Timestamps are in GMT/BST.
[1:21] * gyges (~gyges@gyges.user) has joined #freenet
[1:22] * gyges (~gyges@gyges.user) has left #freenet
[1:41] * gvdm (~gvdm@210-246-21-73.paradise.net.nz) has joined #freenet
[2:25] * sanity_ (~ian@81-178-103-81.dsl.pipex.com) Quit ()
[2:48] * verl (verl@h155n2fls33o877.telia.com) Quit (Read error: 104 (Connection reset by peer))
[2:56] * sanity_ (~ian@81-178-103-81.dsl.pipex.com) has joined #freenet
[2:56] * sanity_ (~ian@81-178-103-81.dsl.pipex.com) Quit (Client Quit)
[4:01] <gvdm> i get the general impression that freenet id kinda dead
[4:01] <gvdm> is*
[4:02] <gvdm> i get the general impression that freenet id kinda dead
[4:02] <gvdm> especially after reading FillaMent
[4:45] * sanity_ (~ian@81-178-103-81.dsl.pipex.com) has joined #freenet
[4:53] * RallosZek (1000@d14-69-248-202.try.wideopenwest.com) has joined #freenet
[5:09] * pittaman (~mistery_b@d54C32DBA.access.telenet.be) has joined #freenet
[5:38] * hackeron (~hackeron@81-86-159-146.dsl.pipex.com) has joined #freenet
[5:47] * nextime (~nextime@213-140-22-64.fastres.net) Quit (Read error: 54 (Connection reset by peer))
[5:51] <hackeron> hmm, what do you guys think of Mute? http://freshmeat.net/projects/mute/
[5:52] <gvdm> well im talking coz hopefully that will attract the rest of the dead channels attention, bu no, i dont even know what it is
[5:52] <gvdm> but*
[5:52] * nextime (~nextime@213-140-22-64.fastres.net) has joined #freenet
[5:53] * oierw` (mathew@cpe-66-91-92-246.hawaii.res.rr.com) Quit (Read error: 54 (Connection reset by peer))
[5:53] * oierw` (mathew@cpe-66-91-92-246.hawaii.res.rr.com) has joined #freenet
[6:01] <hackeron> Just wondering, since freenet gets its anonymity from feeding the data through several computers before it reaches its destination, is data encryption even needed?
[6:10] * gvdm (~gvdm@210-246-21-73.paradise.net.nz) Quit ("I'm quitting")
[6:11] * Hory (~Miranda@82.76.81.56) has joined #FreeNet
[6:25] * sleon|tuX (test@p54A14AF9.dip.t-dialin.net) has joined #freenet
[7:01] <toad_> hackeron: yes, we do need the encryption
[7:01] <toad_> and as far as freenet being dead goes... wait a few days, stick around on the mailing lists...
[7:01] <toad_> certainly the network isn't dead
[7:02] <hackeron> toad_: who said freenet is dead? -- I love and use freenet :)
[7:02] <toad_> the doubter has left :)
[7:02] <hackeron> toad_: also, why do you think encrypting all data is that important? -- slows the hell out of my PC :)
[7:02] <hackeron> if you cant tell where data is coming from anyway, whats the point of encrypting it?
[7:05] <toad_> crypto does not slow freenet very much
[7:05] <toad_> freenet slows freenet
[7:05] <toad_> :|
[7:05] <toad_> symmetric crypto at least is virtually free
[7:06] <hackeron> no, slows my computer down, not freenet. CPU usage is at 100%, computer is crawling
[7:06] <toad_> asymmetric crypto is rather more expensive... we use it 1) to secure node-to-node connections, and 2) to let us have SSKs
[7:06] <hackeron> toad_: still, is there a *reason* for encrypting anything that moves?
[7:06] <toad_> hackeron: it's good practice, and it prevents a wide range of attacks
[7:06] <hackeron> toad_: like?
[7:07] <toad_> hackeron: it's probable that the slowness is caused by routing and so on rather than crypto
[7:07] <hackeron> toad_: routing eats 100% cpu?
[7:07] <toad_> iirc last time it was something like 1/3rd in profiling for asymmetric crypto.. and nil for symmetric
[7:07] <toad_> some for hashing too probably, but certainly less than 10% unless your link is real fast
[7:08] <toad_> hackeron: umm, like the fact that an attacker could see what you are transferring!
[7:08] <hackeron> toad_: SO!?
[7:08] <toad_> furthermore he could see whether you are relaying it
[7:08] <hackeron> toad_: you transmit random data
[7:08] <hackeron> toad_: you store 1gb or more of cache that gets transmitted backwards and forwards
[7:08] <toad_> therefore whether you are the source of the request!!
[7:08] <hackeron> toad_: how would you know if I am the source?
[7:09] <toad_> hackeron: if you are receiving the data but not sending it, you are probably the source
[7:09] <toad_> errr
[7:09] <toad_> request source
[7:09] <toad_> if you are sending it but not receiving it, you are likely the data source
[7:09] <toad_> request source = node that wants the data, data source = node that had the data in its store
[7:10] <hackeron> toad_: well, ok. How is it different that transferring the data once, so you know the md5 of the transferred "encrypted" data, then tracking for that?
[7:10] <toad_> hackeron: freenet does lots of things apart from crypto
[7:10] <hackeron> toad_: exactly. So what is crypto for?
[7:10] <toad_> hackeron: eh?
[7:11] <hackeron> toad_: the data my PC sends is data I didnt request, right?
[7:11] <hackeron> (atleast some of it)
[7:11] <toad_> crypto is 1. for node to node connections, to prevent an attacker who can surveil you seeing what you are requesting. 2. for deniability (we encrypt CHKs). 3. for SSKs.
[7:12] <toad_> hackeron: some of it is, but it is trivial for an attacker who can read all your connections to identify what you are requesting
[7:12] <hackeron> toad_: yeah, thats all fine, but why encrypt the actual data?
[7:12] <hackeron> as in the file being transmitted
[7:13] <hackeron> as long as the attacker doesnt know you requested it, he can figure out what you are transferring anyway, so why encrypt it?
[7:13] <toad_> hackeron: elaborated 2: it's encrypted so that you, or an attacker, cannot know what it is, unless you find the URI somewhere.
[7:13] <toad_> it's encrypted at the key level so that you have deniability, basically
[7:13] <toad_> and so an attacker can't do content based attacks without having to find the keys of all the content he wants to block
[7:13] <toad_> (by fetching them!)
