08:41  * vasi threatens his keyboard with a glass of pepsi
08:41 < vasi> baba :-)
08:43 < baba> hehe
08:44 < baba> if u want to make variants like py23 and py24, and python24 is still in unstable, will all the variants go into unstable?
08:44 < baba> i mean if u use Type
08:45 < vasi> well that doesn't mean variants suck, just that we need better ways to specify tree and section
08:46 < vasi> it's not very hard to just take an item out of Type: though
08:46 < baba> yeah
08:46 < vasi> yeah, for stable/unstable i think the way we handle it now is ok
08:46 < vasi> for -ssl it's not, that needs improvement
08:47 < baba> indeed
08:47 < vasi> it's on my list of things to do, but many things ahead of that
08:47 < vasi> probably best would be to have a Tags: or Sections: field...
08:48 < vasi> so that a program could be both 'gnome' AND an 'editor'
08:48 < vasi> if we allow conditionals, then *poof* you can have a program that says it's in crypto if-and-only-if the -ssl variant is being used
08:49 < vasi> would that be non-sucky? :-)
08:49 -!- akh [~akhansen@ldx4.psfc.mit.edu] has joined #fink
08:49 < vasi> (it's a little more complicated than that because of the .deb locations)
08:49 < vasi> hi akh
08:49 < baba> fine with me
08:49 < akh> hi
08:49 < baba> also, i sometimes don't know which variant to pick
08:50 < vasi> baba, well i sometimes don't know which text editor to pick, but i still like that fink has multiple ones :-)
08:50 < baba> is it possible to use conditional in Description?
08:50 < vasi> baba, i'm not sure...you'll have to read the code
08:51 < vasi> i THINK dmacks may have made that allowed
08:51 < baba> if u could use conditional there, it might help the users when choosing a variant
08:51 < vasi> isn't it rare now to be asked which variant you want?
08:52 < vasi> cuz we don't have the virtual -pm depends on -pm581 | -pm584 | ...  anymore
08:52 < baba> maybe
08:56 < vasi> it doesn't look like conditionals are supported in descriptions, but ask dmacks since he's the one who would really know
09:01 -!- clepple [~clepple@dsl092-164-214.wdc2.dsl.speakeasy.net] has joined #fink
09:02 -!- Feanor [~astrange@feanor.developer.opendarwin] has quit [Read error: 110 (Connection timed out)]
09:04 -!- clepple [~clepple@dsl092-164-214.wdc2.dsl.speakeasy.net] has quit [Client Quit]
09:09  * akh needs my sourceforge.net alias change to take effect soon...
09:10 < vasi> how come?
09:10 < akh> I get administrative messages for all of the spam that gets sent to our lists.
09:10 < vasi> oh yah
09:10 < akh> That's clogging my IMAP account.
09:11  * akh is redirecting them to gmail
09:11 < vasi> i redirect all my mail to gmail...but it's beginning to annoy me as gmail search is not so great
09:11 < vasi> spotlight much nicer
09:12 < akh> It'll keep until I get Tiger installed anyway.
09:13 < akh> Judicious filtering should do the trick for me.
09:13 < vasi> no Tiger yet?
09:13 < akh> Nope--decided to wait a bit.
09:16 < vasi> could be smart, i'm seeing much more disk thrashing with Tiger
09:16 < akh> ick
09:23 -!- drm [~drm@user-152-3-159-22.wireless.duke.edu] has joined #fink
09:23 -!- You're now known as RangerRick
09:27 < akh> Nothing like a KDE build to warm up the office on a cool day.
09:27 < RangerRick> :)
09:28 -!- baba [~baba@YahooBB220041001026.bbtec.net] has quit ["Leaving"]
09:39 < cirdan> morning
09:39 < pogma> cirdan: hi
09:39 < vasi> hi cirdan
09:40 < akh> hi
09:41 < cirdan> df -h
09:41 < cirdan> oops
09:41 < akh> heh
09:41 < vasi> rm -rf cirdan
09:41 < vasi> :-)
09:41 < cirdan> !lart vasi 
09:41  * Melian steals vasi's mojo
09:41 < cirdan> :-)
09:42 < vasi> gotta teach the bot to rm people
09:42 < cirdan> she does sometime
09:42 < cirdan> s
09:43 < vasi> oh, guess i didn't see that
09:52 < drm> pogma: i've analyzed the logs from the bindist build, and it looks like there are a bunch of existing GCC fields which don't need to be there
09:52 -!- TheSin [~TheSin@iphost-64-56-130-194.edm.wiband.net] has joined #fink
09:52 < drm> pogma: i'm actually getting mildly optimistic about the project to move to non-transitional 10.4
09:52 < cirdan> heh
09:52 < pogma> drm: Yeah, they were added seemingly randomly
09:52 < cirdan> !wb TheSin 
09:52 < Melian> Welcome back TheSin, o lonely traveller amongst the TCP/IP packets of chaos.
09:52 < TheSin> man finkbot been MIA for a long time
09:52 < TheSin> hehe thanks cirdan, Melian 
09:52 -!- newmanbe [~newmanbe@db53bfde09bf4278.session.tor] has joined #fink
09:53 < drm> pogma: after all, we don't have to fix all 700 pkgs on msachs' 'broken' list before the new tree, only the ones which use g++
09:54 -!- xhrl [~ThomasW@24.80.39.250] has quit ["leaving"]
09:54 < pogma> drm: right
09:54 < drm> one drawback with the current setup is that its hard to test gcc 4
09:54  * drm is building a gcc4-based fink installation for testing purposes, even as we speak
09:55 < vasi> drm, even the c++ ones are ok if they aren't linked to other c++ things that use g++4
09:55 < vasi> ie: if one isolated lib needs g++3.3, we can leave that so long as nothing needs both it AND another g++4 lib
09:55 < drm> vasi: i agree: an isolated c++ lib is OK, and if all the packages which use it use *only* it then it's fine
09:56 < cirdan> vasi: few things need nothing other than libc++
09:57 < drm> cirdan: there are going to be a few examples, where we might keep the old -shlibs splitoff around for backward compatibility even though it requires g++-3.3 to build
09:57 < pogma> RangerRick's checking method, nm -g diff with nm -g | c++filt seems best
10:04 < vasi> how are we going to create the new tree?
10:04 < drm> well, i only have a half-formulated plan at the moment
10:04 < vasi> ie: are we going to start from scratch and pull stuff in as it's verified working (like the 3.3 transition)?
10:04 -!- chris01 [~chris01@212.126.165.246] has joined #fink
10:04 < vasi> (er, i think 3,3 was that way)
10:05 < drm> step 1: get everything ready for the new tree, which means all c++ stuff to gcc4, and non-c++ stuff can be forced to gcc3.3 when necessary
10:05 < vasi> or will we do like 10.4-trans and assume things work, pull the borked stuff as it's discovered to be borked
10:05 < drm> step 2: also, in the "get ready" stage, remove spurious GCC fields from packages that don't need them
10:05 < vasi> er, and add GCC to packages that do need it i guess?
10:06 < drm> yup :)
10:06 < vasi> are we going to make GCC actually *do* something?
10:06 < drm> step 3: at least for stable, create the new tree with all the versioning of the c++ packages done by hand
10:06 < vasi> ouch :-/
10:06 < drm> step 4: now we can open this tree to the public, probably dumping all non-GCC unstable packages in as well
10:07 < drm> vasi: ouch is right... GCC does nothing... this is all a consequence of the compromise which pogma and I (and others) made back in feb or march
10:08 < drm> vasi: but the ouch is not so bad... turns out that -- at least in the stable tree -- there are lots of spurious GCC fields, and not so many we will need to work on
10:08 < vasi> should we just bump the epoch or something instead, maybe?
10:08 < cirdan>  the problem is how to get from transi to non
10:08 < cirdan> w/o wiping all of /sw
10:08 < vasi> rather than "bump the revision by a random amount"
10:08 < cirdan> vasi: no, no need to touch epoch
10:08 < drm> we need to version them anyway, all the c++ dependencies have to be versioned so that things rebuild in the right order
10:09 < drm> epoch is a pain, i'd rather avoid it
10:09 < vasi> we can't write some script to auto-process the version bumping? i could work on that if you want
10:10 < vasi> cirdan, is dist-upgrade going to work for this move?
10:10 < drm> it will be a bit complicated, but if you can get a working one that would be great
10:10 < cirdan> vasi: i need help to get dist-upgrade working for this job
10:11 < vasi> i'll take a look at scripting that, but i have a few things higher on my list (like un-breaking rsync in HEAD)
10:11 < vasi> cirdan, alright i'll look at your patches soonish too
10:11 < drm> vasi: anyway, first we have to get as many of the c++-using packages in shape for gcc 4 as we can
10:11 < drm> which could take a while
10:11 < vasi> yeah
10:12 < drm> not to mention coding for x86 compatibility :/
10:12 < pogma> identification of those packages is done?
10:12 < pogma> hehe
10:12 < vasi> oy
10:12 < drm> pogma: in the stable tree, yes, more or less
10:12 < pogma> cool
10:12 < vasi> yay!
10:13 < drm> although msachs next run may give us further info
10:13 < vasi> er, so we're going to actually set up the tree pretty quickly before release?
10:13 < pogma> I don't see any need to do it that way
10:14 < pogma> slowly adding working packages works for me
10:14 < vasi> me too, i'm just asking
10:14 < cirdan> vamy partches are only the start of what needs to be done
10:14 < vasi> cuz that's the impression i got from the steps
10:14 < drm> vamy partches?
10:14 < cirdan> right now it provides a way to get from 10.3->10.4 transi
10:14 < vasi> 'vamy'?
10:14 < cirdan> vasi: my
10:14 < pogma> lol
10:14 < vasi> ah :-)
10:14 < cirdan> stupid tab key
10:14 < cirdan> :-)
10:15 < drm> pogma, vasi: i'd like to see a transition in which all the stable packages keep working
10:15 < drm> we can add unstable ones more slowly
10:15 < vasi> drm, so would i, but 'keep working' != 'staying in stable even if they don't actually WORK'
10:15 < drm> vasi: right
10:15 < pogma> drm: but users don't see 10.4 until after it is "released", right?
10:16 < cirdan> we need someting like a hash which maps the cueent dist -> the wanted one, then checks out both trees
10:16 < drm> vasi: are there things in 10.4-transitional/stable right now which don't work?
10:16 < cirdan> then each hash would have a list of subs to do
10:16 < vasi> drm, i don't know but it's likely, given the number of packages in unstable which suddenly developed bugginess
10:16 -!- broeken [~chatzilla@131.211.90.5] has quit ["Chatzilla 0.9.68a [Firefox 1.0.4/20050511]"]
10:16 < drm> pogma: there is a delicate balancing act here: if you make the tree too far in advance of release, you have to keep the damn trees in sync
10:16 < vasi> drm, that's my main worry about multiple trees, yeah
10:17 < cirdan> like a big sub for 10.3/10.4-transi -> 10.4 would be to make dummy packages for all installed c++ packages
10:17 < cirdan> which would clesr out all libs
10:17 < drm> vasi: but all the extensive testing i did was in the stable tree... the unstable tree i just dumped to 10.4T
10:17 < vasi> drm, ok maybe stable's good then :-)
10:17 < pogma> drm: yeah, sucks
10:17 < vasi> cirdan, um you'll have to explain that to me better...i'm not sure why that would be necessary?
10:17 < cirdan> then the tree would be swiches to 10.4, and the 10.4 tree would have all versioned build deps
10:18 < drm> vasi: so again, i propose here to be very careful about stable, with everything working, and a bit more flexible about unstable
10:19 < vasi> drm, ok if you're confident that stable is really good...how much actual testing of stable (in terms of checking that things *work*, not just build ok) do you do?
10:19 < drm> i just build them
10:19 < vasi> cirdan, why are the dummy packages needed?
10:19 < drm> but we got lots of reports after release about stuff that didn't work, and we fixed them
10:19 < vasi> 'after release' being the operative words
10:20 < drm> gotta let your users help test, man! :)
10:20 < vasi> <wild-ass-suggestion>maybe we need a public beta period?</was>
10:20 -!- xerxes1358 [~xerxes__@145.116.2.111] has joined #fink
10:20 < drm> well, we've effectively had that, without using the language, for the past six weeks
10:21 < akh> Since Fink < 1.0, isn't the whole project nominally beta ?  ;-)
10:21 < drm> akh: forever and ever and ever
10:21  * RangerRick wonders if we should start versioning fink based on the os release
10:21 < RangerRick> or the fink release
10:21 < RangerRick> so that we don't have to tell people the difference between Fink and fink
10:21 < drm> hmmm?
10:21 < vasi> cirdan, can't we just do 'fink remains in a dist-upgrade state until all packages are using a version >= the version in the current tree'?
10:22 < drm> oh, Fink vs. fink... well, that silliness was started by chrisp
10:22 < RangerRick> yeah
10:22 < RangerRick> we don't have to perpetuate it though  :)
10:22 < pogma> lets rename fink 'port'
10:22 < vasi> no, 'emerge'!
10:22 < drm> pogma: or maybe 'darwinport'
10:23 < TheSin> woody
10:23 < pogma> call it 'drm'
10:23 < TheSin> thaat'd mess ppl right up
10:23 < pogma> users then will have to type 'drm build <pkg>'
10:23 < pogma> and you probably would, eventually, do so :)
10:24 < drm> and drm will shout back at them: "nooooooooooo"
10:24 < vasi> heh!
10:25 < vasi> cirdan, so 'splain to me what you mean re: dummy packages :-)
10:25 < vasi> oooh, there's a Jared widget, shiny!
10:28 < vasi> (sorry, i'm kinda sleep deprived today)
10:28 < cirdan> hmm
10:28 < cirdan> sorry
10:28 < cirdan> connection probs
10:29 < vasi> s'ok
10:29 < cirdan> hmm
10:29 < cirdan> i guess with perfect versioned deps it's not so much of a problem
10:29 < cirdan> vasi: the problem comes in the tempral state
10:30 < vasi> well the idea is that we need versioned deps, right?
10:30 < cirdan> liba=gcc3.3 libb=gcc4
10:31 < cirdan> foo is upgraded and needs libb, zbar is installed and will be upgraded, but needs liba and libb
10:32 < cirdan> zbar will crash until the libsa are all upgraded
10:32 < cirdan> my idea was to make placeholders so zbar doesn't exist until it's actuaclly build during the dist-upgrade
10:33 < cirdan> well, the update-all after the upgrade
10:33 < vasi> well that kinda works
10:33 < cirdan> not uninstall zbar, but make an empty package wioth with a higher rev, but less than the  10.4 tree
10:33 < cirdan> so an upgrade-allwill still work
10:33 < vasi> problem is that things that use zbar will then die, as well
10:34 < cirdan> vasi: they will die anyway
10:34 < vasi> so you need to go all the way up the dep tree
10:34 < cirdan> vasi: for libs, yes, but i wouldn't for binaries
10:34 < vasi> ummm
10:34 < cirdan> there could be exceptions though
10:35 < cirdan> just because an app uses a binary doen't mean it's required for it to function
10:35 < vasi> that's true, but deps are there for a reason
10:35 -!- ^baba^ [~baba@YahooBB220041001026.bbtec.net] has joined #fink
10:35 < cirdan> yes they are, but most ppl dont use suggests and reccomends, either
10:35 < vasi> fink unfortunately only knows about 'X depends on Y', not 'X mostly depends on Y if you're using it for Z, BUT...'
