Timestamps are in GMT/BST.
[14:18] * hapi (~PircBot@newton.ncc.edu) has joined #freenet
[14:18] * Topic is 'Upgrade to 5096 (5095 is mandatory) | Unstable: Upgrade to 60241 (60234 is mandatory) | Channel logs: http://newton.matcmp.ncc.edu/~lockej/freenet/chanlog/ | #freenet-politics and #freenet-chat are available for offtopic discussions'
[14:18] * Set by toad_ on Fri Sep 24 14:57:57 EDT 2004
[14:27] <toad_> 100x25:10 -> ~ 0.82
[14:27] * cbreak (~cbreak@dhcp-32-094.via-eth.ch) Quit ("leaving")
[14:27] <toad_> 50x25:5 -> ~~~ 0.555 (may go lower)
[14:28] <toad_> 50x25:10 -> ~ 0.965
[14:28] <toad_> 100x25:15 -> ~ 0.98
[14:29] <toad_> 100x25:12 -> ~ 0.92
[14:31] * toad_ is going to try to find equivalent HTLs in terms of performance (stable-ish psuccess) for 50x25, 100x25, and 200x25
[14:33] * KenMan (~dogzilla@pcp403292pcs.mntcrm01.md.comcast.net) has joined #freenet
[14:34] <toad_> hi KenMan !!
[14:35] * hirvox upgrades to 1.5.0-final
[14:38] <toad_> 50x25@10, 100x25@14, and 200x25@20 are all approximately the same line
[14:39] <toad_> roughly the same asymptote line
[14:39] <toad_> at about 0.965
[14:39] <toad_> we need them to run for longer of course...
[14:40] <toad_> but that's VERY promising
[14:41] <toad_> if I can prove the same thing for 400x25@25...
[14:41] <toad_> then we'll be in business...
[14:41] <toad_> we know 400x25@20 sucks...
[14:41] <toad_> as does 400x25@10
[14:41] <toad_> and 400x25RANDOM@20
[14:42] <toad_> but it finally feels like we're on the right track with these simulations...
[14:42] * Ash-Fox (Hal-9000@aay189.neoplus.adsl.tpnet.pl) Quit (Read error: 104 (Connection reset by peer))
[14:43] * Ash-Fox (Hal-9000@aay189.neoplus.adsl.tpnet.pl) has joined #FreeNET
[14:44] <toad_> if we can show this, then we have solid evidence that NGR+LRU scales...
[14:45] <toad_> I would then want to see what difference varying the # conns makes
[14:45] <toad_> it's suspicious that the doubling value is 5, the square root of the number of connections...
[14:45] <hirvox> btw, my node seems to have specialized completely
[14:45] <toad_> hirvox: really? show me
[14:45] <hirvox> what's your ip?
[14:46] <toad_> amphibian.dyndns.org
[14:47] <hirvox> restarting node, you should be able to connect to http://nightwatch.mine.nu:8888 after that
[14:48] <toad_> hirvox: you have to add me by IP
[14:48] <toad_> even if I give you a domain name
[14:48] <toad_> you did, right?
[14:49] <hirvox> wasn't mainport.allowedHosts supposed to accept domain names as well?
[14:51] <toad_> maybe, but it doesn't...
[14:51] <hirvox> the node is up
[14:52] <toad_> hmmm
[14:52] <toad_> 400x25@25 isn't doing as well as hoped...
[14:53] <toad_> but we'll have to see...
[14:53] <toad_> hirvox: connected
[14:54] <toad_> inbound requests histogram appears moderately specialized...
[14:55] <toad_> woah, datastore histogram is specialized...
[14:57] * pittaman (~mistery_b@d51A410CA.kabel.telenet.be) has joined #freenet
[14:59] * i2p_iip (~changate@host.teitel.net) Quit (Remote closed the connection)
[15:01] <toad_> yay, got the ssh working to both PCs...
[15:01] <toad_> both remote dual opterons
[15:10] <toad_> hmmm
[15:10] <toad_> to get it up to 97% success on 400x25 may require HTL 30...