[7:13] <hackeron> toad_: wait, if its encrypted at the key level, that means the source knows who requested it?
[7:14] <toad_> hackeron: no
[7:14] <hackeron> well, if he doesnt, is it a problem to find the keys for the content?
[7:14] <toad_> hackeron: there are 2 kinds of keys, SSKs and CHKs. CHKs are CHK@<routing key = hash of encrypted data>, <decryption key>/<filename>
[7:15] <toad_> the node only knows the routing key
[7:15] <toad_> only the end user node knows the decryption key
[7:15] <toad_> hence all the CHKs in your store are encrypted, and you can't read them, unless you can find the key
[7:16] <toad_> you can't brute force it on CHKs (only on KSKs, and sloppy SSKs - but not good SSKs)
[7:16] <toad_> finding the key is quite hard and usually involves propagating the data you want to attack
[7:17] <toad_> it's encrypted at the link level to prevent an attacker from being able to see what you are sending, what you are receiving, and deduce what you have requested
[7:17] * pittaman (~mistery_b@d54C32DBA.access.telenet.be) Quit ("I'm gone for ever")
[7:17] <toad_> it's encrypted at the key level so you don't have plaintext in your store, and so an attacker can't see plaintext
[7:17] <toad_> what's your CPU?
[7:18] <toad_> are you using an excessive logLevel?
[7:18] <toad_> how much RAM did you give it?
[7:19] * toad_ wonders if that's just pittaman's tagline or if he actually means it
[7:22] <hackeron> toad_: 1ghz Via C3, 512Mb, I think I gave it 16Mb, but its using several hundred anyway :)
[7:23] <toad_> okay, well, if you set the -Xmx (memory allocation param) too low, it will use more CPU for garbage collection
[7:23] <toad_> and it will constantly get OutOfMemoryErrors
[7:23] <toad_> and probably get into infinite loops
[7:23] <hackeron> its actually off atm, just compiling some stuff. Compile unbearably slow with freenet on
[7:23] <toad_> you need it set to at least 128M
[7:23] <toad_> 192M preferably
[7:24] <toad_> linux or doze?
[7:24] <hackeron> bah, what do you take me for :) -- ofcourse linux
[7:24] <toad_> VIA C3 is loosely based on the Cyrix 6x86 chips... it has rather slow floating point, yes?
[7:25] <toad_> first thing i'd do would be to relax the memory usage restrictions anyway - but if you really had set it to 16M, i'd be amazed if it worked AT ALL
[7:25] <hackeron> its actually the nehemiah core, (C3-2), quite different to cyrix these days. Has sse, mmx, all the standard i686 stuff (seems to work with -march pentium3 also)
[7:26] <toad_> sure, but even its SSE is slow, right?
[7:26] <hackeron> original C3 wouldnt work with -march i686
[7:26] <hackeron> well, its passively cooled and not to fast, no.
[7:26] <toad_> odd... the first 686s didn't have SSE
[7:27] <hackeron> hmm
[7:27] <toad_> only came in with the P4
[7:27] <hackeron> well, -mcpu doesnt enable sse :)
[7:27] <toad_> isn't it -march pentium4?
[7:27] <toad_> except the optimizations would be all completely wrong probably for pentium4
[7:28] <toad_> anyway things to check:
[7:28] <hackeron> pentium4 enables sse2, no?
[7:28] <toad_> 1. did you set the memory usage limit too low? you need to set it to around 128M, maybe 192M. if it's too low it won't work and will use 100% cpu. VIRT doesn't matter, it's RES which is the actual memory in use in top.
[7:29] <toad_> hackeron: yes but the P4 is a perverse super-long-pipelines architecture (could be called an expensive DSP even!)
[7:29] <toad_> hackeron: significantly different optimizations to something like the C3 probably
[7:29] <hackeron> well, with 128kbps L2-cache, sure :)
[7:29] <hackeron> even -O3 chockes the C3 pipes
[7:30] <toad_> only 128K? ouch
[7:30] <toad_> at least you have integrated memory controller
[7:31] <hackeron> amazing though, transmeta have 2mb L2-cache, a tenth of the size, same performance, 20Gb takes around 250W peak including north and south bridges :)
[7:31] <toad_> 2. your CPU may just be slow. :( Symmetric crypto probably doesn't use much CPU. Hashing uses a little more. Routing uses more. As do a few other things.
[7:31] <hackeron> if only transmeta sold like these crappy VIAs :(
[7:31] <toad_> hackeron: 250W? that's a lot for a laptop ;)
[7:31] <hackeron> toad_: err, 20ghz, sorry :)
[7:32] <hackeron> toad_: typical 1ghz takes less than a single watt
[7:32] <toad_> ah you propose to overclock them by a factor of 20
[7:32] <toad_> that sounds like fun
[7:32] <hackeron> no, lol
[7:32] <hackeron> I'm talking about clustering
[7:32] <toad_> i see
[7:32] <hackeron> http://www.orionmulti.com/download/pdf/Orion_DT-12_datasheet.pdf
[7:33] <hackeron> tivo sized transmeta cluster :)
[7:33] <hackeron> 24GB ram, 19ghz, 1Tb, 220W peak :)
[7:34] <toad_> looks nice ! :)
[7:34] <toad_> probably expensive... :)
[7:34] <hackeron> pentium4 gives you 3 ghz with 1gb ram and takes something like 400W, lol -- crap pos :)
[7:34] <toad_> also a cluster is only good for cluster-capable apps
[7:34] <toad_> hackeron: well, most apps can't be easily parallelized
[7:34] <toad_> that's the problem
[7:34] <toad_> most that can don't cluster well
[7:34] <hackeron> well, if you want the entire 19ghz, 24GB ram, 1Tb disk - it runs at about 10k, but you can get a few nodes and build up :)
[7:35] <hackeron> toad_: apps that require that sort of speed cluster quite well :)
[7:35] <toad_> it would for example be a lot of work for me to parallelize my simulations
[7:35] <hackeron> toad_: also, look at opemosix
[7:35] <hackeron> toad_: it emulates SMP over network
[7:35] <toad_> hackeron: openSSI is better
[7:35] <toad_> http://openssi.org/cgi-bin/view?page=openssi.html
[7:36] <toad_> openmosix does not really convincingly emulate an SMP box
[7:36] <toad_> but in any case, both of these systems require the apps to be divided into processes - they cannot parallelize threads over a network
[7:36] <hackeron> hmm, I tried it briefly on a 10mbps network, it didnt perform too well on such a slow and busy network, but I could send any process to any box
[7:36] <toad_> because even if that were possible (which it probably isn't), it'd be WAY too slow to be useful
[7:37] <hackeron> err, openmosix parallizes threads over a network, no?