10:36 < cirdan> they just add it to depends
10:36 < cirdan> vasi: you just described an optional feature :-)(
10:36 < cirdan> :-)
10:36 < vasi> variant
10:36 < cirdan> no, optional
10:36 < cirdan> varient means its built different
10:37 < vasi> fine....but anyhow...this whole thing with dummy packages is essentially just a fancy way to keep a list of installed packages that need replacement
10:37 < cirdan> optional means added funcs if x is avail
10:37 < cirdan> vasi: mostly
10:37 < vasi> why don't we just do a recursive remove of the packages instead, and keep a list of them?
10:37 < vasi> and then maybe offer the user a choice of 1) rebuild or 2) go without
10:38 < cirdan> i didnt want to have a secret list hidden from the user, so things like dpkg --get-packages still work
10:38 < vasi> no reason to hide it
10:39 < vasi> but i think that leaving fake packages all over is overkill
10:39 < cirdan> if dpkg doesnt know about it it's hidden :-)
10:39 < cirdan> vasi: it seemed nicer to be able to just keep doing update-alls's until everything was done
10:40 -!- drm [~drm@user-152-3-159-22.wireless.duke.edu] has quit ["Leaving"]
10:40 < cirdan> i didnt wanna have to manage a special seperate list
10:40 < cirdan> because builds fail all the time
10:40 < vasi> er, i mean an on-disk list
10:40 < cirdan> and a user might want to build the most used packages first
10:40 < vasi> not in-memory
10:41 < cirdan> vasi: right, but that means that we wil need to check this on-disk list forever
10:41 < cirdan> every build and install
10:41 < vasi> until we're done with the upgrade, yeah
10:41 < vasi> it's not like we don't already have 300 files that fink reads on startup
10:42 < cirdan> how will fink know when the upgrade's done?
10:42 < vasi> try 'sudo fs_usage | grep fink' someday :-)
10:42 < cirdan> eek
10:42 < cirdan> my eyes!
10:42 < vasi> when everything on the list has been rebuilt or removed
10:42 < cirdan> and if the list gets corrupted?
10:42 < cirdan> gonna have list-old like the dpkg utils do?
10:43 < vasi> if you really want
10:43 < vasi> but 'if something gets corrupted' then fink could do 'rm -rf / w/src/' also
10:44 < vasi> i mean, what if we install fake packages and then the status DB gets corrupted? if stuff gets corrupted, the user is in trouble in any case
10:44 < cirdan> it just seemed just as easy to do an in-memory parse of the package names, desc, and depends, then fake install them w/dpkg
10:44 < vasi> well to be actually protecting the user, we'd have to go over the whole dep tree
10:44 < vasi> which could be somewhat nasty
10:44 < cirdan> since we slurp them in, just add +100 to -rev, and no install/compile scripts
10:45 < vasi> better to append 'fake' to the rev, so it's at least obvious to the user what's going on?
10:45 < vasi> anyway i'll think about it more
10:45 < vasi> what's the plan for updating base, though?
10:46 < cirdan> the temp tree would just have c++ packages, and an update-all of just thouse packages would clear them, then the real 10.4 tree would be set, and update-all would commense
10:46 < vasi> temp tree? wtf?
10:46 < cirdan> iirc the only essential c++ is one ncurses lib
10:46 < vasi> er, apt?
10:46 < cirdan> not essential
10:46 < vasi> dpkg?
10:46 < cirdan> in c?
10:47 < cirdan> dpkg only uses gettext and ncurses
10:47 < vasi> all? i thought it had some c++
10:47 < cirdan> so it'd be safe
10:47 -!- mdmonk [~mr_twitch@64.81.110.110] has joined #fink
10:48 < cirdan> vasi: fink reads current tree, changes info on c++ packages, becomes blind to all but currently modified packages, builds empty debs and installs them, according to what's currently installed,
10:48 < cirdan> when all done, activates 10.4 tree, update-all
10:49 -!- xerxes1358 [~xerxes__@145.116.2.111] has quit ["Leaving"]
10:51 < RangerRick> haha: http://stevenf.com/mt/archives/2005/06/apple_to_switch.php
10:52 < vasi> yeah, saw that
10:53 < akh> yummy!
10:54 < vasi> cirdan, why not just do something amazingly simple: 'fink dist-upgrade' just removes the packages that are not available in the new dist or require an upgrade, put a list on the users desktop....the user can then do 'fink install `cat ~/Desktop/fink.removed`' to get everything back
10:55 < vasi> i don't like message with dpkg and the idea of 'what's installed' behind the scenes
10:57 < vasi> simplicity is a virtue :-)
10:57 < cirdan> i just wanted something more automagic for the user
10:57 < cirdan> and there can be more than one admin
10:57 < cirdan> he muse search to see where /Desktop/fink.removed is? :-)
10:57 -!- newmanbe [~newmanbe@db53bfde09bf4278.session.tor] has quit [Nick collision from services.]
10:57 -!- newmanbe [~newmanbe@f4f383f1ecd68887.session.tor] has joined #fink
10:58 < cirdan> and finding the packages that need to fo into /Desktop/fink.removed is 3/4 of the work to do it all
10:59 < vasi> where we put the file isn't the important part, it can go in / for all i care
10:59 < vasi> finding the packages is super easy
10:59 < vasi> use Fink::Status to get a list of installed packages (3 lines of code?)
10:59 < vasi> find the packages that would need an upgrade, using the code from update-all
11:00 < vasi> do a recursive remove, there's already code for that
11:00 < cirdan> i admit i dont know too much about the packagedb, but if we have %gcc3_pkgs, we can't loop over it and make a dup of the db w/only those packages, clear build/install scripts and add 100 to -rev?
11:00 < vasi> get the list of what's left...the difference between list 1 and list 2 is what's been removed :-)
11:00 < cirdan> seems once we have the g++ packages the rest is easy
11:01 < vasi> no, you have to find everything that depends on them too
11:01 < vasi> it's not super hard, but not easy either
11:01 < cirdan> any non-bin depds will already be tagged with the gcc: flag
11:01 < vasi> but then the stuff about installing dummy packages is essentially leaving the system in an inconsistent state
11:01 < cirdan> if not, it's a package bug
11:03 < chris01> Something seems wrong with our pdb: http://fink.sourceforge.net/pdb/search.php?summary=fink
11:03 < cirdan> oh, and append "temp upgrade empty package" to the descriptioons
11:03 < cirdan> both long and short
11:04 < newmanbe> There is always something wrong with that! :)
11:04 < cirdan> newmanbe: :-p
11:04 < akh> chris01: must be reindexing.
11:04 < cirdan> chris01: the problem is it's on sf :-)
11:05 < newmanbe> "223 packages in 23 sections."
11:05 < newmanbe> The description for the KDE section is wrong: "(version 3.0)"
11:06 < vasi> cirdan, what would REALLY be the easiest thing to do is to 1) find a list of the debs that are ok, copy them somewhere safe  2) tell user to rm -rf /sw and install a new fink  3) fink finds the debs and uses them
11:06 < newmanbe> Yeah! "Found 0 packages without maintainers in the 10.2-gcc3.3 and 10.3 releases:"
11:06 < newmanbe> :)
11:07 < RangerRick> lisppaste: url
11:07 < lisppaste> To use the lisppaste bot, visit http://paste.lisp.org/new/fink and enter your paste.
11:07 < mdmonk> http://dftech.org/.mystuff/fink-xfree-err.txt
11:08 < RangerRick> cool
11:08 < akh> !fc-cache
11:08 < Melian> somebody said fc-cache was http://www.finkproject.org/faq/usage-fink.php#fc-cache
11:08 < RangerRick> hah!
11:08 < RangerRick> no, that's not the fix
11:08 < RangerRick> it's a bug in the X package
11:08 < akh> Ah.
11:08 < RangerRick> it's running fc-cache before fonts.conf is created
11:09 < akh> So at least it's running it.
11:09 < RangerRick> now I can move it to stable!
11:09 < RangerRick> :)
11:09 < cirdan> vasi: except for /sw/var, /sw/etc
11:09 < cirdan> reinstalled from a fresh fink will overwrite any old prefs
11:09 < mdmonk> oh...ok. :) can I modify a build file to switch the order/sequence?
11:09 < RangerRick> the only time fc-cache isn't run now is if peopel install apple's X11
11:09 < vasi> conffiles too
11:09 < RangerRick> mdmonk: might as well wait for me to release the fixed package
11:09 < cirdan> yeah, i meant conffiles when i said /sw/etc
11:09 < mdmonk> haha. ok. :)
11:09 < RangerRick> mdmonk: on tiger, right?
11:09 < mdmonk> si
11:10 < RangerRick> mdmonk: 1. don't bother with xfree86, xorg is better, 2. get the updated info files here: http://ranger.befunk.com/misc/xorg.info and http://ranger.befunk.com/misc/xorg.patch
11:10 < RangerRick> put thos in /sw/fink/dists/local/main/finkinfo/x11-system (you'll probably need to make the directory)
11:10 < RangerRick> and then "fink install xorg"
11:11 < cirdan> or just  /sw/fink/dists/local/main/finkinfo
11:12 < RangerRick> he'd have to make that too, probably
11:12 < RangerRick> figured I might as well have him put it in the right section  :)
11:12 < cirdan> yes, he would
11:12 < cirdan> mkdir -p /sw/fink/dists/local/main/finkinfo/x11-system :-)
11:12 < cirdan> but it's a pain to hevae sections with just a few info files
11:12 < RangerRick> exacitally
11:13 < mdmonk> should I just put the xorg.* files in finkinfo/ or finkinfo/x11-system/ ?
11:13 < RangerRick> either way works, x11-system is more "correct" in theory
11:13 < cirdan> whatever u want
11:14 < cirdan> doesn't matter, its your local tree
11:14 < mdmonk> ahhhh...coolio. 
11:14 < mdmonk> I have to get X built so I can continue working on this article I'm writing on OS X as an attack and pen platform.
11:14 < RageMax> anyone have a working link to the fink .23.10 tarball on the front page?
11:15 < RageMax> it's not listed on sourceforge when you go to it
11:16 -!- vasi [~vasi@modemcable214.145-70-69.mc.videotron.ca] has quit []
11:16 < cirdan> heh
11:18 < mdmonk> the one thing I haven't found for OS X, that would be uber useful, is something like a linux-compatibility layer. e.g. like in freebsd/netbsd/openbsd.
11:18 < cirdan> darwine?
11:18 < cirdan> or just wine
11:18 < mdmonk> isn't that for MS Windows binaries?
11:18 < cirdan> it can be
11:19 < cirdan> you can build stuff for wine, and then anything tht wine was ported to will work
11:19 < RangerRick> he meant linux compat, not windows compat
11:20 < cirdan> it's the same idea
11:20 < cirdan> but u can build stuff for wine
11:21 < RangerRick> mdmonk: if it were to happen, it would be to run linux/ppc binaries, but there's not much point to it
11:21 < RangerRick> otherwise you'd need full-on emulation
11:22 < RangerRick> but it will never happen, not enough reason when most stuff can be ported easily
11:22 < mdmonk> right...
11:23 < mdmonk> but I'm running into troubles porting a few apps. mostly network monitoring apps.
11:24 < RangerRick> probably still easier to port those than write a binary-compat layer  :)
11:25 < mdmonk> yeah. I agree. haha. :)
11:25 < mdmonk> *sigh* you folks are going to make me code again, aren't you. why won't you let me be lazy! hahaha
11:25 < RangerRick> :)
11:25 < RangerRick> I'm not making you do anything, mdmonk.
11:26 < RangerRick> :)
11:26  * RangerRick kicks mdmonk in the nizzuts
11:26 -!- Feanor [~astrange@mp1-248-246.dialup.emory.edu] has joined #fink
11:26 < RangerRick> so I guess I should call matt one of these years and congratulate him  :)
11:26 < mdmonk> I know....but if I want to get these apps working, I need to find a way to compile em on OS X, or find a similar app that is native to OS X.
11:26 < cirdan> RangerRick: matt?
11:27 < RangerRick> cirdan: a friend of mdmonk's and I
11:27 < mdmonk> RangerRick: yeah, I suppose congratulations are in order. or a eulogy. haha
11:27 < cirdan> mdmonk: kernel interaction is different than linux
11:27 < cirdan> ah
11:27 < cirdan> heh
11:27 < mdmonk> cirdan: how so? 
11:27 < mdmonk> cirdan: where could I read up on that?
11:27 < cirdan> kernel interfaces are much differenet
11:27 < cirdan> developer.apple.com :-)
11:27 < cirdan> also the apple mailing lists are good
11:28 < cirdan> use the source, luke!
11:28 < mdmonk> Ok. cool. BRB
11:38 < cirdan> mdthe easiest thing to do is find something that already does what u want ur app to do and use that code to start from
11:39 < mdmonk> yeah, but I also want to understand the underlying code/functionality.
11:39 < cirdan> so look up the func calls :-)
11:39 < mdmonk> I am now. :)
11:40 < mdmonk> I followed your suggestion to go to developer.apple.com
11:40 < cirdan> i mean as u look at existing code
11:40 < cirdan> may help to see the big picture :-)
11:40 < cirdan> mdmonk: did u install the dev docs?
11:40 < cirdan> most if not all the docs from dev.apple.com are in /Developer
11:41 < mdmonk> yes I did. 
11:41 < mdmonk> w00t! so they are local to my system. 
11:41 < cirdan> most
11:41 < cirdan> some docs are newer on the web
11:41 < cirdan> if u have adc u can d/l the newest stuff
11:41 < cirdan> everything
11:41 < mdmonk> i bet the network (code) interfaces are very different.
11:46 -!- mGiff [~mGiff@ottawa-hs-209-217-124-208.d-ip.magma.ca] has joined #fink
11:53 -!- ^baba^ [~baba@YahooBB220041001026.bbtec.net] has quit ["Leaving"]
12:08 -!- bakshi [~bakshi@S0106001217ba40c7.cg.shawcable.net] has joined #fink
12:14 -!- ringerc [~craig@dsl-202-72-144-62.wa.westnet.com.au] has quit ["Whoops, someone let the magic smoke out!"]
12:45 < mdmonk> RangerRick: xorg n xorg-shlibs built and installed. I had to manually go in an rm /usr/X11R6. xorg would not install over, or remove, that directory.
12:45 < mdmonk> retract that.
12:45 < cirdan> mdmonk: correct
12:45 < mdmonk> same error about not finding fonts.conf
12:45 < cirdan> but it should have been removed when u removed xfree/x11
12:46 < mdmonk> right. the xorg or xorg-shlibs pkgs  are still looking for fonts.conf before it's built.
12:47 < RangerRick> mdmonk: what does 'fink list xorg' print?
12:47 < mdmonk> shows xorg-shlibs as installed, and xorg not installed.
12:48 < RangerRick> right, but what version?
12:48 < mdmonk> 6.8.2-35
12:48  * RangerRick wonders if he put the wrong version info file up
12:48 < RangerRick> hrm, no, I think that's right
12:48  * RangerRick checks
12:49 < RangerRick> hrm, 35 is right
12:49 < RangerRick> ah, I see
12:49 < RangerRick> I've got another bug  :)
12:55 < mdmonk> can I run a cmd to create the fonts.conf manually? 
12:56 < cirdan> heh
12:57 < mdmonk> :) debugging is so much fun, huh. haha. RangerRick: is there something I can do to help with the debugging?