[15:10] * toad_ will also try increasing the number of conns but that will slow down the simulation significantly
[15:12] * toad_ points out that 25 is quite low compared to what we use IRL (200 theoretical, 60-100 in reality)
[15:14] * Zorix (Brandon@fl-198-70-221-44.dyn.sprint-hsd.net) has joined #freenet
[15:15] <toad_> http://amphibian.dyndns.org/simulations2-scaling.png
[15:17] <toad_> hmmm
[15:17] <toad_> probably polynomial for a given number of links...
[15:18] <toad_> but i think that was how it worked with classical routing too...
[15:18] * TLF (~francisco@179.Red-81-40-113.pooles.rima-tde.net) Quit ("Adi?s a todos y a todas, y que les vaya bien. || http://www.vitaliano.esp.cc")
[15:19] * toad_ uploads his new jar to the dual-opteron megaserver, as opposed to the home dual-opteron-workstation d-arkangel let me borrow
[15:21] * toad_ thinks he will run 200x50 and 400x50 on it
[15:21] * sanity (~ian@81-178-94-146.dsl.pipex.com) Quit (Read error: 104 (Connection reset by peer))
[15:21] <toad_> or maybe just 200x50 at 2 different htls?
[15:21] <toad_> 400x50 will be sloooow...
[15:21] * sanity (~ian@81-178-94-146.dsl.pipex.com) has joined #freenet
[15:28] <toad_> http://amphibian.dyndns.org/simulations2-scaling.png
[15:28] <toad_> sanity: hi, are you here?
[15:28] <toad_> hmm no apparently not...
[15:29] <toad_> anyone like my graph? :)
[15:32] * TLF (~francisco@209.Red-81-40-113.pooles.rima-tde.net) has joined #freenet
[15:33] * d-ArkAngel (~Rob@cpc2-midd2-5-0-cust16.midd.cable.ntl.com) has joined #freenet
[15:34] <d-ArkAngel> I'm back, sorry to step out of a conversation like that, I had a meeting to attend
[15:36] * TLF (~francisco@209.Red-81-40-113.pooles.rima-tde.net) Quit (Remote closed the connection)
[15:43] * i2p_iip (~changate@host.teitel.net) has joined #freenet
[15:45] * pittaman (~mistery_b@d51A410CA.kabel.telenet.be) Quit (Client Quit)
[15:45] <toad_> :)
[15:45] <toad_> http://amphibian.dyndns.org/simulations2-scaling.png
[15:46] <i2p_iip> <gott> toooooooooooad ! ! ! ! ! !! !! $*Y$R@)&$)
[15:46] <toad_> 100x25@14 and 50x25@10 are almost exactly the same line
[15:46] <toad_> 200x25@20 is pretty close
[15:46] * Ash-Fox (Hal-9000@aay189.neoplus.adsl.tpnet.pl) Quit (Nick collision from services.)
[15:46] * Ash-Fox (Hal-9000@abh154.neoplus.adsl.tpnet.pl) has joined #FreeNET
[15:46] * pittaman (~mistery_b@d51A410CA.kabel.telenet.be) has joined #freenet
[15:46] <toad_> but you have to go up to 30 HTL for 400x25 to reach approximately the same levels
[15:46] <toad_> I'm trying 400x50...
[15:46] <i2p_iip> <gott> toad, I love you !
[15:46] <i2p_iip> <gott> toad, I love you !
[15:47] <toad_> hello gott :)
[15:47] <toad_> we're not adding a constant with every doubling
[15:47] <toad_> but we're not doubling with it
[15:47] <toad_> so it's sublinear
[15:47] <toad_> but not logarithmic
[15:47] <d-ArkAngel> :-)
[15:48] <toad_> if we don't change the number of connections
[15:48] <toad_> i'm trying 400x50@25...
[15:48] <i2p_iip> <gott> toad, jdk 1.5 is rather intriguing
[15:48] <i2p_iip> <gott> do you use it ?
[15:48] <toad_> it may be the # conns becomes important later on and restores it to logarithmicity
[15:48] <toad_> gott: I use a beta
[15:48] <i2p_iip> <gott> yes
[15:48] <toad_> gott: so far
[15:49] <i2p_iip> <gott> you are currently running 1.5 then
[15:49] <i2p_iip> <gott> I find it much faster than 1.4
[15:49] <d-ArkAngel> you was that 1.5 final is out toad?