[7:37] <toad_> yeah, I'd like to have an SSI cluster eventually... when I manage to get my parents onto linux :)
[7:37] <toad_> hackeron: no
[7:37] <toad_> hackeron: nothing does
[7:37] <toad_> hackeron: threads have to be within a single domain for acceptable performance
[7:38] <toad_> unless they're not really threads and don't talk to one another in which case you could do it but it'd be pointless
[7:38] <toad_> well it wouldn't be pointless - but it'd only be useful on a few apps
[7:38] <hackeron> ah, I see
[7:38] <toad_> the whole point of SMP is that you have consistent caches and memory
[7:38] <toad_> and tons and tons of bandwidth between chips
[7:39] <hackeron> yeah, suppose thats a problem. Still, where do you think the future is, clustering or making single monster CPUs?
[7:39] <toad_> both, of course
[7:39] <toad_> and monster SMPs, and small SMPs, and single chip SMPs
[7:39] <hackeron> well, openmosix spread ram all over the network
[7:39] <toad_> not exactly
[7:39] <toad_> single chip SMPs will get big - but slowly because of software issues
[7:39] <hackeron> it emulated the combined ram amount so to the programs, it seemed like I had several gigs of ram :)
[7:39] <toad_> heh
[7:40] <toad_> you don't actually get to use it like that...
[7:40] <toad_> network swap is possible but slow
[7:40] <toad_> well, if it's 100M ethernet it's not too slow... if it's 1G ethernet, it's probably fine
[7:40] <hackeron> well, they have their distributed filesystem
[7:40] <toad_> there are loads of distributed filesystems
[7:40] <hackeron> well, the transmetas have 1gbps onboard network on EACH node :)
[7:41] <toad_> but mosix is really old tech, it's very primitive, it only distributes a small fraction of the total system image
[7:41] <hackeron> thats 12nodes in a tivo sized box :)
[7:41] <toad_> openSSI is _much_ nicer
[7:41] <hackeron> hmm, I'll check openSSI out, thanks :)
[7:42] <toad_> hopefully openSSI will go into linux 2.7 some time in the next few years
[7:42] <hackeron> PS, you know NPTL or those so called light threads? -- dont they stop program from being clusterable?
[7:42] <toad_> not exactly. only true processes can be migrated.
[7:43] <hackeron> but they are "light" processes. They dont even show up in ps
[7:43] <toad_> ANY kind of threads have to stay on the same node - you may be able to move the entire process with all threads, but you can't have one thread on one node and another on another node
[7:43] <toad_> well it might be possible with some clever memory management - but it'd be really slow if the threads actually communicated with one another much
[7:43] <hackeron> toad_: hmm, well, say apache opens new threads that you can migrate, no?
[7:44] <toad_> hackeron: apache can easily run with processes instead of threads
[7:44] <toad_> hackeron: it always used to
[7:44] <toad_> hackeron: oh, and apache is pathetically slow anyway :)
[7:44] <hackeron> lol
[7:44] <hackeron> so you dont like apache? :)
[7:44] <toad_> i like apache
[7:44] <toad_> but it is slow
[7:45] <hackeron> know any good alternatives?
[7:45] <toad_> for db-driven dynamic sites it's probably no slower than anything else
[7:45] <hackeron> I'm learning zope atm :) -- its about 100x slower than apache, but fast enough, lol
[7:45] <toad_> for static content there are zillions of possibilities
[7:45] <toad_> some of which can do 10k simultaneous clients at speed
[7:46] <toad_> hackeron: why is it so slow? surely stuff gets cached?
[7:46] <toad_> obviously it has to construct each page from dozens of database queries - but popular pages presumably are cached?
[7:46] <hackeron> toad_: well, its 100% python, and they have these "zope objects" instead of using a relational database, or filesystem directories/files.
[7:47] <toad_> ah okay
[7:47] <toad_> in other words it's all in RAM
[7:48] <toad_> hackeron: finished compiling?
[7:48] <hackeron> well, they have several storate types, a single file, directories and 2 other experimental ones
[7:48] <hackeron> toad_: on a C3-2? -- you're kidding right :) -- compile will take *days*, lol
[7:48] <toad_> well so they implement their own mysql clone?
[7:48] <toad_> what are you compiling? X? Mozilla? OpenOffice?
[7:48] <hackeron> I'll just background it, kill mlnet and start freenet, -- actually, I'll update my freenet first and change firewall settings back.
[7:49] <toad_> check the -Xmx
[7:49] <hackeron> toad_: well, wouldnt call it a clone, its not really relational :)
[7:49] <hackeron> toad_: whoza whatsit?
[7:49] <toad_> hackeron: you said you'd given it 16M of RAM
[7:49] <toad_> hackeron: you using start-freenet.sh, right?
[7:50] <hackeron> toad_: I'm using gentoo's init script
[7:50] <toad_> edit it, what does the line say that has -Xmx in it?
[7:50] <toad_> hmmm
[7:50] <toad_> well did you play with the config?
[7:50] <toad_> or the gentoo files even?
[7:50] <hackeron> a little bit, it was a while ago
[7:50] <toad_> you said you'd changed the memory allocation...
[7:50] <hackeron> freenet-0.5.2.1-r8 < new enough?
[7:51] <hackeron> actually, screw the freenet on gentoo. I'm going to install it manually.
[7:52] <toad_> no
[7:52] <toad_> 0.5.2.1 is ancient
[7:52] <hackeron> its the latest on portage, lol
[7:52] <toad_> if it is actually 5102, it's new enough
[7:52] <hackeron> gentoo is *horribly* outdated
[7:52] <toad_> well, -r8 might actually be 5102
[7:52] <toad_> hackeron: really? i thought gentoo was the geeky distro that has everything up to date?
[7:53] <hackeron> toad_: nah, thats LFS :)
[7:53] <toad_> as opposed to debian, the relatively user-unfriendly OS with the nice package system
[7:53] <toad_> (but not as unfriendly as gentoo)
[7:53] <toad_> (and rapidly getting friendlier)
[7:54] * toad_ installed debian the other day on a friend's PC.. it took something like 10 minutes of actual interaction, with a long package download... I only had to update the kernel and had 3D working
[7:54] <hackeron> well, I'm currently on arklinux. Its very friendly, and uses apt4rpm with very responsive devs. Got them to put unace on the repository by just saying, what, no unace? on their irc channel :)
[7:54] <toad_> was pretty amazing really.. the new installer is NICE
[7:54] <hackeron> ah, debian made a new release?