12:58 -!- xerxes1358 [~xerxes__@145.116.2.111] has joined #fink
12:59 < cirdan> !trout xerxes1358 
12:59  * Melian takes the trout and shoves it up xerxes1358's left nostril.
12:59 < cirdan> :-D
12:59 < RangerRick> mdmonk: I think I know what's going wrong
12:59 < mdmonk> it's the firewall, right?
13:00 < RangerRick> um, no
13:00 < mdmonk> sorry. :) used to be a firewall admin, and whenever someone would say "I think I know what's wrong", they would say "I think it's the firewall." hahaha
13:00 < RangerRick> ah  :)
13:13 -!- vasi [~vasi@modemcable214.145-70-69.mc.videotron.ca] has joined #fink
13:16 < RangerRick> mdmonk: downlaod again from http://ranger.befunk.com/misc/xorg.info and http://ranger.befunk.com/misc/xorg.patch and put them in your local dir, and do "fink rebuild xorg"
13:16 < mdmonk> haha. ok. :)  
13:16 < RangerRick> (sorry)
13:16 < RangerRick> thanks for being a guinea pig though  ;)
13:17 < RangerRick> hehe, I got a little pissy at a guy who posted to my blog about kde 3.4.1  :)   http://ranger.befunk.com/blog/mtc.cgi?entry_id=598
13:17 < mdmonk> I copied fonts.conf and fonts.dtd from another xorg install (different OS), to /private/etc/fonts/ and the xorg packages installed okie dokie. :)
13:18 < mdmonk> but I'll try the new xorg.info and .patch to see if they work as normally expected.
13:18 < RangerRick> ok
13:18 < RangerRick> well you woudln't know if you've got the files now
13:18 < RangerRick> hehe
13:18 < RangerRick> but oOK
13:18 < mdmonk> I'll remove the files again.
13:19 < vasi> RR, did i tell you about that weirdness i was seeing with xterms disappearing?
13:19 < mdmonk> no problem. the Powerbook understood that it would be used as a guinea pig for software testing. 
13:19 < RangerRick> vasi: erm, no...
13:19 < vasi> well occasionally when i launch xterm, it immediately quits
13:20 < RangerRick> any logs to Console.app?
13:20 < vasi> usually a message about being unable to grab a console
13:20 < RangerRick> odd
13:20 < vasi> cuz of permission problems with /dev/ttyp*
13:20 -!- dreamind [~dreamind@C2107.campino.wh.tu-darmstadt.de] has joined #fink
13:20 < vasi> it's apparently luit that's trying to change the perms
13:20 < vasi> because of my LANG setting
13:21 < vasi> try using a utf8 LANG, see if it happens to you too
13:21 < RangerRick> hrm, ok
13:21 < RangerRick> if you could send me an e-mail, I can pass it on to torrey
13:21 -!- schihei [~schihei@pD902375B.dip.t-dialin.net] has joined #fink
13:21 < RangerRick> this x.org?
13:21 < RangerRick> or xfree86?
13:21 < vasi> yep
13:21 < RangerRick> or both?
13:21 < vasi> xorg
13:21 < RangerRick> ok
13:22 < vasi> with applex11tools, but i don't think that makes a difference
13:22 < RangerRick> I don't really understand any of X's code, I just build it  :)
13:22 < vasi> you wouldn't have any idea how the perms on ttyp's get set in the first place, would you?
13:22 < vasi> i don't understand what's changing them
13:23 < RangerRick> nope
13:23 -!- schihei [~schihei@pD902375B.dip.t-dialin.net] has left #fink ["Leaving"]
13:24 < cirdan> omg, wtf
13:24 < RangerRick> ?
13:24 < cirdan> apple just announced intel chips starting late 2006
13:24 < RangerRick> for PCs?
13:24 < akh> nope
13:25 < akh> Starting with minis
13:25 < slyrus> OS XX?
13:25 < RangerRick> was gonna say
13:25 < RangerRick> mini is a PC  :)
13:25 < cirdan> heh
13:25 < RangerRick> I mean, as opposed to wifi, or some other crap rumors have been
13:25 < RangerRick> actual computers
13:25 < RangerRick> systemss
13:25 < RangerRick> OSX  :)
13:25 < cirdan> thix sux0rs!
13:25 < mdmonk> the macmini is pc-based? no way
13:25 < cirdan> it will be
13:25 < mdmonk> gah!
13:25 < RangerRick> mmm
13:25 < RangerRick> welcome to support hell
13:26 < mdmonk> why??? I'm trying to get MS *out* of my house. :)
13:26 < mdmonk> can't seem to convince the wife that Mac would be better for her to use.
13:26 < vasi> not gonna make development easier!
13:26 < RangerRick> it'll still be a mac, just with an intel chip
13:26 < RangerRick> :)
13:26 < vasi> unless they have some really funky magic
13:27 < cirdan> tim eto prep fink for darwin x86
13:27 < cirdan> :-)
13:27 < mdmonk> RR: rebuilding xorg
13:27 < akh> Probably wouldn't hurt.
13:28 < vasi> once the transition's over, porting will be easier at least
13:28 < cirdan> that's what it'll be
13:28 < cirdan> same AIPs, just low level hw will change
13:28 < vasi> but legacy software? endianness?
13:29 < cirdan> all fux0rd
13:29 < cirdan> :-)
13:30 -!- rjmunro [~chatzilla@gateway.textmatters.com] has joined #fink
13:30  * cirdan holds on to his ppc chip and screams
13:30 < RangerRick> hehe
13:30 < vasi> and who's going to buy a new PPC now until 2007?
13:31 < akh> OTOH, how about OS X (or XI, or whatever) for other x86 boxes?
13:31 < vasi> i really don't see that happening....but somebody will definitely write a VM
13:31 < RangerRick> yeah
13:31 < RangerRick> vmware for osx would be phat
13:31 < RangerRick> pearpc without the ppc emulator
13:31 < vasi> or MOL on a PC
13:32 < vasi> depending how you look at things :-)
13:32 < cirdan> how will apple survive as a hw company?
13:32 < cirdan> i can't figure
13:32 < akh> ipods.
13:33 < cirdan> no
13:33 < vasi> presumably they'll crack down on any business interest that tries a VM
13:33 < cirdan> i love my beautiful stable ibm chipp'd mac
13:33 < cirdan> vasi: opensource :-)
13:33 < cirdan> in india
13:34 < vasi> yeah, but you avg mac user type isn't going to run OS X in a VM that's not super polished
13:34 < vasi> and OSS VM's....just aren't polished
13:35 < vasi> At least Linux on Macs will probably be nicer :-)
13:35 < vasi> why Intel and not AMD, though?
13:35 < cirdan> more marketing $$?
13:35 < vasi> (actually, they probably bid the two of them off against each other :-)
13:36 < rjmunro> How come http://fink.sourceforge.net/pdb/search.php?summary=postgresql gives no results, when I just did "fink install posgresql" successfully?
13:36 < RangerRick> I think the pdb is messed again
13:37 < vasi> but uh....so just before Leopard, we're going to have trees for 10.3, 10.4-transitional, 10.4, 10.4-x86 and 10.5-seed ?
13:37 < vasi> oh, 10.5-seed-x86 too, don't forget that
13:37 < vasi> yay
13:37  * vasi runs in terror
13:37 < RangerRick> hahaha
13:37 < rjmunro> postgresql installed successfully, but I can't see how I'm supposed to get it to start.
13:37 < cirdan> fink info postgresql
13:37 < RangerRick> rjmunro: sudo /sw/bin/pgsql.sh start
13:38 < vasi> what about the codewarrior users? afaik, there's not even a CW PPC anymore
13:39 < RangerRick> heh
13:39 < vasi> oooh they DO have an emulator, *huge sigh of relief*
13:40 < cirdan> vasi: i bet it sucks worse then the 68040 one did
13:40 < vasi> well if it's fast enough for Jobs to demo Excel live, that's not bad
13:40 < vasi> er wait, have they actually said x86 yet?
13:41 < cirdan> yes
13:41 < cirdan> everything they see is on a pent4
13:41 < vasi> ah
13:41 < cirdan> the whole show is running tiger on x86
13:41 -!- rjmunro [~chatzilla@gateway.textmatters.com] has quit [Remote closed the connection]
13:42 < vasi> so then how the heck do they get ok emulation of PPC on x86
13:42 < vasi> ?
13:42 < vasi> that's some crazy shit
13:42 < cirdan> every os x release has been built and runs on x86
13:42 < cirdan> 6
13:42 -!- LawjoskarAway [~larry@cpe-024-168-176-124.nc.res.rr.com] has joined #fink
13:43 < akh> So they're using the Darwin-x86 kernel?
13:43 -!- hennker [flullup@dsl-082-082-237-064.arcor-ip.net] has joined #fink
13:43 < vasi> apparently
13:43 < vasi> i can't imagine they decided to write a new kernel, too ;-)
13:44 < akh> Yeah, that would be excessive.
13:44 < akh> No more Altivec, then?
13:45 -!- lisppaste [~lisppaste@common-lisp.net] has quit [Connection reset by peer]
13:45 < vasi> *shrug* i wouldn't expect it
13:45 < akh> heh--lisppaste must be all broken up about this.
13:45 < RangerRick> maybe they'll add altivec onto intel chips  ;)
13:45 < vasi> but at this point i'm not about to make any confident assertions about what apple would do
13:45 < cirdan> i want my 64bit, bithces
13:46 < vasi> oooh it's gonna have to be a 64bit chip
13:46 < vasi> otherwise no way it can emulate worth crap
13:46 < RangerRick> so they gonna emulate x86 on ppc too?
13:46 < RangerRick> :)
13:47 < cirdan> heh
13:47 < cirdan> gonna have to
13:47 -!- theid [~theid@207.177.103.77] has joined #fink
13:52 < vasi> the DarWINE group is gonna love this
13:52 < theid> no more qemu for them I guess
13:52 -!- lisppaste [~lisppaste@common-lisp.net] has joined #fink
13:53 < vasi> CrossOver and VMWare are probably gonna price VirtualPC out of the market
13:54 < akh> That's a shame. 
13:54 < akh> :-)
13:54 < mdmonk> <sarcasm> oh darn </sarcasm>
13:56 < cirdan> w00t!
14:03 -!- drm [~drm@user-152-3-151-141.wireless.duke.edu] has joined #fink
14:03 -!- xerxes1358 [~xerxes__@145.116.2.111] has quit [Operation timed out]
14:03 < drm> wow
14:04 < RangerRick> yup
14:04 < RangerRick> :)
14:04 < RangerRick> chilly in hell this time of year, I hear
14:04 < drm> hehe
14:04 < drm> you know, it seemed to me that there was no reason to go to wwdc this year
14:05 < drm> doh!
14:05 < akh> yup
14:05 < RangerRick> so what are the chances we can get some of these developer kit loans from apple for free to work on fink?
14:06 < RangerRick> I don't really want to pay $999 for a box I don't get to keep
14:06 < vasi> RR, you have to be a Select/Premier ADC member too....maybe you are already, i dunno
14:07 < RangerRick> I am
14:07 < RangerRick> select's the $500 one right?  :)
14:07 < drm> well, all i can say is, do they want to keep 200,000 fink users happy or not?
14:07 < vasi> :-)
14:07 < RangerRick> considering we're the 20% that doesn't use xcode  :)
14:07 < vasi> eh, we use gcc, that's even better :-)
14:07 < drm> i didn't renew my ADC select
14:07 -!- cianhughes [~cian@cian.ws] has joined #fink
14:08 < RangerRick> vasi: but all our stuff isn't built around all the extra junk to do fat binaries
14:08 < vasi> i'm sure it's part of gcc if xcode has it
14:08 < drm> "universal" sounds so much nicer than "fat"
14:08 < drm> "oh, have you noticed how universal aunt martha is getting lately?"
14:08 < vasi> lol
14:09 < RangerRick> hahaha
14:09 -!- xerxes1358 [~xerxes__@145.116.2.111] has joined #fink
14:09 < theid> sounds like Rosetta isn't a full emulator, working much in the way of WINE
14:09 < drm> that's ok, we don't want to emulate fink stuff anyway
14:10 < vasi> we'll have to see if we want to have fink-universal or separate fink-ppc and fink-x86
14:10 < vasi> testing will become a bitch
14:10 < theid> separate, please
14:10 < drm> i think we'll want both, actually
14:10 < vasi> 3 trees!?
14:10 < drm> three different binary-foo directories
14:11 < vasi> well it means testing 3 binaries...
14:11 -!- Omni|Sleep is now known as Omni|Work
14:11 -!- bakshi [~bakshi@S0106001217ba40c7.cg.shawcable.net] has quit ["<head exploded>"]
14:11 < slyrus> who10.5-transitional-transitional?
14:11 < cirdan> right, but the same patch and .info for all
14:11 < drm> exactly
14:11 < vasi> code that tests for PPC to look for Apple is gonna be annoying
14:12 < drm> make everything so that it *can* build fat
14:12 < cirdan> ok, we need to start asking for donations
14:12 < RangerRick> <ColPanic> your momma's so universal, she contains everything!
14:12 < drm> and then give users the option of non-fat building
14:12 < drm> rangerrick: lol
14:12 < cirdan> vasi: yeah, lots of ifdefs will change
14:12 < cirdan> !trout RangerRick 
14:12  * Melian takes the trout and shoves it up RangerRick's left nostril.
14:12 < cianhughes> anyone here in the hall?
14:12 < cirdan> ?:-)
14:12 < mdmonk> what do you mean by "build fat"
14:13 < drm> mdmonk: "build universal" in steve's terminology
14:13 < drm> cianhughes: wish i was
14:13 < mdmonk> ah. k
14:13 < mdmonk> static builds vs dynamic?
14:13 < drm> before today they were called "fat binaries"
14:13 < RangerRick> mdmonk: NeXT had the ability to put multiple architectures in a single binary
14:13 < mdmonk> static includes vs dynamic I mean
14:13 < cianhughes> i'm too young, spoke to senior director, who told me, come back next year when you're 18...
14:13 < drm> no, you build a single executable which works on both processors
14:13 < mdmonk> oh wow!
14:14 < cirdan> yeah
14:14 < mdmonk> it would be lots of extra code in the bin.
14:14 < cirdan> man ld, i think
14:14 < vasi> oooh, intel compilers are coming
14:14 < mdmonk> wonder what security implications that would bring up.
14:14 < drm> cianhughes: they can't enforce a non-disclosure agreement signed by a minor
14:15 < drm> so they simply don't disclose :/
14:15 < cianhughes> drm: I know, that was the problem
14:15 < cirdan> yeah, snu knows it well
14:15 < cianhughes> though she said she didn't particularly mind me being an ADC member
14:15 < vasi> kinda ironic how XBox is moving to PPC and emulating x86, while Macs are moving to x86 and emulating PPC
14:15 < mdmonk> if the binary was built to include op codes for each architecture, then would that binary be vuln to an attack using a non-native build in that binary?
14:15 < cirdan> vasi: yup
14:16 < vasi> mdmonk, huh?
14:16 < mdmonk> vasi: why is Apple moving to x86? any ideas?
14:16 < vasi> i don't see how that follows
14:16 < vasi> probably for laptop performance
14:16 < cianhughes> So is ANYONE going to buy new macs in the next 18 months?