[15:49] <toad_> argh
[15:49] <toad_> you can just make out orange on that graph...
[15:50] <toad_> it's going to settle at maybe 0.93
[15:50] <i2p_iip> <gott> 1.5 final is not yet out
[15:50] * lolo-laptop (~lostlogic@68.251.84.226) Quit ("Client exiting")
[15:50] <toad_> otoh small changes in htl make large differences in some of these graphs...
[15:50] * toad_ starts a 400x50@27
[15:51] <d-ArkAngel> yet when I look at the sun website it's there for download as JDK 5.0
[15:52] <d-ArkAngel> not 5.0RC like it was yesturday
[15:52] * lolo-laptop (~lostlogic@68.251.84.226) has joined #freenet
[15:53] <toad_> hi lolo-laptop
[15:53] * hirvox downloaded and installed 1.5 final about an hour ago
[15:54] <toad_> theory:
[15:54] <toad_> normally, every doubling produces an increase in required htl of 6
[15:54] <toad_> for 25 connections
[15:54] <lolo-laptop> hey
[15:54] <lolo-laptop> I'm 2 steps closer to having my node up again
[15:54] <lolo-laptop> I have my DSL back up
[15:54] <toad_> but when HTL is close to # nodes, this is violated
[15:54] <lolo-laptop> I have a new case for my dead computer
[15:55] <toad_> hence 50x25@10 instead of @8
[15:55] <toad_> ?
[15:55] <lolo-laptop> as soon as I have a new PSU, I'll be able to get the node back up and with a 60G DS probably
[15:55] <toad_> so then it's 6 every time
[15:55] <toad_> so 50x25@10, 100x25@14, 200x25@20, 400x25@26 ?
[15:55] <toad_> well, 400x50@26?
[15:55] <toad_> lolo-laptop: cool!
[15:55] <toad_> what are you using now?
[15:56] <toad_> brother just got an opteron, he's putting it together... hopefully i can borrow it for simulations while he's out...
[15:56] <lolo-laptop> nice
[15:56] <toad_> only 1 processor
[15:56] <toad_> and I have access to two DP opterons remotely
[15:56] <lolo-laptop> I'm upgrading to dual Athlon-2100s during all this chaos...
[15:56] <toad_> but every little helps
[15:56] <toad_> nice, i've heard bad things about MP mobos though...
[15:57] <toad_> athlon MP mobos i mean
[15:57] <lolo-laptop> well I've been the using dp mobo with a single proc for a long time now (a year or so) I'm just finally getting dual procs for it
[15:57] <lolo-laptop> Tyan Thunder K7
[15:57] <toad_> nice
[15:57] <toad_> does that need registered ram?
[15:57] <lolo-laptop> yeah
[15:57] <d-ArkAngel> tyans are nice, my dual opteron is a tyan
[15:58] <lolo-laptop> and a frigging AMD-GES power supply
[15:58] <lolo-laptop> that blew up
[15:58] <lolo-laptop> and that I therefore need a $200 replacement for
[15:58] <toad_> d-ArkAngel hard-core is a tyan?
[15:58] <d-ArkAngel> yup
[15:58] <toad_> then both the dual opteron systems i have accounts on are tyans...
[15:58] <toad_> the other one is called tyan
[15:59] <toad_> strangely enough :)
[15:59] <toad_> anyway
[15:59] <lolo-laptop> Yeah, Tyan does good stuff, it is just also expensive stuff
[15:59] <toad_> anyone like my new graph: http://amphibian.dyndns.org/simulations2-scaling.png ?
[15:59] <lolo-laptop> yeah, back to freenet
[15:59] <d-ArkAngel> tyan were the first to market with decent dual oppy boards :-)
[15:59] <lolo-laptop> toad_: that's interesting
[15:59] <toad_> lolo-laptop: I think so
[16:00] <toad_> lolo-laptop: approximately the same performance for 50x25@10, 100x25@14, 200x25@20, and 400x25@30
[16:00] <lolo-laptop> do the lines cut off because you didn't bother simulating beyond that number of requests
[16:00] <toad_> I'm trying to see if I can reduce the 400x25
[16:00] <toad_> lolo-laptop: yes, or because it's still working on it
[16:00] <lolo-laptop> gotcha
[16:01] <d-ArkAngel> I think the relation is something like this...