[7:55] <toad_> well, anything with APT is obviously nice
[7:55] <toad_> hackeron: they're still working on sarge
[7:55] <toad_> hackeron: should be out soonish
[7:55] <toad_> hackeron: very nice installer, even if it is text mode
[7:55] <hackeron> yeah, apt is ok. Has some major drawbacks, but best package system out there for opensource stuff :)
[7:55] <toad_> what are the major drawbacks?
[7:56] <hackeron> it requires a deb for one
[7:56] <hackeron> (or an rpm on the hacked version)
[7:56] <toad_> (btw it is possible to get debian to rebuild a package and all its dependancies locally...)
[7:56] <hackeron> what if you cant freely distribute, but can freely download a piece of software?
[7:56] <hackeron> toad_: yeah, apt-build rules :)
[7:57] <hackeron> makes gentoo sorta redundant as it has all the latest stuff anyway
[7:57] <toad_> hackeron: well, it also requires a repository
[7:57] <toad_> gentoo is more up to date - debian is still on a fork of X4.3.0, for example
[7:57] <hackeron> toad_: yeah, and it has to refetch the *entire* repository
[7:57] <hackeron> toad_: say 1 package changes? -- refetch almost 4Mb anyway :)
[7:57] <toad_> hackeron: it does?
[7:57] <toad_> well you have to fetch the Packages.gz yes
[7:58] <hobx_> Tried Ubuntu?
[7:58] <hackeron> toad_: well, I'm designing a packaging system that will use debs, but uses a relational database with commulative changelogs you can just rsync with --partial.
[7:58] <toad_> anything that isn't freely distributible will have major problems..
[7:58] <toad_> hobx_: i tried the other one... what's it called..? MEPIS
[7:58] <hobx_> The name is stupid, but Shuttleworth seems to have the right idea.
[7:58] * sleon|tuX is now known as sleon|away
[7:58] <hackeron> toad_: it will also support another package type to be able to install software not freely distributable.
[7:58] <hobx_> My brother has it on his laptop, and it looks alright.
[7:58] <toad_> yeah there are some nice debian-based ones for end users
[7:59] <hackeron> hobx_: try arklinux
[7:59] <hackeron> hobx_: and I hated ubuntu, its very unpolished.
[7:59] <toad_> hackeron: DEBs can download stuff automatically if you're allowed to
[7:59] <hobx_> I can't be bothered to reinstall anything
[7:59] <hobx_> I haven't done an OS install for like five years.
[7:59] * toad_ is happy with debian, having unfucked it by removing the stupid cronjob that was messing a lot of stuff up
[8:00] <toad_> of course i compile my own kernels
[8:00] <toad_> stupid cronjob was written by me of course
[8:00] <toad_> it called /etc/init.d/umountnfs.sh
[8:00] <hobx_> The problem with sid is that is very periodic.
[8:00] <toad_> that was a bad idea
[8:00] <hobx_> Recently it has been very crashy for me, especially Firefox.
[8:00] <toad_> firefox crashes sometimes yeah.. the kernel crashes more for me though
[8:00] <hackeron> toad_: oh, and I'll also have forced srcurl for the watch system :) -- debian looks quite bad here: http://dehs.alioth.debian.org/no_updated.html
[8:01] <hackeron> and this: http://dehs.alioth.debian.org/no_watch.html
[8:02] <hackeron> toad_: really? -- I found firefox just plain unstable. So is konqueror 3.4 I suppose. I'm back to konqueror 3.3-alpha1 -- most stable konqueror ever :)
[8:02] <toad_> somebody told me debian was only going to support 4 architectures in future to speed releases up.. is this true?
[8:02] <hackeron> toad_: probably not, lol
[8:02] <toad_> hackeron: firefox is .. a little unstable
[8:02] <toad_> it's not unusably so for me
[8:02] <hobx_> Some of candidates suggested it
[8:02] <hobx_> If you care you should vote.
[8:02] <toad_> where?
[8:03] <hobx_> I don't know. Debian Community election.
[8:03] <hackeron> toad_: well, konqueror-3.3-alpha1 hasnt crashed for me in weeks, wonder why they fucked up up for the final release...
[8:03] <hackeron> hobx_: sine when does debian listen to the community?
[8:04] <hobx_> hackeron: I didn't say they did. I said they elect the project leader.
[8:04] <hackeron> hobx_: remember gshield vs shorewall? -- why was it removed? -- one of the senior debian devs said he doesnt like the guy that develops/maintains gshield, lol
[8:04] <hobx_> Yeah, that is always they whole story...
[8:05] <hackeron> incidentally, that was round about the time some devs left to develop gentoo :)
[8:05] <hackeron> sure gentoo sucks, but atleast they are not fascist
[8:05] <hackeron> (well, except their gnome devs, but who cares about gnome anyway)
[8:05] <toad_> hobx: hmm.. any exit polls or similar things?
[8:05] <hobx_> get over it
[8:05] <toad_> who is is now?
[8:05] <hackeron> toad_: should I use freenet-latest.tgz?
[8:06] <hobx_> toad_: I don't know. I don't really care, I just read about it.
[8:06] <hackeron> hobx_: you use ubuntu?
[8:06] <toad_> hackeron: if you don't like it, maintain the package yourself
[8:06] <hobx_> no, I use Debian sid.
[8:06] <hackeron> toad_: dont like what?
[8:06] <toad_> hackeron: yes
[8:07] <toad_> hackeron: the dev who said he didn't like the guy was the maintainer for the package?
[8:07] <hackeron> toad_: nah, he didnt like the actual gshield developer and just removed the package from the repository despite many people offering to maintain it.
[8:08] <toad_> hmmm
[8:08] <hackeron> toad_: it was just "I dont like this guy, so screw him.. What? you like the package? -- fuck you!"
[8:08] <toad_> well like oskar said there may have been more to it than that
[8:09] <hackeron> toad_: I've read quite a few theads about it at the time, there may have been, but from what I could tell, the debian dev was just pushing shorewall even though its far more complicated"
[8:10] <toad_> hmmm
[8:10] <hobx_> Either way, people are allowed to do whatever they want.