14:16 < drm> vasi: i think what you wanted to say to mdmonk was "no"
14:16 -!- mwt [~mike@208.42.143.196] has joined #fink
14:16 < vasi> cianhughes, that's what i've been wondering
14:16 < mdmonk> haha. yeah, probably. :)
14:16 < cirdan> haha
14:17 < vasi> i guess they figured with the way the ipod is selling, they can deal if Mac sales slow for a year
14:17 < cirdan> vasi: slow? u mean sto
14:17 < cirdan> unless sales abound
14:17 < mdmonk> since it wouldn't even be able to execute code that isn't built for the architecture it's running on.
14:17 < vasi> and then in early 2006 they'll release a really cheap Mini
14:17 < cirdan> and i bet most things will be fat binaries for a very long time
14:17 < mdmonk> cirdan: I agree. seems like the smart way to go for most vendors. especially how cheap drive space is now days.
14:18 < vasi> well that's ok, i don't mind fat binaries
14:18 < drm> i think a more important question is whether any commercial software house is gonna ship a new release on PPC...we may see the offerings stagnate
14:18 < cirdan> vasi: i can't imaging there is that much a price diff betweek the x86-64 and the g5
14:18 < vasi> cirdan, not because it's actually cheaper to produce, just so they can make up for lost sales
14:18 < cirdan> drm: if they use xcode it shouldn't be a problem
14:18 < cirdan> vasi: no, i mean the chip self
14:19 < cirdan> your not getting a dell when u buy apple
14:19 < vasi> cirdan, yes i agree with you
14:19 < cirdan> no matter the processor
14:19 < cirdan> it'll still be close to the same prioe
14:19 < mdmonk> I never thought that Apple would go to x86 architecture. 
14:20 < cianhughes> i presume we won't be able to run 10.4-intel on Dells out of the box, but i suspect that someone will hack it into working fairly fast.
14:20 < vasi> yeah, probably the PearPC folks
14:20 < cianhughes> thats what I was thinking
14:21 < cirdan> cianhughes: openfirmware will be the #1 showstopper i bet
14:21 < cianhughes> or CherryOS :) *cough*GPL Infringement*cough*
14:21 < cirdan> also all the apple boards will be custom, as always
14:21 < vasi> the possibilities for windows and linux emulation on Mac-x86 will be nice :-)
14:21 < cianhughes> is it going to use openfirmware?
14:21 < mdmonk> see, Apple should have consulted the folks here in #fink before making any rash decisions.  :)
14:21 < cirdan> cianhughes: u really think they will use an award bios POS?
14:21 < cianhughes> it will be a-la-WINE i presume
14:22 < vasi> OF is portable, so i wouldn't be surprised
14:22 < cirdan> cianhughes: suns use it also
14:22 < cirdan> it's an open standard
14:22 < cianhughes> yeah, i suppose so
14:22 < cianhughes> I was thinking they might use something new
14:22 < cianhughes> i wish i had a webcast
14:22 < cirdan> vasi: no, it means that companies wont build for mac, like they dont build for linux
14:22 < cirdan> they'll just rely on wine :-(
14:23 < cianhughes> shudder
14:23 < cirdan> or cedega
14:25 -!- aweil [~aweil@201.255.0.174] has joined #fink
14:25 < mdmonk> RangerRick: the updated-new (from the .info .patch files you provided) xorg n xorg-shlibs built successfully and installed. 
14:25 < mdmonk> it built/installed the fonts.conf and fonts.dtd. they're in /private/etc/fonts/ just like expected. w00t!
14:25 < RangerRick> cool
14:26 -!- dmacks [~dmacks@203-137.dialup.cloud9.net] has joined #fink
14:26 < mdmonk> now apps that require xfree86 or xfree86-shlibs, will they recognize xorg and xorg-shlibs as valid replacement pkgs?
14:26 -!- msachs [~msachs@A17-201-21-9.apple.com] has joined #fink
14:26 < msachs> 'lo
14:26 < cirdan> yoyo
14:26 < dmacks> newmanbe: I fixed PDB descriptions for GNOME/KDE versions.
14:26 < cirdan> mdmonk: yes
14:27 < dmacks> OTOH, the actual population of the database breaks whenever SF feels like it:(
14:27 < mdmonk> thanks! 
14:27 < vasi> do we actually know *why* it breaks?
14:28 < drm> hi msachs
14:28 < cirdan> vasi: because it's sf?
14:28 < dmacks> vasi: I summarized it on -devel in response to Max.
14:28 < msachs> hey drm
14:28 < drm> guess you're not as surpised as we are :/
14:28 < msachs> Guess again :)
14:28 < msachs> Maybe a bit less.
14:28 < dmacks> We *reallY* need to make it atomic and/or move it off SF.
14:28 < msachs> I knew that we had OS X working on it for a long time.
14:28 < drm> yeah, one had heard rumors
14:29 < vasi> thanks dmacks
14:29 < vasi> yeah, transactions would be nice!
14:29 < msachs> I'll be at WWDC this week, I'll probably be able to get my hands on an x86 box to run buildfink on next week.
14:29 < RangerRick> msachs: can you get us boxes?
14:30 < msachs> One thing y'all should know is that only 4.0 will be able to emit x86 code.
14:30 < drm> ah ha
14:30 < cirdan> msachs: we are working on it
14:30 < drm> now the secret 4.0 plot is out
14:30 < RangerRick> msachs: paying $999 to give a machine back means I won't be buying it
14:30 < cirdan> RangerRick: yup
14:30 < dmacks> IMO once an injection is taking >20 minutes, we may as well let it get nuked. As long as we roll back to the last full injection.
14:30 < msachs> Ben: I'll ask around.
14:30 < drm> RangerRick: msachs may not be the right guy to ask... i'm gonna try some other routes too
14:31 < msachs> drm: Probably a good idea.
14:31 < RangerRick> drm: yeah, i figured we need to spam every apple person we know about it   :)
14:31 < cirdan> hehe
14:31 < drm> i figure that the 200,000 downloads of the fink installer on 10.3 should speak volumes
14:31 < drm> even if steve likes to brag about a billion quicktimes or whatever
14:31 < cirdan> time to start collecting donations to fund our new research/porting efforts
14:32 < cirdan> drm: hehehe
14:32 < msachs> I can certainly get a regular x86 buildfink run going, though.
14:32 < drm> that would be cool, msachs
14:32 < drm> much appreciated
14:32 < dmacks> Yeah!
14:32 < msachs> np
14:33 < msachs> There's a "transitioning to 4.0" talk one of my co-workers is giving at WWDC, I'll be going to that, hopefully he'll cover how to deal with C++ :)
14:33 < cirdan> haha
14:33 < cirdan> drm and pogma should be guest speakers there :-)
14:34 < msachs> C++-specific stuff and weird templatey things is an area I don't know too much about.
14:34 < cirdan> for all the work done and ppl brought to apples
14:34 -!- mwt [~mike@208.42.143.196] has quit ["Leaving"]
14:35 < dmacks> msachs: Do you have any information about Radar #4049905? It's a gcc3.3 bug that was closed "fixed in gcc4.0" but no info about the bug itself.
14:36 < msachs> dmacks: "An easy workaround is to use a variable for calculation of the offset expression as shown in the diff"
14:36 < msachs> And we aren't fixing 3.3 bugs which have workarounds at this point.
14:36 < msachs> So the diff
14:36 < msachs> Is instead of doing a->data[4*i]
14:36 < msachs> Do int k = 4*i;
14:36 < msachs> a->data[k]
14:36 < msachs> a->data[k+1]
14:36 < msachs> ...
14:37 < dmacks> Okay...thanks. I don't need it fixed since there are many work-arounds, just wondering what the actual bug was.
14:38 < dmacks> (it behaves differently when under a debugger, or even if I slap printf() in the code)
14:38 < msachs> Doesn't say.
14:38  * dmacks will use that workaround, and get on with my life.
14:38 < msachs> Yep :)
14:40 < dmacks> clef: glib2 and atk1 are now up-to-date in 10.3; need to be tested in 10.4; will get pango1 and hopefully gtk+2 tonite. This was all much more painful than I expected:(
14:40 < drm> msachs: BTW, at the moment, in an effort to jump-start the switch to the non-transitional 10.4 tree, i may be forcing gcc-3.3 on more packages
14:40 < msachs> drm: Okay, as long as my finkfilter_sysgcc can unforce them.
14:41 < drm> msachs: paradoxically, i'm striving for g++-4.0 but am willing to settle for gcc-3.3 when no c++ code is involved
14:41 < drm> once we get the new tree in place, we can go back and address the gcc-3.3 issues
14:41 < msachs> I mean that's one of the points of my builds, to let you folks do things like that and still let us see what needs to be done to get everything over to 4.0.
14:41 < drm> msachs: yeah, will do it the way its been done before, with env variables mostly
14:41 < msachs> drm: Here's a question for you.
14:42 < msachs> I'm going to want to do builds which produce universal binaries now.
14:42 < msachs> Which means modifying CFLAGS and LDFLAGS.
14:42 < msachs> I can patch the info files to add SetCFLAGS and SetLDFLAGS or modify them appropriately.
14:42 < vasi> or you can patch fink
14:42 < msachs> How many packages don't honor those fields?  I know there are some which use nonstandard build systems.
14:42 < drm> yeah, down the road, patching fink is what we want to do
14:43 < vasi> msachs, i can't give you a number...but there are a lot of apps that use weird build systems
14:43 < cirdan> why not branch/patch had?
14:43 < drm> we need to understand how the new system works, make a design decision about when to build universal binaries and how to control that, and proceed
14:43 < msachs> Okay, well, that still runs into the same issue of packages whose build systems don't use that env vars.
14:43 < drm> yes, exactly... those problems will exist in any event
14:43 < vasi> i'd say at least a few hundred?
14:43 < msachs> Is there a Fink policy saying that packages should honor that?
14:43 < drm> vasi: really? yeah, maybe
14:43 < cirdan> msachs: i think seperate debs would be better for the diff arches
14:43 < vasi> msachs, no there isn't
14:43 < drm> msachs: nope, we just port what's out there
14:44 < vasi> we just try to build packages
14:44 < cirdan> vasi: some packages honor certain vars, but not others
14:44 < vasi> yeah exactly
14:44 < msachs> Hm, that'll make life complicated.
14:44 < msachs> I might have to gcccwrap it.
14:44 < cirdan> dpkg and friends use the arch in the path to the deb
14:44 < msachs> *gccwrap
14:44 < cirdan> we could setup symlinks, but...
14:44 < vasi> msachs, that's one way to do things
14:44 < drm> msachs: also, we need to get pogma involved in this discussion quite soon, since he will undoubtedly be the point guy for getting gnu libtool to play fair with universsal binaries
14:45 < msachs> cirdan: If you want to do that, there is an undocumented and unsupported option to gcc_select, --wrap, which will set that up for you.  I wrote that code last summer :)
14:45 < cirdan> yeah
14:45 -!- finkooo [~blankboy@106.Red-80-39-87.pooles.rima-tde.net] has joined #fink
14:45 < finkooo> hi!
14:45 < drm> msachs: we don't like to use gcc_select in fink, largely because folks might be doing multiple builds
14:45 < finkooo> somebody know why fails fink when i try to apt-get update?
14:45 < finkooo> 403 Forbiden
14:45 < cirdan> msachs: no, i mean dpkg looks at dists/unstable/main/binary-darwin-powerpc
14:45 < RangerRick> !403
14:46 < Melian> rumour has it, 403 is http://fink.sf.net/faq/usage-fink.php#four-oh-three
14:46 < vasi> msachs, we already have the path-prefix mechanism which we've been using to force gcc
14:46 < Omni|Work> Reading the gcc.gnu.org stuff, it seems like gcc-4.0 isn't really ready for prime-time yet.
14:46 < vasi> i would suggest just setting up a different path-prefix
14:46 < Omni|Work> But, I don't know about the Apple fork.
14:46 < msachs> Oh, also, you guys should know, XCode 2.1 fixes a lot of 4.0 ICEs which I've seen in my builds.
14:46 < vasi> 4.0 is fine :-)
14:46 < msachs> We've synced with FSF 4.0 final.
14:46 < msachs> It fixes the netpbm ICE, for one.
14:46 < cirdan> cool.
14:46 < drm> msachs: cool... do you know the build number by chance?
14:46 < Omni|Work> Oooh...
14:46 < drm> the gcc build number, i mean
14:47 < msachs> Apple's XCode 2.1 one?  Uh, 5xxx.
14:47 < msachs> I don't think it's in FSF CVS, we switched to our own private repo to do some x86 stuff.
14:47 < Omni|Work> Why doesn't Software Update update XCode stuff?
14:47 < drm> yeah, well, i need to test for it in fink to find out which compiler is being used, msachs :)
14:47 < vasi> gcc -v would be nice :-)
14:48 < drm> gcc -v will tell us
14:48 < msachs> drm: I /think/ it's 5026
14:48 < cirdan> anyone wanna mirror xcode 2.1 for the main fink devs? :-)
14:48 < msachs> Omni: I don't know.
14:48 < cirdan> cause apple will be slamme dfor a long time
14:48 < finkooo> RangerRick: 
14:49 < finkooo> sorry , what version i need to use with tiger
14:49 < finkooo> ?
14:50 < vasi> we're probably going to have to set up a better system for testing packages...cuz it's already tough enough to test on two versions of OS X, with two arches as well that's gonna be a non-starter
14:50 < vasi> maybe recruit testers, or set up a list of packages that need testing per arch/OS so that devs can test for other devs
14:50 < RangerRick> finkooo: 10.4-transitional
14:51 < msachs> drm: My boss'll be swinging by in a minute to take me to Mocsone, I'll ask him for a definitive answer.
14:51 < vasi> er, 'testing' != 'building', i mean testing that things really do work
14:51 < theid> I'd go for the package testing list
14:51 < finkooo> RangerRick: i download the install binary from the web-fink
14:51 < msachs> vasi: Let me know if there's anything else I can do to help with that.
14:51 < theid> could set up a website really easily to do that, too
14:51 < dmacks> compare.php was an attempt in that direction.
14:51 < drm> msachs, vasi: we need to give the process some thought
14:51 < finkooo> 0.7.1-Installer
14:51 < msachs> drm and I talked about having my automated builder/checker wired up to fink-commits.
14:51 < finkooo> there is?
14:52 < msachs> getting buildfink to a state where it's getting the best data it can and then setting it up to just keep looping and automatically sending out reports is another thing I'd like to get done.
14:52 < vasi> maybe move to a debian-like system, where auto-builds go into unstable....at which point testers are notified, and after testing it moves to an intermediate dist automatically
14:53 < regeya> I'
14:53 < vasi> but that's just another random thought, i don't know what's right to do yet
14:53 < regeya> INTEL???!?
14:53 < vasi> yeehaw
14:53 < theid> well, Debian-like could mean having a "testing" branch that goes into "unstable" after somebody checks it out
14:53 < cirdan> vasi: exactly
14:53 < dmacks> Yeah...would be good for the hypermultivariant packages that nobody tests all of for each update.
14:53 < finkooo> somebody uses fink with tiger????
14:54 < cirdan> finkooo: we all do
14:54 < dmacks> finkooo: Many do, actually.
14:54 < regeya> heh...someone on the gentoo forums has already posted an 'rip apple 06/06/05', erm, post
14:54 < finkooo> what package version uses?