[16:01] <toad_> or in some cases on other graphs because machines were rebooted, or kicked to run more interesting sums
[16:01] <toad_> sims
[16:01] <toad_> i'm trying 400x50@27
[16:01] <toad_> that may do significantly better than 400x50@25 and 400x25@25
[16:02] <toad_> you see the notation i assume? nodes x rtSize @ HTL
[16:02] <toad_> d-ArkAngel: hrrm?
[16:02] <d-ArkAngel> just trying to think before I speak :-)
[16:02] <toad_> drum roll...
[16:03] <d-ArkAngel> :-P
[16:03] * dystopia_ (admin@gaia-wan0.ipv6.freeshell.bofx.net) has joined #freenet
[16:03] * dystopia (admin@gaia-wan0.ipv6.freeshell.bofx.net) Quit (Read error: 104 (Connection reset by peer))
[16:03] * dystopia_ is now known as dystopia
[16:04] <toad_> oooh
[16:04] <toad_> dark green is just starting to show itself
[16:04] <d-ArkAngel> I think the important number is where it's leveling out
[16:04] <toad_> it's in almost the right place...
[16:04] * thelema|away (~thelema@adsl-65-65-202-4.dsl.stlsmo.swbell.net) has joined #freenet
[16:04] <toad_> hi thelema|away !
[16:04] <toad_> you're a long way away if I recall correctly?
[16:05] <toad_> maybe he left his computer running? :)
[16:05] <d-ArkAngel> so I wonder if we could do something like ballon plot of connections vs nodes with psuccess as baloon size
[16:05] <toad_> dark green could settle anywhere between 0.97 and 0.955...
[16:06] <toad_> you mean nodes versus htl?
[16:06] <d-ArkAngel> hm...
[16:06] <i2p_iip> <gott> you are much more attractive now than previously, toad
[16:07] <toad_> if dark green settles anywhere near red and blue, then we've got it - it's very nearly logarithmic
[16:07] <toad_> we'll have to run an 800x100@34 overnight, but it'd be pretty promising
[16:08] <toad_> that's if it's even possible to run 800x100 - it might use a ton of ram...
[16:08] * TLF (~francisco@222.Red-81-40-116.pooles.rima-tde.net) has joined #freenet
[16:08] <lolo-laptop> I think a problem that is showing up is that the 'good HTL threshold' seems to change too much with network size... Idealy over some threshold network size/HTL number the size should be able to keep increasing without the HTL needing to grow, eh?
[16:08] <toad_> 400x50 uses 143MB RES
[16:08] <toad_> so 800x100 would use ~ 550MB RES
[16:09] <toad_> lolo-laptop: well ideally but i don't think that is practical
[16:09] <toad_> what we have is O(log N) at best
[16:09] <toad_> not O(1)
[16:09] <toad_> O(1) is impossible
[16:09] <toad_> unless you centralize
[16:09] <toad_> and have other bits be worse than O(log N)
[16:10] <toad_> if it's double->6HTL, then for a 100K node network, the HTL would still have to be rather high...
[16:10] <lolo-laptop> nod
[16:11] <toad_> if we say 20 for 200, 100K is 5,000 times bigger...
[16:11] <toad_> 5000 is approximately 4096, which is 2^9...
[16:11] <toad_> so we'd need... ouch... 20+6*9 maxHTL
[16:11] <toad_> maxHTL would need to be at least 74
[16:12] <toad_> and it looks like it's not quite log, it looks like the increment increases slowly with each doubling
[16:12] <toad_> so it could need to be a lot more than that...
[16:12] <toad_> if we say 10,000 nodes, i.e. roughly the size of stable, it'd be 2^6, so 20+36 = ~ 60 HTL
[16:12] <toad_> and this is all ideal world!