[8:10] <hackeron> toad_: there was no technical reason. The package had an active developer, the package is still widely used, there were maintainers for it on debian and many people that wanted it back on the repository. It was just removed, no questions asked.
[8:10] <hobx_> If there is a problem one can fork.
[8:10] <toad_> that does not align with my general experience which is that debian will package even other distributions
[8:10] <hackeron> hobx_: yeah, thats why debian has so many forks :)
[8:11] <hobx_> yeah, and they all suck.
[8:11] <hobx_> So maybe the problem wasn't with Debian...
[8:11] <toad_> well i don't really know how the power structure works
[8:11] <hackeron> the forks suck, but they were made because debian is not community driven.
[8:11] <toad_> but obviously people with authority need to be accountable - hence the election
[8:12] <toad_> how is debian not community driven?
[8:12] <hobx_> Eh.
[8:12] <hobx_> The election is just stupid. The people in charge are accountable - their only asset is the loyalty of their users and developers.
[8:13] <toad_> hobx_: isn't it better to resolve these things within the structure?
[8:13] <toad_> hobx_: moving everyone over to another fork is infeasible
[8:13] <toad_> it could happen but it'd be an enormous disruption
[8:14] <toad_> anyway i should remember never to discuss politics with hobx_ :)
[8:14] <hobx_> But normally the threat is enough.
[8:14] <hackeron> PS, have any of you compared apt-build to redhat spec files and srpms? -- the deb package format is not even that impressive :)
[8:14] <hackeron> apt is, but it works all the same with rpms.
[8:15] <toad_> However, it is
[8:15] <toad_> equally unacceptable for team members to be abused by other
[8:15] <toad_> developers. Developers who are unable to communicate in a reasonable
[8:15] <toad_> way should not expect to obtain useful feedback, and this should be
[8:15] <toad_> recognised by the community.
[8:15] <toad_> heh
[8:15] <hobx_> yeah, well fuck off!
[8:15] <hobx_> amphibian slime!
[8:15] <toad_> :)
[8:15] <hackeron> toad_: that sounds so nice, doesnt it :)
[8:16] <toad_> hackeron: from the first election candidate's platform
[8:16] * toad_ tends to agree with him that debian's problems are social rather than technical
[8:16] <hobx_> see populist crap. That is what democracy gives you
[8:16] <toad_> but i don't really have any grounds to judge that apart from occasional things i've heard
[8:17] <hobx_> I think stepping down the number of platforms would be pretty smart.
[8:17] <toad_> hobx_: obviously a benign dictator would be better.. as long as he was a perfect person
[8:17] <hobx_> Three is enough.
[8:17] <toad_> hobx_: three? IA64, x86-64/EMT64, and x86
[8:17] <toad_> all distros support those
[8:17] <hobx_> toad_: No, the idea of benign despot is stupid. What is best is people without power because others are free to do as they wish.
[8:18] <toad_> hobx_: that's the bottom line, and it's a welcome safeguard. that doesn't mean it's enough though.
[8:18] <hobx_> - IA64 + Mac
[8:18] <toad_> debian is a massive community
[8:18] <toad_> hobx_: isn't one of debian's key distinctives that it runs on everything?
[8:18] <toad_> that certainly is a big part of the reason why the installer took so much work
[8:18] <hobx_> Is that why you use it?
[8:19] <toad_> hobx_: what are the current release platforms btw?
[8:19] <hobx_> They are wasting their time on like 0.0001% of the users.
[8:19] <hobx_> toad_: Too many.
[8:19] <toad_> hobx_: i use it because it's good enough, it's free in every sense, and I don't need to reinstall :)
[8:19] <hobx_> yeah, me too.
[8:20] <hobx_> So that is what Debian ought to work on, not supporting a hundred minority platforms and thus never being able to release.
[8:20] <toad_> ftp://debian.blueyonder.co.uk/pub/debian-iso/ only lists 3 platforms
[8:20] <toad_> i386, mips and sparc
[8:20] <hackeron> toad_: did distupgrade work for you? -- I tried that a while back and it updated glibc without updating packages compiled against an old glibc - naturally, system never booted up again :)
[8:21] <toad_> ahhh... http://ftp.ticklers.org/debian-cd/images/3.0_r4/
[8:21] <toad_> hackeron: it almost always works fine
[8:21] <toad_> for me
[8:21] <toad_> occasionally a package breaks, but I _am_ running sid
[8:21] <toad_> that sounds like a bug in apt to me.. it might be some sort of cock-up by the maintainers though
[8:21] <hackeron> toad_: yes, *almost* always :) -- not something you can truly rely on. I suppose thats a problem with all distributions including source distributions though.
[8:22] <hobx_> well, stable always works...
[8:22] <toad_> S390 is in the official releases? woah
[8:22] <hackeron> yes, but I dont find stable very usable really.
[8:22] <toad_> what actually uses S390? mainframes?
[8:22] <hobx_> hackeron: But nobody is telling you you have to use it.
[8:22] <toad_> stable is always too far behind... at least it used to be. just before a release stable is pretty good. :)
[8:23] <hobx_> Use whatever you want, but don't whine about something you are getting for free.
[8:23] <hobx_> I use stable on my server.
[8:23] <toad_> hobx_: "I whine because I care"
[8:24] <hackeron> hobx_: no, I'm not complaining here, I'm saying its a problem with all distributions in general, just saying I occationally had to re-install debian when I used it.
[8:25] <hackeron> surely not nearly as much as I had to re-install windows ;)
[8:25] <toad_> err just after a release
[8:25] <toad_> just before a release, testing is pretty good
[8:25] <hackeron> toad_: when was last release? 3 years ago?
[8:25] <toad_> hackeron: woody wasn't that long ago
[8:25] <hackeron> wasnt it?
[8:25] <hobx_> yes
[8:25] <hobx_> at least three years
[8:26] <toad_> <tt>[19 Jul 2002]</tt> <strong><a href="http://www.debian.org/News/2002/20020719">Debian GNU/Linux 3.0 released</a></strong>
[8:26] * toad_ wonders why firefox includes the tags when i copy...
[8:26] <hackeron> toad_: yeah, 3 years ago :)
[8:26] <toad_> okay, a long time ago :)
[8:26] <hobx_> hackeron: I have never had to reinstall debian.
[8:27] <hobx_> The only reason I can come up with is to get off sid and join testing instead. Downgrading is difficult.
[8:27] <hackeron> well, I donno, perhaps I did something wrong, but a distupgrade updated glibc to an new major version, naturally breaking all packages on system.