14:54 < vasi> theid, except that would be confusing because with Debian 'unstable' comes first, then ppl test it and then it goes into 'testing'
14:54 < drm> finkooo: go to sf.net/project/fink, select "files" and then find Fink-0.8.0.Installer.dmg
14:54 < finkooo> i try with Fink-0.7.1-Installer
14:54 < finkooo> ok
14:54 < drm> sorry, finkooo, sf.net/projects/fink
14:55 < vasi> so using the same names 'testing' and 'unstable', but with the opposite meanings, would be kinda funny but very mixed up
14:55 < dmacks> "fink.sf.net" too
14:55 < drm> dmacks: no
14:55 < drm> its not linked from there, because our "PR" guy hasn't put an announcement out yet
14:55 < finkooo> how i remove the last version of fink, for i try to install this drm ?
14:55 < dmacks> Ah right.
14:56 < drm> finkooo: sudo rm -R /sw
14:56 < theid> vasi: right, by bad
14:56 < dmacks> vasi: So wait, fink *isn't* Debian?
14:56 < theid> *my bad
14:56 < vasi> dmacks shhh, you'll confuse the newbies
14:56  * dmacks shuts up.
14:56 < cirdan> hehe
14:56 < msachs> Will any Fink people be at WWDC this year?
14:57  * Omni|Work hunts down XCode 2.1
14:57 < dmacks> Not I
14:57  * cirdan wishes he had the plane fare/ticket $$
14:57 < finkooo> drm: thanks
14:57 < Omni|Work> I was kinda sad to learn about the Intel switch.
14:57  * Omni|Work :-(
14:57 < theid> from an IM: "Apple will have to use the 'think similar' marketing campaign now"
14:57 < finkooo> if the new ibooks uses Intel the OSX will be more slow?
14:57 < Omni|Work> Partly because it's pretty clear that if you care about performance/watt, AMD is the way to go.  And partly because I like the PPC architecture a lot better.
14:57 < dmacks> theid: ha!
14:57 < drm> msachs: no
14:58 < cirdan> heh
14:58 < cirdan> Omni|Work: yeah, i agree on both counts
14:58 < cirdan> bbl
14:58 < msachs> Ah, pity.
14:58 < drm> indeed
14:58 < finkooo> drm: 
14:58 < drm> i really didn't think there was going to be much of interest for fink this year
14:58 < drm> doh!
14:58 < msachs> :)
14:59 < finkooo> if the new apple machines, runs on x86 will be more slow
14:59 < finkooo> ?
14:59 < vasi> Adobe gets told, MS too, even Wolfram, what are we chopped liver? ;-)
14:59 < drm> vasi: apparently :)
14:59 < vasi> finkooo, i suspect they will be faster
14:59 < drm> but we've known that for a long time
14:59 < msachs> Well, I'll be there.
14:59 < vasi> mmmm, chopped liver is yummy
14:59 < finkooo> vasi: but, PPC is a lot beeter 
14:59 < msachs> Don't feel bad, vasi.  I didn't know either :)
14:59 < dmacks> Now it turns out I'll probably be in CA next weekend. If I was richer than God I could arrange new flights and WWDC tix now.
14:59 < drm> msachs: ah, sorry, i misspoke... should've said that you'll be the only fink person at wwdc, Matt
14:59 < finkooo> and my battery runs 5 hours,
15:00 < msachs> drm: Yeah, what am I, chopped liver? ;)
15:00 < drm> msachs: and while i'm at it, i mustn't forget Murr, our other in-house guy
15:00 < RangerRick> drm: don't insult the guy who could be getting us x86 test hardware!
15:00  * RangerRick grins
15:00 < msachs> dmacks, drm: Never underestimate SJ's ability to drop bombshells at WWDC :)
15:00 < vasi> where's the perl wizard when i need him?
15:00 < msachs> vasi: I might be able to help.
15:01 < dmacks> Rigjh:)
15:01 < dmacks> *right
15:01 < msachs> Ben: What would you want in terms of x86 hardware other than a box churning away with buildfink and occasional one-off builds for testing random things?
15:01 < drm> msachs: i think vasi was referring to Murr... are you acquainted with him? you aware of his background?
15:01 < vasi> not Apple related, just still unsure what to do with the perllocal.pod
15:01 < msachs> drm: Nope.
15:01 < drm> msachs: Murr aka Matthias Neeracher was Mr. MacPerl prior to OS X
15:02 < msachs> drm: Ah, okay, yes.
15:02 < drm> vasi: why don't we have fink itself install an empty one?
15:02 < cirdan> heh
15:02 < vasi> i thought pudge took over before OS X?
15:03 < drm> vasi: maybe
15:03 < dmacks> vasi: doesn't work if /user/bin/perl changes versions.
15:03 < drm> maybe i mean prior to OS 9
15:03 < msachs> vasi: Haven't really been following that thread too closely, what's going on with it?
15:03 -!- finkooo [~blankboy@106.Red-80-39-87.pooles.rima-tde.net] has quit ["leaving"]
15:04 < vasi> ok, basically it's the versioned vs. non-versioned thing that leaves me annoyed
15:04 < dmacks> Simplest (from a fink POV) is to nuke the file instead of leaving empty. But I agree...need Randal or *someone* to tell us what's what.
15:04 < drm> vasi: ah yes, we need to fix that
15:04 < vasi> currently i have perllocal.pod in /sw/lib/perl5/darwin/ (non-versioned) AND /sw/lib/perl5/5.8.6/darwin-thread-multi-2level (versioned)
15:04 < msachs> Okay, off to Mocsone.
15:04 < msachs> drm: gcc-5026 shipped with XCode 2.1
15:04 < vasi> so 'perldoc perllocal' only sees the versioned one, which sucks
15:04 < msachs> See y'all!
15:04 -!- msachs [~msachs@A17-201-21-9.apple.com] has quit ["wwdc"]
15:04 < drm> thanks
15:04 < vasi> msachs, have fun :-)
15:05 < vasi> pudge is already way too busy afaik....Randal only shows up when things break
15:05 < vasi> who else would know?
15:05 < dmacks> Since we requilb the whole perllocal.pod on-the-fly at every -pm* install/removal, maybe we should switch to a single perllocal.pod for all (nonversioned and all versions of versioned)
15:05 < drm> vasi: it gets rewritten every time, which is gonna make it hard to change
15:05 < vasi> drm, it gets rewritten under fink command
15:06 < vasi> so we can change that easily
15:06 < drm> vasi: easy to imagine fink doing something different; harder to imagine how to do the transition
15:06 < vasi> dmacks, then what do we do about different versions of perl installed?
15:06 < dmacks> ?
15:06 < drm> dmacks: no, a better way would be to have the versioned perllocal.pod include all the non-versioned entries when it is written
15:07 < vasi> yeah, drm that makes most sense
15:07 < drm> but this has the problem that if you installed a non-versioned pm, the perllocal.pod for the versioned ones won't get updated... hmmm
15:07 < vasi> another issue is with apple's perl
15:07 < drm> dmacks, vasi: on second thought, the best thing would be to convince perl to consult both of them
15:07 < dmacks> drm, vasi: yeah.
15:08 < vasi> time to hack on perldoc i guess, yay
15:08 < vasi> that's some really old perl code
15:08 < dmacks> drm: What's wrong with a single consolidated file?
15:09 < vasi> dmacks, we don't want users using 586 to see 581 modules
15:09 < vasi> but 'which perl is current' can change during runtime
15:09 < dmacks> vasi: the entries are already tagged with where they are installed. And since perl5.8.6 see perl5.8.1 modules anyway....
15:10 < vasi> dmacks, are you sure 586 already sees 581?
15:10 < vasi> i didn't see that behavior
15:11 < drm> dmacks: i'm not sure that apple's 586 sees 581, unfortunately
15:11 < drm> in fact i rather doubt it... it requires a compile-time flag
15:11 < dmacks> Hm, maybe not that combination, but perl goes out of its way to pick up lower, bincompat versions of its libs. I assume we follow that behavior?
15:11 < drm> our 586 in 10.3 did this however
15:12 < vasi> oy vey
15:12 < dmacks> ...and I assume our 5.8.7 will pick -pm586 and -pm584
15:12 < Clef> anyone getting the intel dev kit?
15:12 < drm> yeah, i tried to convince the apple guys to do it our way but obviously didn't succeed
15:12 < drm> hi Clef
15:12 < Clef> hi
15:12 < drm> we're trying to beg some free ones from apple
15:13 < dmacks> (OTOH, 5.9 has broken bincompat so maybe this will become less an issue)
15:13 < TheSin> Clef, I am
15:13 < TheSin> Just doing the paper work for it now
15:13 < TheSin> stupid apple
15:13 < dmacks> But my point remains: perllocal.pod entries are already tagged with which perlXXX (or none) they are.
15:13 < vasi> hi Clef...do you know when you'll have a new Mac::Carbon ready? i'm kinda waiting on it so i can update some packages
15:14 < TheSin> Oops didnt' mean to type that
15:14 < drm> Clef: on another topic, your sagasu package is one of the few things left that uses gnome-1, and there is a gnome-2 version available... do you know anything about who uses this? do we need to keep the gnome-1 version around?
15:14 < Clef> you can update it
15:14 < vasi> dmacks, so we should convince perldoc to check that?
15:14 < drm> Clef: it comes up now because it won't build with gcc 4.0
15:14 < vasi> clef, ok sure :-)
15:14 < Clef> i doubt many folks use sagasu.
15:14 < drm> me too
15:14 < Clef> no need to keep it
15:14 < drm> i plan to ask some of the japanese guys though
15:14 < dmacks> vasi: s/perldoc/user/
15:15 < drm> i might make a sagasu2 for gnome2
15:15 < vasi> dmacks, does anybody actually user 'perldoc perllocal'?
15:15 < vasi> cuz there's really nothing else perllocal.pod is good for, right?
15:15 < dmacks> Correct.
15:15 < drm> or just replace it
15:16 < dmacks> cpan bundle *may* use it, but that doesn't work so well with fink-installed pkgs anyway.
15:16 < vasi> so this isn't really worth a lot of work, whatever we do
15:16 < dmacks> Right,
15:16 < Clef> vasi: you can take over mac-carbon if you want also :)
15:17 < vasi> clef, ok
15:17 < Clef> mostly i just pkged it cause noone had yet
15:17 < vasi> i just like my Mac::Glue :-)
15:17  * akh will write a FAQ item about the PDB update issue so that we can just point people there.
15:17 < vasi> i've got some command line scripts to activate exposé that i built with it, they're fun to run remotely while someone else is sitting at my Mac
15:18 < dmacks> heh
15:18 < vasi> (in addition to more useful things)
15:18 < akh> Sounds fun.
15:18 < dmacks> Sort of the Quartz version of xflip/meltdown/xroach :)
15:18 < vasi> used to use them so i could make Quicksilver do Expose
15:18 < vasi> but somebody finally wrote a QS plugin, yay
15:19 < dmacks> bbl
15:19 < vasi> so yeah, if we want a perllocal solution, and it's not worth a lot of work, let's just dump everything into the non-versioned perllocal
15:19 -!- dmacks [~dmacks@dmacks.student.supporter.pdpc] has quit ["leaving"]
15:19 -!- chris01 [~chris01@212.126.165.246] has left #fink []
15:19 -!- ismith [ismith@c-24-14-254-14.hsd1.il.comcast.net] has joined #fink
15:19 < vasi> bye dmacks
15:21 < drm> vasi: the problem with that one is that you'll have entries from the wrong version too
15:21 < vasi> yeah, but as dmacks pointed out, 'perldoc perllocal' also shows the version of perl for each module
15:21 < Clef> tried xcode 2.1 yet?
15:22 < drm> ah, sorry, missed that
15:22 < drm> not yet Clef
15:22 < ismith> is it possible to install gnome on tiger yet?
15:22 < Clef> yes
15:23 -!- vasi is now known as vasiAway
15:24 < ismith> i keep getting a "can't set build lock" error on gnome-terminal when i try to install either gnome-terminal alone, or as part of bundle-gnome.  any ideas?
15:25 < akh> We need more information.
15:25 < akh> use lisppaste
15:25 < akh> lisppaste:  URL
15:25 < lisppaste> To use the lisppaste bot, visit http://paste.lisp.org/new/fink and enter your paste.
15:25  * RangerRick is not looking forward to the millions of messages about the x86 switch on the fink list
15:25 < RangerRick> gonna have to ignore it again for a while  ;)
15:26 < lisppaste> ismith pasted "gnome-terminal build lock" at http://paste.lisp.org/display/8852
15:26 < akh> It will only affect new boxen, anyway.
15:26 < ismith> hold on
15:26 < ismith> it didn't have that "already installed thing" a minute ago
15:26 < ismith> i'll be back once i've got that sorted
15:27 < lisppaste> akh annotated #8852 with "It tells you what to do:" at http://paste.lisp.org/display/8852#1
15:27 < ismith> i re-ran the command
15:27 < ismith> hence the new error
15:27 < Clef> fink wont werk on my x86zorz i ar erunning windoes95
15:27 < akh> No, no.  What I just pasted.
15:27 < akh> Clef:  haha
15:27 < RangerRick> woot, I just reproduced the --accepting-rule bug
15:28 < TheSin> so I guess are gonna have to add arch trees now too huh
15:29 -!- regeya [~shane@adsl-sp3-cdale176.micgi.com] has left #fink ["Leaving"]
15:29 -!- Sortova [~tarus@66.45.100.223] has joined #fink
15:30 < RangerRick> woot, 100k/s downloading xcode 2.1  :)
15:30 < TheSin> ooo 2.1
15:31 < Clef> 251KB/sec
15:31 < Clef> sorry up to 279
15:32 < Clef> cable is nice
15:32 -!- Feanor [~astrange@feanor.developer.opendarwin] has quit []
15:32 < Clef> its not the 4 MB/sec i get at work tho..
15:32 < drm> man, i can't tell ya how much trouble i had grabbing the final tiger seed when i was in china
15:34 < theid> oh man, 751 MB... that's going to take forever and a day
15:34 -!- Fang [~Fang@2002:53cd:a59f:1:0:0:0:1] has quit [Remote closed the connection]
15:34  * akh should install Tiger one of these days.
15:34 < TheSin> where do I go to buy this new dev platform
15:34 < TheSin> I can't find it
15:34 < TheSin> theid, hope your not on dialup
15:35 < theid> close to -- 128 kbit
15:35 < drm> thesin: presumably from the adc site
15:35 < TheSin> drm, that is where I am
15:35 < TheSin> and yes I have a seed :P
15:35 < TheSin> just can't find it in the store there
15:36 < TheSin> maybe it's not posted yet?
15:36 < RangerRick> TheSin: I don't think you can do it if you just have a seed, you have to have a select or premier membership
15:36 < RangerRick> seed is just an asset that gives you prerelease software
15:36 < TheSin> RangerRick, I have one
15:36 < theid> I think I'll download it to a computer with a faster connection first, then do sftp overnight
15:36 < TheSin> I have a premier I think or select
15:36 < TheSin> the one with 5 seeds
15:36 < RangerRick> ok
15:36 < RangerRick> then you're fine
15:36 < RangerRick> they may just not have posted it yet
15:36 -!- atoussaint [~atoussain@199.125.244.126] has joined #fink
15:37 < RangerRick> ah
15:37 < RangerRick> go to developer.apple.com
15:37 < RangerRick> it's linked from there
15:37 < TheSin> ahh okay thanks RR
15:37 < atoussaint> Hi there. How do I start Apache2 at startup. Fink distribution?