[8:28] <hobx_> I distupgraded through that glibc update as well. It updated all my packages.
[8:28] <hobx_> So I don't know why it broke for you.
[8:28] * greycat (rfc1413@wooledge.org) has joined #freenet
[8:28] <toad_> The release team has
[8:28] <toad_> been hampered by communication issues between teams, resulting in
[8:28] <toad_> release-blocking issues not becoming apparent until they are too late
[8:28] <toad_> to avoid
[8:28] <toad_> is this true?
[8:29] <toad_> hackeron: I successfully updated between dists and glibc versions many times, including once from slink to potato
[8:29] <toad_> AAAAGES ago
[8:29] <hackeron> me neither. Still, upgrading every single package on system pretty much equals a re-install :) -- just mount /home on a separate partition and you can distupgrade any distribution :)
[8:29] <hackeron> toad_: major or minor versions?
[8:29] <toad_> hackeron: hmmm
[8:30] <toad_> woody was 3.0, right?
[8:30] <toad_> i'm sure i went from potato (2.1?) to woody
[8:30] <greycat> yes
[8:30] <greycat> potato is 2.2; slink is 2.1
[8:30] <toad_> well what was just before woody? potato?
[8:30] <toad_> so major versions yes
[8:30] <toad_> was probably on testing of course
[8:30] <toad_> so from potato to pre-woody
[8:31] <hackeron> well, maybe update got interrupted for some reason, I donno. I didnt know much about updating package back then :)
[8:31] <hackeron> packages*
[8:32] <hackeron> hmm, latest freenet doesnt like me: "grep: error while loading shared libraries: libc.so.6: cannot open shared object file: No such file or directory"
[8:32] <greycat> uh... I'd say it's grep that doesn't like you.
[8:32] <toad_> :(
[8:36] <hobx_> hackeron: Do "ls $LDPATH | grep libc.so.6"
[8:37] <greycat> it's rather unlikely that $LDPATH has any value
[8:37] <hobx_> whatever
[8:37] <hackeron> server freenet # ls $LDPATH
[8:37] <hackeron> README freenet.conf freenet.pid seednodes.ref start-freenet.sh update.sh
[8:37] <hackeron> freenet-ext.jar freenet.jar preconfig.sh seednodes.ref.bz2 stop-freenet.sh
[8:37] <hackeron> hmm
[8:37] <hackeron> start-freenet.sh overrides $LDPATH ?
[8:37] <greycat> hackeron: it's just like "ls", because $LDPATH is nothing
[8:37] * hobx_ think you nerds ruined my joke
[8:37] <greycat> there is no $LDPATH.
[8:37] <hackeron> ah, ok
[8:38] <hobx_> you can't do ls on a path statement anyways if we are going to be super anal today.
[8:39] <toad_> hobx_: you're implying you're not a nerd?!
[8:39] <toad_> okay, geek/nerd dichotomy maybe
[8:40] <hackeron> toad_: http://deadbeefbabe.org/paste/229?_nevow_carryover_=1112705059.53127.0.0.10.965504472956
[8:41] <hackeron> err, I mean http://deadbeefbabe.org/paste/229/0
[8:41] <toad_> hackeron: your system is fscked
[8:41] <toad_> sed, grep, head and uname aren't working due to library issues
[8:41] <hackeron> sure they are:
[8:41] <toad_> i suggest you continue the emerge
[8:41] <hackeron> server freenet # grep grep *
[8:41] <hackeron> start-freenet.sh:#if java -help 2>&1 | grep "[-]server" >/dev/null ;
[8:41] <hackeron> start-freenet.sh: if echo $SUN_VERSION | grep "^1.[0-3]" ; then
[8:41] <hackeron> start-freenet.sh: if echo $SUN_VERSION | grep "^1.4.0" ; then
[8:41] <hackeron> start-freenet.sh: if echo $SUN_VERSION | grep "^1.4.1" ; then
[8:42] <toad_> okay in that case it could be the LD_whatever in the start-freenet.sh
[8:42] <toad_> get rid of it
[8:42] <hackeron> start-freenet.sh: LD_ASSUME_KERNEL=2.2.5
[8:42] <hackeron> start-freenet.sh: export LD_ASSUME_KERNEL
[8:42] <hackeron> start-freenet.sh: LD_ASSUME_KERNEL=2.4.1
[8:42] <hackeron> start-freenet.sh: export LD_ASSUME_KERNEL
[8:42] <hackeron> start-freenet.sh: LD_ASSUME_KERNEL=2.2.5
[8:42] <hackeron> start-freenet.sh: export LD_ASSUME_KERNEL
[8:42] <hackeron> which one?
[8:43] <hackeron> seems it sets LD_ASSUME_KERNEL=2.2.5
[8:43] <greycat> run it with "sh -x start-freenet.sh" and see what it's doing
[8:44] <hobx_> I am not a nerd. I'm just disproportionately intelligent.
[8:44] <hackeron> + LD_ASSUME_KERNEL=2.4.1
[8:44] <hackeron> + export LD_ASSUME_KERNEL
[8:44] <hackeron> ++ java -version
[8:44] <hackeron> ++ head -n 1
[8:44] <hackeron> head: error while loading shared libraries: libc.so.6: cannot open shared object file: No such file or directory
[8:45] <hackeron> hmmm
[8:45] <greycat> does it work if you comment out the LD_ASSUME_KERNEL lines?
[8:45] <hackeron> seems to be
[8:45] <hobx_> What on earth does freenet-start have to know the kernel version for?
[8:46] <hackeron> yup, working fine
[8:46] <toad_> to tell the JVM not to crash :(
[8:46] <hackeron> toad_: well, it has a bad side effect here :)
[8:46] <hobx_> java rules!
[8:47] <hackeron> hobx_: indeed!
[8:47] <hobx_> I wonder if hating the only language I am fluent in reflects a deeper self-loathing...
[8:48] <hackeron> lol
[8:48] * hobx_ codes pretty good ruby as well, though...
[8:48] <hackeron> hobx_: learn python, you'll love it
[8:48] <hobx_> no, I won't
[8:48] <hobx_> python is crap
[8:48] <hackeron> learn it, you'll love it!
[8:49] <hobx_> I have learned it, it was crap.
[8:49] <hackeron> what did you write in it?
[8:49] <hobx_> The tutorial examples.
[8:50] <hackeron> heh :)
[8:50] <hobx_> Do you know ruby?