15:38 < TheSin> sudo daemonic enable apache2
15:38 < TheSin> then restart
15:39 < atoussaint> the same goes for mySQL?
15:39 < vasiAway> no, then do '/Library/StartupItems/daemonic-<something>/daemonic-<something> start'
15:39 < mdmonk> TheSin: if you restart launcher, will that handle the starting up of apache2 when you enable it that way (daemonic ....)
15:39 < mdmonk> I mean Launcher.
15:40 < TheSin> no
15:40 < mdmonk> ok.
15:40 < akh> Just saw a new article on the Mac/X86 thing--part of the motivation was faster chips for notebooks.
15:40 < akh> Makes sense.
15:41 < RangerRick> yeah, not surprised
15:41 < vasiAway> yup
15:41 -!- vasiAway is now known as vasi
15:42  * TheSin goes and cries
15:42 < TheSin> :D
15:42 < akh> There, there.  A 3 GHz powerbook will dry those tears.  ;-)
15:43 < TheSin> not if it's x86
15:43 < TheSin> I'd rather a 2Ghz ppc
15:43 < theid> I always thought the PPC chips ran a bit cooler for the same Gflops
15:44 < theid> but not according to SJ
15:44 < TheSin> anyhow where do I get teh info on the Developer Transition Kit
15:44 < TheSin> I can't see what I'm getting for 999
15:44 -!- drm [~drm@user-152-3-151-141.wireless.duke.edu] has quit ["Leaving"]
15:44 -!- atoussaint [~atoussain@199.125.244.126] has quit []
15:45 < RangerRick> you're getting a lease of a machine  :P
15:46 -!- mds-work [~mds-work@destiny.EECS.tufts.edu] has joined #fink
15:46 < vasi> theid, they used to
15:46 -!- ringerc [~craig@dsl-202-72-144-62.wa.westnet.com.au] has joined #fink
15:47 < theid> still, the G5 was way cheaper than Opteron
15:47 < theid> I think the biggest thing may be Windows programs (games) if that's a possibility
15:53 < ismith> does fink compile the source in /sw/src?  i.e., if i want to modify the source, can i unzip /sw/src/foo.tar.gz, make my change, re-tar it and fink install it?
15:53 < akh> Nope--that'll cause an MD5 error.
15:54 < ismith> oh.  is there a way to do that (let fink do the compiling for me), or will i just have to do it by hand?
15:54 < akh> You can edit (or create if it doesn't exist) a patch script.
15:54 < akh> (.patch file)
15:55 < ismith> ok
15:55 < akh> Which more or less involves building (not necessarily installing) by hand
15:55 -!- atoussaint [~atoussain@199.125.244.126] has joined #fink
15:56 < vasi> ismith, if you just want to make your own packages you can write your own .info/.patch files, it's not too hard
15:56 < vasi> (see the tutorial in the doc section of our website)
15:56 < ismith> building i don't mind
15:56 < akh> Check the .patch files on your system for examples.
15:56 -!- vasi [~vasi@modemcable214.145-70-69.mc.videotron.ca] has quit []
15:57 < ismith> thanks
15:57 < atoussaint> tks theSin, it worked.
16:00 -!- akh [~akhansen@ldx4.psfc.mit.edu] has quit []
16:01 < atoussaint> How do i install php in Apache2, fink distro?
16:07 -!- You're now known as RangerAway
16:07 < cirdan> back
16:09 < cirdan> so...
16:09 < cirdan> anyone finish the d/l yet?
16:14 < cirdan> haha
16:14 < cirdan> http://uploads.offtopic.com/files/empjobs2.JPG
16:14 < dreamind> hi folks
16:15 < dreamind> hm, andbody here who was _not_ sad about the switch to x86 cpus? - I'm kinda pissed.
16:15  * cirdan is sad
16:18 < mdmonk> highly irritated about the CPU bit.
16:18 < Omni|Work> I'm also sad.
16:19 < dreamind> hm, I think this sadly shows just one point - in opposite to software for linux and other free operating systems, open source software developed specifically for mac os x is not at any point free. its tied to what apple decides to do.
16:19 < TheSin> atoussaint, php4 or php5
16:19 < cirdan> hey TheSin 
16:19 < atoussaint> php5
16:19 < cirdan> dreamind: just like os developement for ibm mainframes, and sun boxes?
16:20 < Omni|Work> dreamind: Well, that's true, but I don't the CPU debacle demonstrates that point.  :-)
16:20 < dreamind> yup.
16:20 < Omni|Work> If, OTOH, Apple were to drop Darwin and go with some random weird proprietary OS core, it definitely would.
16:20 < dreamind> well anyhow - at least today I think I won't buy any machine with a x86 cpu from apple.
16:21  * Omni|Work nods.
16:21 < Omni|Work> I might, regardless.
16:21 < TheSin> hey cirdan 
16:21 < TheSin> atoussaint, fink install libapache2-ssl-mod-php5
16:22 < Omni|Work> I think their decision is a sad one, but I don't think it's exactly a bad one.
16:22 < TheSin> I think it's Sad and Bad
16:22 < Omni|Work> It sounds like the PowerPC architecture was just not keeping up.  :-(
16:22 -!- mGiff [~mGiff@ottawa-hs-209-217-124-208.d-ip.magma.ca] has quit [Client Quit]
16:22 < dreamind> TheSin: ack!
16:23 < TheSin> Clock speed was nothing
16:23 < TheSin> RISC is better even at lower clocks
16:23 < Omni|Work> But, if Apple maintains the high quality of the hardware, it doesn't matter a lot to me what CPU is inside, even if I think the architecture of the CPU is horribly ugly.
16:23 < TheSin> look at all the benchs
16:23 < dreamind> Omni|Work: well I think more apple was pissed by ibm building 3 core cpus for m$ and not delivering G5 cpus at the same cpu speed.
16:23 -!- ringerc [~craig@dsl-202-72-144-62.wa.westnet.com.au] has quit ["Whoops, someone let the magic smoke out!"]
16:23 < TheSin> Xeon 3.6s and G5s where close
16:23 < TheSin> sure it didn't win everything but the clock speed was almost half
16:23  * Omni|Work nods.
16:23 < TheSin> and half the heat
16:24 < Omni|Work> I know clock speed isn't a good measure.  I run AMD after all.  :-)
16:24 < TheSin> means quiet, less bulky PCs
16:24 < cirdan> hehe
16:24 < mdmonk> and maybe all the op codes for hacks in tools like metasploit will work against x86 Apples.
16:24 < cirdan> TheSin: they are both toasters
16:24 < cirdan> just one is small, 1 slice, the other is a commercial furnace :-)
16:24 < TheSin> cirdan, not the same though
16:24 < TheSin> I have lots of dual G5s (not the 2.7) but I don't even have a CPU fan at all
16:25 < atoussaint> theSin: my fink says: Failed: no package found for specification 'libapache2-ssl-mod-php5'!
16:25 < TheSin> I hope at least they will use EM64T or what ever the 64 bit P4s are
16:25 < TheSin> Melian, tell atoussaint about unstable
16:25 < cirdan> yeah, 64but, but still intel. ugh.
16:25 < TheSin> totally
16:25 < Omni|Work> I wish they'd use the AMD ones.
16:26 < TheSin> I'm not against Intel
16:26 < Omni|Work> AMD makes much better chips than Intel right now.
16:26 < TheSin> would have been cool if Intel just made ppcs for Apple
16:26 < dreamind> TheSin: I hope so too, but I still don't like that little endian shit.
16:26 < TheSin> I just hate the though of going back to SiSC
16:26 < dreamind> TheSin: ack ;)
16:26 < cirdan> TheSin: yeah, that blows
16:26 < cirdan> risk is good
16:26 < dreamind> risc.
16:26 < cirdan> :-p
16:26 < cirdan> i know
16:26 < cirdan> you ever watch hackers?
16:27 < dreamind> some time ago
16:27 < Omni|Work> I liked that movie.  :-)
16:27 < Omni|Work> I should get a copy.
16:27 < cirdan> great movie
16:27 < Omni|Work> Bitorrent.  Buying the DVD of it would just be wrong.  :-)
16:27 < dreamind> well I think its kinda nonsense what they tell about hacking stuff =)
16:27 < dreamind> (in the movie)
16:28 < Omni|Work> I thought of it more as metaphor than reality.
16:28 < Omni|Work> A movie with all the technical detail accurately portrayed would bore even me.  :-)
16:28 < dreamind> Omni|Work: ok ;)
16:30 -!- euthydemus [~mhough@host81-157-60-186.range81-157.btcentralplus.com] has joined #fink
16:33 -!- Murr [~neeri@A17-202-20-71.apple.com] has joined #fink
16:43 -!- att [~att@rrcs-24-123-50-242.central.biz.rr.com] has joined #fink
16:44 -!- ismith [ismith@c-24-14-254-14.hsd1.il.comcast.net] has quit ["Leaving"]
16:59 < aweil> argh.. 
17:00 < cirdan> heh
17:00 < aweil> i'll expect not to see mac's with intel inside sticker.. :S
17:00 < cirdan> hey Murr 
17:00 < cirdan> aweil: you'll see a lot of ppl go postal
17:00 < aweil> cirdan: postal?
17:03 -!- atoussaint [~atoussain@199.125.244.126] has quit []
17:11 -!- vasi [~vasi@modemcable214.145-70-69.mc.videotron.ca] has joined #fink
17:11 < vasi> wow: "Macintosh computers using Intel microprocessors do not use Open Firmware."
17:12 < cirdan> yeah
17:12 < cirdan> even sadder :-(
17:12 < cirdan> *sniff*
17:12 < mdmonk> perhaps it's an extreme view, but I feel a bit betrayed by Apple going to Intel chips. :(
17:13 < dreamind> vasi: where?
17:13 < vasi> in the Universal Binary PDF that i'm sure everyone is reading 
17:13 < vasi> mdmonk, well i don't think they would do it if they felt it wasn't considerably better
17:14 < vasi> i mean, SJ isn't exactly known for giving up on ideas he likes
17:14 -!- att [~att@rrcs-24-123-50-242.central.biz.rr.com] has quit []
17:14 < slyrus> so do we get 64-bit address spaces under x86?
17:14 < mdmonk> true. perhaps I am just too closely associating x86 architecture with MS Windows.
17:15 < mdmonk> and my issue is with MS Windows. :) So I'll back off of the x86 bashing. haha.
17:16 < Omni|Work> dia and pygtk-py2[34] both refuse to make on my 10.4 system.
17:17 < Omni|Work> x86 is an icky architecture, but that ickiness largely doesn't result in anything bad at the OS level.
17:17 < Omni|Work> Aside from the awful ancilliary hardware that's made for that platform.
17:17 < vasi> er, the apple pdf refers to x86-32
17:17 < vasi> Omni|Work, what error do you get?
17:18 < Omni|Work> I'm going to have to take another go at compiling it.
17:18 < vasi> if you show me the error, i may be able to tell you what to do
17:18 < Omni|Work> But it there are a LOT of pointer type mismatch warnings before the final error shows up.
17:18 < vasi> !lisppaste
17:18 < Melian> methinks lisppaste is a bot that lets you post large chunks of text without flooding the channel. Use it at http://paste.lisp.org/new/fink
17:25 < aweil> question: I tried fink install openoffice.org and failed (i'm using 10.3). so I went to /sw/src/openoffice.org../ and built it like openoffice.info says. It's, after more than 24hs finishing. after this, is possible to install it?
17:25 < aweil> i mean: i can follow install script steps..
17:26 < aweil> but i'll like that it be installed for fink system..
17:26 < aweil> is that possible? 
17:29 -!- asparagui [~billy@mo-71-1-109-215.dhcp.sprint-hsd.net] has joined #fink
17:33 < Omni|Work> vasi: How about I paste the last few lines of the compile, then if that doesn't work, a log of the entire thing.
17:33 < vasi> aweil, no
17:33 < Omni|Work> Or, maybe by some weird quirk, it will work this time.
17:33 < vasi> Omni|Work, that would be good
17:34 < vasi> aweil, fink is designed to build entire packages on its own....it sets up the environment particularly, so if you try to do parts yourself it's likely that something isn't quite right...and later there would be problems
17:34 < aweil> vasi: :-( if i follow the steps in installation script. and after that, modify the openoffice.info to make it like a virtual package?
17:35 < vasi> aweil, you can do it if you want, but if you ever have a problem with fink afterwards, don't ask anybody about it...cuz we won't be able to be sure if it's fink or if it's what you did
17:36 < vasi> i highly recommend for the future that you install the package 'ccache-default'....it caches the partial results of a fink build, so that if it breaks and you have to try again it it will go much faster the second time
17:36 < lisppaste> Omnifarious pasted "last few lines of dia compile" at http://paste.lisp.org/display/8857
17:37 < vasi> also, if you have trouble with a fink build and wish to fix it yourself, fix it by modifying the .info file and starting over
17:37 < vasi> then you can send that to the maintainer and everything's cool
17:37 < aweil> vasi: yes yes of course!!!! I'll be not asking for help aobut what I compiled.
17:38 < Omni|Work> vasi: I should try doing that for inkscape.
17:38 < aweil> vasi: thansk for the info on ccache-defualt
17:38 < Omni|Work> I discovered it has unregistered dependencies on libxslt in 10.4
17:38 < vasi> no problem aweil....i'm sorry your build broke
17:38 < vasi> Omni, so email the maintainer
17:39 < Omni|Work> Though, inkscape has a weird problem that makes it's usefullness limited anyway.
17:39 < aweil> vasi: Well.. I tried to know what happend.. i started from that directory, and build seemed to work.. but was unsure (also it's the first time that i used sources and was unsure about how them work (sources in the fink way)).
17:39 < Omni|Work> Under Apple's X11 server, when a window states that it's a top-level modal window, Quartz keeps it on _top_.
17:40 < vasi> Omni, you see where it says 'static declaration of 'parse_path' follows non-static declaration'
17:40 < Omni|Work> It even keeps it on top of the little sub-windows that open for selection lists inside a dialog and stuff.
17:40 < Omni|Work> vasi: Looking...
17:40 < aweil> vasi: and I didn't change anything (related to the bug that fink install raises), it worked directly with the same script..
17:40 < vasi> hmm that's weird Omni, email the maintainer :-)
17:40 < aweil> vasi: but run by hand
17:41 < vasi> aweil, but fink sets environment variables and such
17:41 < Omni|Work> vasi: Yes, I see that.
17:41 < vasi> so it's likely that something in your build is different from a regular fink OOo package
17:41 < aweil> vasi: yes!!!
17:41 < vasi> and so that's why it may cause breakage in the future :-/
17:42 < vasi> Omni, so that's the real error....if you want to try fixing it, make a new .patch file for dia that makes sure they're both non-static
17:42 < aweil> vasi: i mean that the problem wasn't with the parameters to configure or something like that.. so I didnt fall to the step of midifyng the .info to check error there..
17:42 < vasi> i'm sorry aweil
17:43 < vasi> but to manage packages, fink has to know what the packages are exactly
17:43 < vasi> if things are different, you get more problems in the future generally
17:43 < aweil> vasi: :-) no problem!
17:45 < vasi> aweil, please at least mail the maintainer and show him the error you saw
17:45 < vasi> that way the error can be fixed
17:46 < aweil> vasi: i'll do
17:48 -!- att [~att@cpe-65-24-92-24.columbus.res.rr.com] has joined #fink
17:48 -!- drm [~drm@dirac.cgtp.duke.edu] has joined #fink
17:48 -!- vasi [~vasi@modemcable214.145-70-69.mc.videotron.ca] has quit []
17:52 < neo> hello
17:52 < neo> is anybody alive ? :)
17:53 < drm> yes
17:53 < neo> :)
17:53 < neo> a little question
17:53 < neo> how do you play divx on tiger ?