[8:50] <hackeron> no, I dont
[8:50] <hackeron> ruby is crap
[8:50] <hobx_> Ruby is better than python in every way.
[8:51] <hackeron> well, I read some ruby book a while ago, how is it better than python exactly? -- I dont understand the point of forcing methods. Other than that, the code is just ugly.
[8:52] <hobx_> Better OO, better support for lambda, continuations, anonymous code blocks, language extensions, etc etc
[8:53] <hobx_> I don't understand why people like Python. It is the vanilla of languages.
[8:53] <hackeron> hobx_: lambda has been obseleted in python in favor of nested functions
[8:53] <hackeron> hobx_: lambda is ugly
[8:53] <hobx_> sure. Whatever you say.
[8:54] <hackeron> hobx_: "language extensions"? "anonymous code blocks"? -- wtf
[8:54] <hackeron> hobx_: language extensions = modules
[8:54] <hackeron> hobx_: anonymous code blocks? -- just dont specify global
[8:54] <hobx_> no
[8:55] <hobx_> I think you should have read the book harder.
[8:55] <hackeron> hobx_: http://www.secnetix.de/~olli/Python/lambda_functions.hawk
[8:55] <hobx_> But honestly I hardly care to argue.
[8:55] <hobx_> If Python works for you, use it.
[8:55] <hackeron> hobx_: read that about anonymous functions and lambas in python
[8:55] <hobx_> But I assure you I don't like it. I haven't seen a single appealing feature in it.
[8:58] <hackeron> Python just makes makes you think differently, like 99% of the time you dont need regex where you would use it in perl or ruby :)
[9:00] * FallingBuzzard (~srademach@207.152.112.129) has joined #freenet
[9:05] * Morten^Test (~Miranda@83.73.56.94.ip.tele2adsl.dk) has joined #freenet
[9:06] * Morten^Test (~Miranda@83.73.56.94.ip.tele2adsl.dk) has left #freenet
[9:11] <hackeron> Hmm, when I click on Spread Freenet, I just see "Error", is this normal?
[9:14] * NullAcht15 (~NullAcht1@dsl-213-023-039-193.arcor-ip.net) has joined #freenet
[9:15] <hackeron> hmm: 'Pooled threads running jobs: 57 (47.5%) [Rejecting incoming connections and requests!]' < why?
[9:16] <hobx_> I don't do the kind of coding where you use regex, so I wouldn't know.
[9:18] <hackeron> any ideas about my freenet problem?
[9:24] <toad_> hackeron: which problem?
[9:24] <Elly> hey, toad_
[9:24] <hackeron> toad_: when I click on "Spread Freenet" - it just says "Error"
[9:24] <Elly> I reverse-engineered EzBoard's community chat system and implemented a protocol handler in Java
[9:26] <toad_> Elly: congratulations
[9:26] <toad_> hackeron: you are accessing it from another machine, right?
[9:26] <hackeron> toad_: yup
[9:26] <hackeron> toad_: is that bad?
[9:26] <toad_> you have to tell it to let you control the dist servlet from outside.. there's an option
[9:27] <toad_> i think it's distribution.generatorAllowedHosts
[9:27] <toad_> find it in the config file
[9:27] <hackeron> the config file is almost empty
[9:27] <hackeron> Thats all I have there:
[9:27] <toad_> ah
[9:27] <hackeron> mainport.bindAddress=*
[9:27] <hackeron> mainport.allowedHosts=127.0.0.1,192.168.0.0/4
[9:27] <hackeron> listenPort=25062
[9:27] <hackeron> seedNodes=seednodes.ref
[9:27] <hackeron> ipAddress=81.86.159.146
[9:27] <toad_> hmmm
[9:27] <toad_> you didn't chmod a+x *.sh
[9:28] <toad_> you only chmod'd +x start-freenet.sh
[9:28] <toad_> right?
[9:28] <toad_> no, that wouldn't do it...
[9:28] <hackeron> start-freenet.sh was already +x
[9:28] <toad_> it's supposed to call --config the first time
[9:28] <toad_> but it didn't because it crashed
[9:28] <toad_> shut down the node and do it now
[9:28] <hackeron> ok, doing
[9:28] <toad_> one way is stop-freenet; rm freenet.conf; start-freenet
[9:28] <toad_> that should work
[9:30] <hackeron> ah, it is, its asking me questions now :)
[9:31] <hackeron> is inputBandwidthLimit in KBps?
[9:31] <toad_> bytes/sec
[9:32] <hackeron> averageInputBandwidthLimit < what exactly is that?
[9:33] <toad_> if you want it to on average transfer less than X bytes/sec e.g. if you have a monthly cap
[9:33] <toad_> don't think that actually works though
[9:33] <toad_> not sure
[9:33] <hackeron> heh, ok :)
[9:34] <hackeron> when freenet starts, it shows:
[9:34] <hackeron> INFO: Native CPUID library 'freenet/support/CPUInformation/libjcpuid-x86-linux.so' loaded from resource
[9:34] <hackeron> INFO: Non-optimized native BigInteger library 'net/i2p/util/libjbigi-linux-none.so' loaded from resource
[9:34] <hackeron> is that anything to worry about?
[9:34] * Kiske (~ksno@217.170.33.158) has joined #freenet
[9:34] * Kiske (~ksno@217.170.33.158) has left #freenet
[9:44] <sanity_> toad: what is the equation to create the 1/d network again?
[9:45] <hobx_> int u = r.nextDouble();
[9:45] <hobx_> int d = (int) Math.pow(n, u);
[9:46] <hobx_> Though really you should be able to calculate that...
[9:47] <sanity_> i don't do math
[9:47] <sanity_> n is the number of nodes?
[9:49] <hobx_> aye
[9:50] <hobx_> that is an approximation, of course.
[9:50] <hobx_> But it works well enough
[9:51] <hobx_> Note that if you have edges in both directions you need to randomize that, and then you can use n / 2 instead of n.
[9:51] <hackeron> would you say its safe to try the unstable version of freenet, or are there major problems with it?
[9:51] <hobx_> It has been known to forward your name to the FBI...
[9:52] <hackeron> so its not anonymous?
[9:53] <nextgens> hobx_: not sure it keeps on doing it ;-)
[9:53] <sanity_> nah, edges are unidirectional
[9:54] <sanity_> hobx: so: nodes[x].addNeighbour(nodes[(int) Math.pow(nodes.length, r.nextDouble())]);
[9:54] <sanity_> oh, wait
[9:54] <sanity_> needs to be relative
[9:55] <sanity_> nodes[x].addNeighbour(nodes[x + ((int) Math.pow(nodes.length, r.nextDouble())) % nodes.length]); ?