17:53 < aweil> neo: hello
17:53 < drm> i don't know
17:53 < neo> hello
17:54 < neo> i used mplayer on debian
17:54 < neo> with the marrillat repository
17:55 < aweil> another question: i installed fink binaries about 3 months ago, and then switched to sources. i have a fink.old directory inside /sw, can I remove it safely?
17:55 < drm> yes
17:57 < neo> how can i read my Ogg files ?
17:57 < neo> s/read/play/
17:58 < aweil> ogg123?
17:58 < aweil> bmp?
17:58 < neo> a graphical one, please
17:58 < neo> there isn't any plugin for itunes ? :/
17:59 < aweil> bmp is graphical!
17:59 < neo> oh ok
17:59 < aweil> (sadly, like xmms ;-) )
17:59 < neo> last question
18:00 < neo> do you know what line have i to add to my sources.list to have access to the unstable branch
18:01 < aweil> i use: deb file:/sw/fink unstable main crypto
18:01 < aweil> but recommend that you check the faqs cause i'm not sure about teh commands to update the indexes
18:03 < neo> ok thanks
18:06 < cirdan> neo: there is an itunes plugin
18:08 < neo> oh
18:09 < neo> do you have any url for me ?
18:10 < neo> i have it
18:10 < neo> thank you :)
18:11 < cirdan> neo: http://www.google.com/
18:12 -!- xerxes1358 [~xerxes__@145.116.2.111] has quit ["Leaving"]
18:14 -!- xerxes1358 [~xerxes__@145.116.2.111] has joined #fink
18:17 -!- Sortova [~tarus@66.45.100.223] has quit ["Client exiting"]
18:18 < aweil> just as a color note: the directory where building the openoffice has 9.8G !
18:19 < mdmonk> yikes!
18:23 -!- TheSin [~TheSin@iphost-64-56-130-194.edm.wiband.net] has quit ["Client exiting"]
18:26 < cirdan> heh
18:28 < aweil> but, as a more colored note, http://article.gmane.org/gmane.os.apple.fink.devel/10835 says it tooks 37gb of space!!! (my hd has 30gb!! ha!)
18:28 < cirdan> weird
18:29 -!- drm [~drm@dirac.cgtp.duke.edu] has quit ["Leaving"]
18:33 < cirdan> dam! drm!
18:42 -!- euthydemus [~mhough@host81-157-60-186.range81-157.btcentralplus.com] has quit []
18:44 < gecko2> hmm, anyone tried Xcode 2.1 yet?
18:49 -!- att [~att@cpe-65-24-92-24.columbus.res.rr.com] has quit []
18:50 -!- asparagui [~billy@mo-71-1-109-215.dhcp.sprint-hsd.net] has quit ["outtie"]
18:52 < pogma> gecko2: it is downloading here
18:52 < gecko2> i finished downloading it
18:53 < pogma> fat compilers
18:53 < pogma> I think that's the major change :)
18:54 < gecko2> yea
18:54 < gecko2> ./usr/libexec/gdb/gdb-i386-apple-darwin <-- ahhh
18:55 < gecko2> ./usr/libexec/gcc/darwin/i386/as <-- and more of them
19:02 -!- aweil [~aweil@201.255.0.174] has quit [Read error: 110 (Connection timed out)]
19:17 -!- hennker [flullup@dsl-082-082-237-064.arcor-ip.net] has quit ["leaving"]
19:20 -!- You're now known as RangerRick
19:20 -!- drm [~drm@dirac.cgtp.duke.edu] has joined #fink
19:21 < pogma> drm!
19:21 < drm> hi pogma
19:21 < drm> surprise surprise :)
19:22 < RangerRick> hehe
19:22 < lisppaste> pogma pasted "question" at http://paste.lisp.org/display/8859
19:23 < pogma> drm: please help me answer
19:23 < pogma> drm: Yeah, got any hardware that runs darwin8 x86?
19:23 < drm> pogma: nope
19:24 < drm> (no to hardware qustion)
19:24 < drm> well first off, the guy needs to read his docs
19:24 < drm> and to understand that this system is *not* the same as LD_LIBRARY_PATH
19:24 < RangerRick> hah
19:25 < pogma> have to go, jessica requires me to sit next to her, bbl
19:30 -!- asparagui [~billy@mo-71-1-109-215.dhcp.sprint-hsd.net] has joined #fink
19:38 -!- asparagui [~billy@mo-71-1-109-215.dhcp.sprint-hsd.net] has quit ["outtie"]
19:38 -!- ^baba^ [~baba@YahooBB220041001026.bbtec.net] has joined #fink
19:39 < drm> hi ^baba^
19:44 < ^baba^> hi drm
19:45 -!- ^baba^ is now known as baba
19:45 < baba> universal?
19:45 < drm> universal is steve jobs' term for fat
19:45 < drm> "fat binaries" are now "universal binaries"
19:45 < baba> i see
19:45 < drm> (maybe you didn't hear what happened whil you were sleeping?)
19:46 < newmanbe> Fat binaries, brings back memories...
19:46 < baba> will apple shift towards Intel chips?
19:46 < drm> yes, baba
19:46 < baba> oh dear1
19:46 < drm> which is why we need fat binaries :)
19:46 < baba> 68m ppc nightmare!
19:47 < drm> indeed
19:47 < drm> although they ported Mathematica in 2 hours
19:47 < pogma> drm: no! no fat binaries
19:47 < pogma> drm: different x86 binaries, cool, not fat
19:47 < drm> pogma: did you see my msg on fink-devel?
19:48 < pogma> drm: Okay, I'll respond on that thread
19:48 < drm> baba: do you use sagasu, or do you know anybody who does?
19:49 < baba> sagasu is a package?  or just a japanese word?
19:50 < baba> oh GNOME tool
19:50 < drm> its a package which i guess helps to search through text, even japanese text
19:50  * drm is searching for sagasu users, ironically enough
19:50 < baba> hehe
19:50 < baba> maybe todai ppl
19:50 < drm> yeah
19:51 < baba> i remember some of them use GNOME
19:52 -!- Zugwrack [~Zugwrack@cpe-24-27-120-83.houston.res.rr.com] has joined #fink
19:53 < drm> i wonder if XCode 2.1 has an intel assembler for cross-compiling
19:53 < drm> err, x86 i mean
19:53 < newmanbe> From talk on #opendarwin, I *think* it does.
19:53 < Zugwrack> That was my understanding
19:53 < newmanbe> gcc still, not icc.
19:54 < drm> yeah
19:54 < cirdan> heh
19:54 < cirdan> hey drm
19:54 < pogma> xcode-2.1 has i386 tools
19:54 < drm> pogma: and by the way, they are called "universal binaries" now
19:54 < pogma> they are fat, despite marketing
19:54 < drm> pogma: as in "have you noticed how universal aunt martha is getting?"
19:54 < cirdan> can we please get what i have in my branch to head?
19:54 < cirdan> anyone mind? :-)
19:54 < cirdan> heh
19:54 < newmanbe> Fat was fine with PowerPC/68k.
19:55 < cirdan> i still like fat
19:55  * newmanbe does too.
19:55 < cirdan> no euphomisims here
19:55 < cirdan> :-)
19:55 < newmanbe> It is not universal.
19:55 < cirdan> right
19:55 < cirdan> it's not platform-independant
19:55  * newmanbe looks at a package on the Debian website to find other platforms. :)
19:55 < drm> newmanbe: you mean it doesn't give SPARC code?
19:55 < Zugwrack> Well I am going to hold off and see how changing over to x86 architecture affects the way that the "new" macs will have to work...I am worried that other than the OS that there will not be much point in owning a mac..
19:55 < cirdan> hehe
19:56 < cirdan> Zugwrack: will prolly still get good hardware, stable, and just be able to get inflecked with windoes bins via wine now, not just virtPC
19:56 < cirdan> whee.
19:56 < cirdan> can;t. wait.
19:56 < drm> pogma: does libtool support fat building for other OS's?
19:57 < drm> (saw your msg)
19:57 < Zugwrack> Errr..yeah...that too...heh
20:00 < Zugwrack> So did anyone hear a time frame for them to come up with one of the macs with a AMD processor? Or is it already prototyped?
20:00 < cirdan> Zugwrack: nothing
20:01 < cirdan> just apple partnered with intel
20:01 < cirdan> ugh.
20:01 < Zugwrack> makes me shudder too.....
20:01 < drm> wintel vs. mactel
20:01 < cirdan> ewww
20:01 < cirdan> make it stop
20:01 < pogma> drm: libtool supports fat building on darwin, it passes all it's tests when built tripple fat (i386 ppc ppc64)
20:01 < newmanbe> If they had to go x86, /me would have rather had them choose AMD.
20:01 < pogma> drm: but that is not the point
20:02 < Zugwrack> My bad guys..someone at work following the announcement told me AMD...apologies
20:02 < drm> pogma: the point is autoconf? or the assumptions people make when using autoconf?
20:02  * newmanbe goes to take of the "Made on a Mac" badge on a website.
20:02 < pogma> drm: other issues include dirs for some packages (e.g. ruby) being powerpc-apple-darwin won't work in i386
20:02 < newmanbe> s/of/off
20:02 < drm> Zugwrack: the keynote address was presented using a mac-on-intel prototype
20:02 < pogma> drm: assumptions made using autoconf
20:02 < pogma> drm: checking endianness, sizeof() etc
20:03 < Zugwrack> drm: Thanks...
20:03 < cirdan> right, quite a few things hardcode the processor
20:03  * newmanbe plays Bing Crosby's White Christmas on the piano. :)
20:03 < cirdan> it's gonna be a looong year
20:03 < drm> pogma: ok, but, these are errors :)
20:03 < cirdan> just when upstream ppc support was getting decent
20:03 < cirdan> apple's gotta ruin it all
20:03 < RangerRick> hehe
20:03 < cirdan> !lart steve jobs
20:03  * Melian explains, ever so gently, that if steve jobs doesn't give the channel more information, they can't help
20:03 < cirdan> bah
20:04 < cirdan> Melian: that wasn't very good
20:04 < Melian> bugger all, i dunno, cirdan
20:04 < drm> !lart steve jobs
20:04  * Melian installs PocketPC on steve jobs's PDA
20:04 < cirdan> haha
20:04 < cirdan> how true
20:04 < cirdan> mm
20:04 < cirdan> thunder strom
20:04 < cirdan> :-)
20:05 < cirdan> pogma: even dpkg/apt itself hardcodes the target platform
20:05 < cirdan> in every deb
20:05 < drm> just in the name, cirdan? or in other ways?
20:06 < drm> filename
20:06 < newmanbe> That kept /me from install a PowerPC binary on an x86.
20:06 < newmanbe> s/install/installing
20:06 < cirdan> more than that
20:06 < Zugwrack> Cya all...have to go eat...after mowing front and back lawns in 95 degree weather I am hungry after cooling down..heh
20:06 -!- Zugwrack [~Zugwrack@cpe-24-27-120-83.houston.res.rr.com] has quit ["Leaving"]
20:07 < cirdan> it's hardcoded into every control file
20:07 < drm> well, ok, but we would call the architecture "universal"
20:08 < newmanbe> Fat!
20:08 < cirdan> drm: universal is wrong
20:08 < drm> maybe need to change dpkg to let us install universals as well as arch-specific
20:08  * RangerRick notes it's really easy with rpm  <grin>
20:08 < cirdan> the debs are not 'universal'
20:08 < drm> cirdan: whatever... i don't want to argue about the name, i want to argue about the principal :)
20:08 < cirdan> RangerRick: hence rpm sucks :-)
20:09 < cirdan> i vote against fat binaries
20:09 < cirdan> for fink anyway
20:09  * newmanbe seconds cirdan.
20:09 < drm> if the contents of the deb are suitable for installation on more than one architecture, we should go for it
20:09 < cirdan> larger files, longer compile time
20:09 < cirdan> boo
20:09 -!- asari [~asari@newsodan.sodan.ecc.u-tokyo.ac.jp] has joined #Fink
20:09 < asari> logs?
20:09 < drm> to addrss longer compile time, i suggested there would be a user configure option
20:10 < newmanbe> !logs
20:10 < Melian> logs are at http://meme.b9.com/cview.html?channel=fink&date=today or http://fink.aquaflux.org (currently down)
20:10 < Omni|Work> I don't understand why there's really any difference between and rpm and a deb.
20:10 < cirdan> building kde once every 2 weeks is enough, i dont need it taking longer
20:10 < drm> asari: do you use the package "sagasu" or do you know anyone who does?
20:10 < asari> thanks, newmanbe :)
20:10 < cirdan> Omni|Work: it's all in the total system
20:10 < Omni|Work> Ahh, not in the package format itself.
20:10 < cirdan> Omni|Work: rpm is like tar with a install.log :-)
20:11 < cirdan> dpkg tries to force you not to screw up
20:11 -!- dreamind [~dreamind@C2107.campino.wh.tu-darmstadt.de] has quit [Read error: 60 (Operation timed out)]
20:11 < asari> drm: no, i haven't even heard of it
20:11 < drm> asari: ok... its a gnome searching tool, supposed to be good for searching japanese text
20:11 < drm> asari: i'm trying to find users to discuss upgrading to gnome 2.x, but maybe there aren't any :)
20:11 < cirdan> drm: so, we'd have to always override configure things like arch=darwin-universal and all, for fat builds
20:11 < RangerRick> cirdan: that's totally not true, but I won't argue such things right now  :)
20:11 < asari> drm: sounds like "namazu"
20:12 < cirdan> RangerRick: shhh!
20:12 < cirdan> or is this for #od?
20:12 < cirdan> :-)
20:12 < newmanbe> No, DarwinPorts doesn't think ahead like use. :-p
20:12 < RangerRick> cirdan: no, if it were #od, I *would* argue such things right now
20:12 < asari> drm: is it this one? http://www3.sympatico.ca/sarrazip/dev/sagasu.html
20:12 < newmanbe> s/use/us
20:12 < cirdan> heh
20:12 < drm> asari: it's already a fink package
20:13 < cirdan> no, i mean i told toby to tell u hot to change the process title in ps
20:13 < drm> asari: and yes, that's the pkg
20:14 < asari> seems a little old now
20:15 < drm> yeah... the fink version is way out of date... i will update it
20:16 < asari> great
20:18 -!- aweil [~aweil@201.255.0.72] has joined #fink
20:22 < Omni|Work> I will have to try Sagasu.
20:32 -!- akh [~akhansen@68-118-244-23.dhcp.oxfr.ma.charter.com] has joined #fink
20:39  * newmanbe realises he doesn't care if there are fat binaries because he builds from source. :-)
20:42 < akh> Yah
20:42 < asari> RangerRick: are you here?
20:42 < Omni|Work> I suspect the move from PPC to x86 will be much less traumatic then the move from 68000 to PPC.
20:43 < Omni|Work> The state of software as a whole is very different now from where it was then.
20:43 < RangerRick> asari: yes
20:43 < asari> RangerRick: building "ant" fails in Japanese environment.
20:44 < asari> RangerRick: I recently tried installing Mac OS X as English environment and build ant, and succeeded...