[9:58] <hobx_> yeah
[9:58] <hobx_> that is right.
[9:59] * hobx_ has implemented this a couple of times now :-)
[10:08] * Ash-Fox (UNKNOWN@adg87.neoplus.adsl.tpnet.pl) Quit (Nick collision from services.)
[10:08] * Ash-Fox (UNKNOWN@adg87.neoplus.adsl.tpnet.pl) has joined #FreeNET
[10:24] * Hory (~Miranda@82.76.81.56) Quit (Read error: 110 (Connection timed out))
[10:47] <sanity_> hobx: bugger. any idea how to get an average in a circular space?
[10:47] <sanity_> hobx: ie. finding a point such that the sum of the distances between it and sampled points is minimised?
[11:15] * NullAcht15 (~NullAcht1@dsl-213-023-039-193.arcor-ip.net) Quit (Read error: 110 (Connection timed out))
[11:17] * NullAcht15 (~NullAcht1@dsl-082-082-144-172.arcor-ip.net) has joined #freenet
[11:20] <hackeron> how much should the outbound message overhead be? -- is 62% much?
[11:28] * verl (verl@h155n2fls33o877.telia.com) has joined #freenet
[11:37] * pittaman (~mistery_b@d54C32DBA.access.telenet.be) has joined #freenet
[12:08] <hackeron> is there a way to see a list of most popular SSK requests from my node? (to give me ideas what to browse)
[12:17] * cbreak (~cbreak@84-72-98-199.dclient.hispeed.ch) has joined #freenet
[12:28] * gggggg (gggggg@dyn134.kif8.nas.panafonet.gr) has joined #freenet
[12:28] * oierw` (mathew@cpe-66-91-92-246.hawaii.res.rr.com) Quit (Read error: 54 (Connection reset by peer))
[12:29] * oierw` (mathew@cpe-66-91-92-246.hawaii.res.rr.com) has joined #freenet
[12:31] * gggggg (gggggg@dyn134.kif8.nas.panafonet.gr) has left #freenet
[12:43] * nextgens nnt
[13:04] * cbreak is now known as cbreak|WoW
[13:20] * RallosZek (1000@d14-69-248-202.try.wideopenwest.com) Quit ("Leaving")
[13:27] * oierw` is now known as oierw
[13:29] <oierw> hackeron: no
[13:31] * cbreak|WoW is now known as cbreak
[13:35] * Hory (~Miranda@82.76.81.56) has joined #FreeNet
[13:35] <hackeron> oierw: ok, thanks. Suppose thats a good thing.
[13:36] * FallingBuzzard (~srademach@207.152.112.129) has left #freenet
[14:01] * sleon|away (test@p54A14AF9.dip.t-dialin.net) Quit (Read error: 60 (Operation timed out))
[14:09] * sleon|away (~sleon|tuX@p54A135DF.dip.t-dialin.net) has joined #freenet
[15:29] * NullAcht15 (~NullAcht1@dsl-082-082-144-172.arcor-ip.net) Quit (Read error: 104 (Connection reset by peer))
[15:32] * NullAcht15 (~NullAcht1@dsl-082-082-142-046.arcor-ip.net) has joined #freenet
[15:58] <hackeron> vote for freenet: http://www.planetpeer.de/index.php?option=com_poll&task=results&id=4
[16:50] * jaguar (jepel@frank.psimonkey.org.uk) Quit (Read error: 54 (Connection reset by peer))
[17:26] * NullAcht15 (~NullAcht1@dsl-082-082-142-046.arcor-ip.net) Quit (Read error: 110 (Connection timed out))
[17:26] * NullAcht15 (~NullAcht1@dsl-082-082-140-184.arcor-ip.net) has joined #freenet
[17:32] * oierw` (mathew@cpe-66-91-92-246.hawaii.res.rr.com) has joined #freenet
[17:32] * oierw (mathew@cpe-66-91-92-246.hawaii.res.rr.com) Quit (Read error: 54 (Connection reset by peer))
[17:39] * bluephile (bluephile@oh-wstnrsrv-cuda1d-5-2.clvdoh.adelphia.net) has joined #freenet
[17:53] * ca (~ca@83.72.201.52.ip.tele2adsl.dk) has joined #freenet
[17:57] * greycat (rfc1413@wooledge.org) Quit ("This time the bullet cold rocked ya / A yellow ribbon instead of a swastika")
[18:06] * pittaman (~mistery_b@d54C32DBA.access.telenet.be) Quit ("YOU FUCK.")
[18:29] * Hory (~Miranda@82.76.81.56) Quit ("CyberLore.net - Recommendations on the best games, freeware and websites.")
[19:08] * NullAcht15 (~NullAcht1@dsl-082-082-140-184.arcor-ip.net) Quit (Read error: 110 (Connection timed out))
[19:08] * NullAcht15 (~NullAcht1@dsl-082-082-142-128.arcor-ip.net) has joined #freenet
[19:12] * toad_ (toad@toad-with-underline.active.supporter.pdpc) Quit (Remote closed the connection)
[20:14] * cbreak (~cbreak@84-72-98-199.dclient.hispeed.ch) Quit ("leaving")
[20:15] * toad_ (toad@toad-with-underline.active.supporter.pdpc) has joined #freenet
[20:15] * ChanServ sets mode +o toad_
[20:20] * ca (~ca@83.72.201.52.ip.tele2adsl.dk) Quit ("Leaving")
[20:31] * sleon|away (~sleon|tuX@p54A135DF.dip.t-dialin.net) Quit (Remote closed the connection)
[21:45] * Ash-Fox (UNKNOWN@adg87.neoplus.adsl.tpnet.pl) Quit (Nick collision from services.)
[21:45] * Ash-Fox (UNKNOWN@adg87.neoplus.adsl.tpnet.pl) has joined #FreeNET
[21:50] * NullAcht15 (~NullAcht1@dsl-082-082-142-128.arcor-ip.net) Quit (Remote closed the connection)
[22:33] * Ash-Fox (UNKNOWN@adg87.neoplus.adsl.tpnet.pl) Quit (Nick collision from services.)
[22:33] * Ash-Fox (UNKNOWN@ade91.neoplus.adsl.tpnet.pl) has joined #FreeNET
These logs were automatically created by Jay Oliveri with his gimp hapi on irc.freenode.net.