20:44 < RangerRick> asari: yeah, I don't have a japanese environment, so I'm not sure how to go about reproducing it
20:44 < asari> yeah :(
20:45 < RangerRick> asari: any ideas what constitutes a "japanese environment" so that I can turn it off during the build?  :)
20:45 < asari> LANG or LC_ALL was not that
20:45 < RangerRick> hm
20:46 < asari> and java still outputs Japanese massages even I set LANG=C
20:47 < aweil> other question: installing xorg package will not interfere with default xfree provided by panther, am i right?
20:47 < RangerRick> aweil: no, that is not right
20:47 < RangerRick> aweil: they all use /usr/X11R6
20:47 < aweil> RangerRick: :-(
20:47 < aweil> RangerRick: thanks!
20:47 < akh> But you can install applex11tools and get X11.app back.
20:48 < RangerRick> yeah
20:49 < aweil> akh: ah.. cool.. I heard that xorg will have enhancements over xfree86.. apple's xserver is based on that, right?
20:49 -!- drm [~drm@dirac.cgtp.duke.edu] has quit ["Leaving"]
20:50 < RangerRick> xfree86 4.3 < apple X11 (panther) < xfree86 4.4 < x.org 6.7 < apple X11 (tiger) < x.org 6.8    :)
20:50 < RangerRick> as far as bugfixes and optimization
20:51 < RangerRick> apple's x11s have some features over xfree86 and x.org, but you can get most of those features by installing applex11tools
20:51 < aweil> RangerRick: wow. thanks! 
20:52 < aweil> Ok.. is compatible being "conservative" and installing xorg from fink? :-) (yes i know it's on unstable stage..)
20:53 < aweil> Is apple's x11 optimized for it's hardware and xorg's one not?
20:54 < newmanbe> I think Apple's has optimisations for Quartz.
20:54 < aweil> sorry.. i know that you may have not idea about this.. (i don't want to be hated here :-) )
20:54 < newmanbe> There have been too few people needing help recently. :) I was getting bored.
20:54 < RangerRick> they're all optimized for hardware
20:54 < RangerRick> apple made a library to aid the X server to talk to quartz
20:55 < RangerRick> and every x11 greater than xfree86 4.3 uses
20:55 < RangerRick> it
20:55 < RangerRick> :)
20:55 < mdmonk> RR: the xorg build we did this morning works much better than my previous xfree86 install.
20:55 < aweil> cool!!
20:55 < aweil> thanks!
20:55 -!- clepple [~clepple@dsl092-164-214.wdc2.dsl.speakeasy.net] has joined #fink
20:57 < Omni|Work> RangerRick: I've noticed a weird difficulty with Apple's X11.
20:57 < RangerRick> Omni|Work: ?
20:58 < Omni|Work> In apps that put up a window where they say "This is a modal, on top window!" Quartz does it very religiously.
20:58 -!- baba [~baba@YahooBB220041001026.bbtec.net] has quit [Read error: 110 (Connection timed out)]
20:58 < Omni|Work> So much so that when that window spawns a little temporary top-level window for a drop-down list or something, that window is udner the dialog.
20:58 < Omni|Work> This is mostly noticeable in some of the inkscape dialogs.
20:59 < akh> There was a post on the lists about that.
20:59 < RangerRick> can't help with that, open a bug to apple  :)
20:59  * Omni|Work nods.
20:59 < Omni|Work> I should.
20:59 < newmanbe> The black hole called Radar!
20:59 < Omni|Work> I'm not sure how yet.  :-)  But I can probably figure it out if I poke around their site.
21:00 < Omni|Work> The true difference between an Open Source project and a closed source one is how easy it is to report bugs.  :-)
21:01 < Omni|Work> Have fun all.
21:01 -!- Omni|Work is now known as Omni|AFK
21:02 < cirdan> hmm
21:02 < cirdan> !seen drm
21:02 < Melian> drm <~drm@dirac.cgtp.duke.edu> was last seen on IRC in channel #opendarwin, 13m 39s ago, saying: 'benn you're gonna lose in the long run :)'.
21:02 < cirdan> d'oh
21:02 < cirdan> fink is low on cash
21:03 < akh> Time to charge $ to answer questions that are in the docs.
21:04 < cirdan> hehe
21:04 < cirdan> we'd be very wealthy 
21:04 < akh> Yup.
21:04 < akh> mmm...unstable bindist server farm.
21:04 < cirdan> next month fink will not have enough cash in the checking acct, it costs us $25/month to have a bank account
21:05 < akh> Ah, yeah--the wife has the same problem.
21:05 < cirdan> we currently have $35
21:05 < cirdan> so one more month
21:05 < cirdan> :-)
21:05  * cirdan needs to get drm to apply for a corperate credit card :-)
21:08 < aweil> do you accept Argentinian pesos? ;-)
21:09 < mdmonk> I have some euros remaining from my visit to Italy.
21:09 < cirdan> sure
21:09 < cirdan> just take it to the local money exchange :-)
21:10 < cirdan> i gotta call our lawyer about our tax-exempt status
21:10 < cirdan> !lart the irs
21:10  * Melian takes out the irs with the trash
21:10 < akh> !w00t
21:10 < cirdan> !wot
21:10 < cirdan> !woot
21:10 < Melian> woot!
21:11 < akh> ah
21:12 -!- emp [~emp@67-42-224-94.blng.qwest.net] has quit [Read error: 104 (Connection reset by peer)]
21:13 -!- emp [~emp@67-42-224-94.blng.qwest.net] has joined #fink
21:13 < akh> [OT] /me had to edit my sources.list on my Debian box.
21:13 < cirdan> why?
21:14 < akh> 3.1 was officially released today.
21:14 < cirdan> yeah, got the email
21:14 < cirdan> i run unstable anyway
21:14 < cirdan> :-)
21:14 < cirdan> woot!
21:14 < cirdan> hmm
21:15 < cirdan> Melian: woot!
21:15 < Melian> Woo-Hoo!
21:15 < cirdan> heh
21:15  * akh couldn't get to non-US/unstable.
21:15 < cirdan> weird
21:16 < aweil> hehe i told about the argentinian pesos cause we got them devaluated from 1 to 1 dolar to 3 for each dolar .. 
21:16 < cirdan> ?
21:16 < cirdan> dont quite follow
21:17 < cirdan> holy crap...
21:17 < cirdan> -users turned into -talk
21:17 < akh> ?
21:17 < cirdan> 56 new msgs
21:17 -!- tux0beliver [~tux0beliv@141.151.18.62] has joined #fink
21:18 < akh> sounds like Sourceforge latency.
21:18 < cirdan> yeah
21:18 < tux0beliver> hey peeps!
21:18 < akh> hi
21:19 -!- theid [~theid@207.177.103.77] has quit [Read error: 110 (Connection timed out)]
21:25 -!- dmacks [~dmacks@netspace.org] has joined #fink
21:25 -!- msachs [~msachs@206.13.39.65] has joined #fink
21:26 < msachs> hey
21:26 < msachs> I just got some interesting news.
21:26 < msachs> There's a talk at WWDC tomorrow by one of the x86 guys on their experiences porting Fink and FireFox.
21:26 < msachs> (To Tiger/x86.)
21:27 -!- xerxes1358 [~xerxes__@145.116.2.111] has left #fink ["Leaving"]
21:27 < msachs> Apparently the work involved in porting Fink was pretty trivial.
21:27 < dmacks> Wow! Is there gonna be a archived video?
21:28 < msachs> I'll be taking notes.
21:28 < msachs> If there is a video, I don't think people are supposed to redistribute it.
21:28 < dmacks> Probably the only difficulty is how specific some patches are ("make it work on ppc/osx under fink" not "make it *also* work...")
21:29 < msachs> The engineer was just here talking to me and my boss, he said that the only thing was there were a couple of places where Fink was assuming that Darwin == powerpc.
21:29 -!- asari [~asari@newsodan.sodan.ecc.u-tokyo.ac.jp] has quit ["Quitting!"]
21:29  * dmacks uses fink (the core program) on linux, no problem.
21:29 < msachs> dmacks: Well that's not Darwin, is it? :)
21:30 < dmacks> Good point. (/me wrote that before I got the "assume darwin==ppc" msg)
21:31 -!- msachs [~msachs@206.13.39.65] has quit []
21:31 < RangerRick> yeah, I use fink on linux as well for the rss-feed generator
21:32 < dmacks> And actually the pdb generator runs (okay "runs") on SF's whatever-the-hell-underpowered-box :)
21:33 < cirdan> hehe
21:33 < cirdan> hey dmacks
21:33 < cirdan> RangerRick
21:33 < cirdan> fink used to run on darwin x86
21:33 < cirdan> bbraun helped there some
21:33 < cirdan> but we neglected te support :-)
21:36 -!- swix_ [om@u1.omx.ch] has quit [Read error: 104 (Connection reset by peer)]
21:37 -!- swix [om@u1.omx.ch] has joined #fink
21:37 -!- tux0beliver [~tux0beliv@141.151.18.62] has quit ["Get MacIrssi - http://www.g1m0.se/macirssi/"]
21:37 < dmacks> RangerRick: Ya wanna add the XTools->xfree86->apple-watever->xfree86-something->xorg timeline to Fink FAQ or Using X11 doc?
21:39 < aweil> add it!! :D
21:39 < akh> It's there (sans XTools)
21:39 < cirdan> hehe
21:45 < dmacks> Ah, found it. Kinda convoluted (bare numbers means "versions of the fink xfree86 pkg"?) :(
21:45 < dmacks> Maybe we need a doc-a-thon, make a sweep and update everything to mention Tiger and current pkg versions.
21:52  * akh is kind of working on it.
21:52 < akh> I probably could commit what I have.
21:52 < dmacks> akh: /me was hoping someone would say "hey, I can host that at my place and even provide the beer"
21:52 < akh> I could do that, too.
21:54 < newmanbe> cirdan: When did Melian start monitoring #od?
21:54 < newmanbe> Haha. /me saw what drm was last seen saying in #od.
21:55 -!- jerwin [~jeremy@66-44-58-38.s38.tnt3.lnhva.md.dialup.rcn.com] has joined #fink
21:56 < mdmonk> later all. 
21:56 < newmanbe> Bye.
21:56 -!- mdmonk [~mr_twitch@64.81.110.110] has quit ["time"]
22:00 -!- emp [~emp@67-42-224-94.blng.qwest.net] has quit [Read error: 145 (Connection timed out)]
22:02 -!- emp [~emp@67-42-224-94.blng.qwest.net] has joined #fink
22:03 < newmanbe> Ah, I didn't see that Melian is in #opendarwin.
22:03 -!- Omni|AFK [~hopper@Omnifarious.sustaining.supporter.pdpc] has quit [Read error: 104 (Connection reset by peer)]
22:04 -!- Omni|AFK [~hopper@2002:d8fe:1010:1:204:e2ff:feb2:7b2d] has joined #fink
22:04 < akh> She swings both ways?
22:05 < dmacks> Yeah...she wants dependencies and reliable compiling, but prefers tcl.
22:05 < dmacks> #od: I keed, I keed.
22:06 -!- jerwin [~jeremy@66-44-58-38.s38.tnt3.lnhva.md.dialup.rcn.com] has quit ["Leaving"]
22:07 < akh> heh
22:19 -!- bakshi [~bakshi@S0106001217ba40c7.cg.shawcable.net] has joined #fink
22:21 < dmacks> Any Todai folks here?
22:27 < akh> maybe not
22:28 < dmacks> Yup:(
22:31 < aweil> another "faq like" question :-): is there a way to compile packages depending on gcc 3.1 with the current gcc? (maybe modifing the local packagename.info) ?
22:33 < dmacks> That's a two-parter...the easy (second) part is under the general "compiling my own package or an existing packaeg with non-standard options"
22:34 < aweil> dmacks: is there some first part?
22:34 < dmacks> The hard part "make it work with gcc3.1" is the difficult one. Everyone's working to eliminate all traces of gcc3.1 (since Tiger doesn't have it). But some things really do need it at this time.
22:34 < dmacks> Make that "make it work with gcc3.3 or gcc4.0"
22:35 < aweil> ah.. oki
22:35 < aweil> but for trying, making the "fake" package should work..
22:36 -!- Experimen [~Jimmy@cpc4-stoc5-3-0-cust74.midd.cable.ntl.com] has joined #fink
22:36 < Experimen> So you guys seen the WWDC video with Steve Jobs?
22:36 < dmacks> Right. You're welcome to try stuff, and we'd love to hear any success stories so we can update the packages.
22:36 < Experimen> The intel rumor is true.
22:36 < newmanbe> We heard.
22:37 < aweil> dmacks: thanks!
22:37 < newmanbe> !lart not-working QuickTime movies
22:37  * Melian blasts not-working QuickTime movies with a huge firehose then strangles not-working QuickTime movies with it
22:37 -!- Experimen [~Jimmy@cpc4-stoc5-3-0-cust74.midd.cable.ntl.com] has quit [Remote closed the connection]
22:39 < dmacks> Experimen: #fink's latency is less than /.'s even, and usually much more accurate:)
22:39 < akh> Mmm...faster powerbooks.
22:41 < dmacks> http://www.ilovebacon.com/060605/darth.jpg
22:42 < newmanbe> 'dat you? :)
22:43 < dmacks> Naw...I didn't dress up when I went.
22:44 < dmacks> Only movie I've ever attended dressed-up is Rocky Horror.
22:45  * newmanbe makes bundles that install OpenOSX CD stuff. :-p
22:46 -!- You're now known as RangerAway
22:48  * akh will go this weekend.
22:49 < Knghtbrd> newmanbe: eww
22:51 < newmanbe> "And all of the source code is completely editable, so if phpShop doesn't do it, you may modify it so it does."
22:51 < Knghtbrd> EW EW EW EW
22:51 < newmanbe> If you know PHP, chances are, you don't need OpenOSX.
22:51 < Knghtbrd> OpenOSX is teh devile
22:51 < Knghtbrd> I mean really
22:52 < newmanbe> I am thinking of writing them a letter telling them they spell Fink wrong, and that it is not archaic.
22:52 < akh> Fink archaic?  Eftsoons!
22:52 < dmacks> newmanbe: Considering that they did (and possibly/probably still do) use Fink packages in their distro...
22:52 < Knghtbrd> newmanbe: I would rather you just clone their product offerings for free to piss them off  ;)
22:53 < newmanbe> Yep.
22:53 -!- theid [~theid@207.177.103.77] has joined #fink
22:53 < RangerAway> OpenOpenOSX
22:53 < newmanbe> Haha.
22:53 < akh> hehe
22:53 < dmacks> Ha!
22:54 -!- akh is now known as akh_sleeping
22:54 < akh_sleeping> zzzzzzzz
22:57 -!- newmanbe [~newmanbe@f4f383f1ecd68887.session.tor] has quit ["Leaving"]
23:07 < cirdan> heh
23:08 < cirdan> F(ree)OpenOSX
23:08 < cirdan> FOpenOSX
23:22 -!- slyrus [~slyrus@adsl-68-121-172-169.dsl.pltn13.pacbell.net] has quit [Remote closed the connection]
23:30 -!- Murrito [~neeri@c-24-7-119-74.hsd1.ca.comcast.net] has joined #fink
23:42 -!- aweil [~aweil@201.255.0.72] has quit ["Leaving"]
23:58 -!- aweil [~aweil@201.255.0.72] has joined #fink
--- Log closed Tue Jun 07 00:00:39 2005