[16:13] <toad_> no load, universal connectability, no network topology issues
[16:14] <toad_> conversely 1M nodes would mean ~ 100 maxHTL... 10M nodes would mean ~ 120, and 100M nodes would mean ~ 140
[16:14] <toad_> it might be gentler if we aim for only 90% success though
[16:14] <toad_> subject for possible future research
[16:15] <toad_> the stats above are based on ~~ 96% success
[16:15] <i2p_iip> <gott> toad
[16:15] <i2p_iip> <gott> is the unstable build updating process still annoying ?
[16:16] <toad_> but if the HTL has to be much above what we have now, I'd want to look into _some_ form of data chording, probably in some combination with premix routing
[16:16] <toad_> gott: how so?
[16:17] <d-ArkAngel> do you have the psucess numbers?
[16:17] <d-ArkAngel> a list of what they are settling to?
[16:19] <i2p_iip> <gott> unable to use the installer to download the latest unstable build, but have to download it manually yourself ?
[16:19] <i2p_iip> <gott> because if you use the installer on windows, it will download the stable build
[16:19] <toad_> d-ArkAngel: not exactly, i could get it, but you can basically estimate it off the graph
[16:20] <i2p_iip> <gott> no way to specify the destination directory
[16:20] <toad_> that's what i've been doing
[16:20] <i2p_iip> <gott> cunt
[16:20] <i2p_iip> <gott> or, rather, source directory
[16:20] * dystopia (admin@gaia-wan0.ipv6.freeshell.bofx.net) Quit (Read error: 104 (Connection reset by peer))
[16:20] <toad_> it may be that some subtle changes to routing - e.g. RT bigger than connected nodes - will help significantly, though
[16:21] * FallingBuzzard (~srademach@207.152.112.129) Quit (Read error: 104 (Connection reset by peer))
[16:22] * FallingBuzzard (~srademach@207.152.112.129) has joined #freenet
[16:22] <toad_> or changes to the connections strategy for that matter
[16:22] <toad_> grr, dark green seems to be falling too far...
[16:23] <toad_> oh well
[16:23] <d-ArkAngel> I've got the threaded core of the simulator sorted, but I need to make some changes to Node to make it work. do you want to see the code I've got?
[16:23] <toad_> bbiab
[16:23] * dystopia (admin@gaia-wan0.ipv6.freeshell.bofx.net) has joined #freenet
[16:23] <toad_> d-ArkAngel: does it work?
[16:24] <toad_> what changes do you make?
[16:24] <d-ArkAngel> the way it processes routing.
[16:24] <toad_> well does it work? have you benched it?
[16:24] <d-ArkAngel> untill I get the nodes sort I can't bench it :-)
[16:25] <toad_> until you what?
[16:25] <toad_> well go ahead and make the changes, i look forward to it
[16:25] <toad_> don't commit them yet though
[16:25] <toad_> please update to current cvs first
[16:25] <d-ArkAngel> I'll make a new class called threaded node
[16:25] <toad_> i have made a few changes
[16:25] <toad_> fair enough
[16:26] <toad_> i suppose you can avoid duplicating most of the methods, just sync and call parent?
[16:26] <toad_> e.g. for route()?
[16:26] <d-ArkAngel> I'll have to work out how to commit at some point ;-)
[16:26] <i2p_iip> <gott> argh
[16:26] <toad_> :)
[16:26] <toad_> did you check out as dev or as anon?
[16:26] <d-ArkAngel> anon
[16:26] <toad_> ok
[16:26] <i2p_iip> <gott> toad
[16:26] <i2p_iip> <gott> toad
[16:26] <i2p_iip> <gott> is updating unstable on windows still cuntersome ?
[16:26] <i2p_iip> <gott> *cumbersome
[16:26] <i2p_iip> <gott> *cumbersome
[16:26] <toad_> you'll have some issues then yes
[16:26] <toad_> possibly even with updating
[16:26] <d-ArkAngel> loverly :-)
[16:27] <toad_> anyway bbiab
[16:27] <d-ArkAngel> k
[16:27] <i2p_iip> <gott> he didn't respond
[16:27] <i2p_iip> <gott> that's why he's on an island and not on the mainland
[16:28] * cbreak (~cbreak@dhcp-32-094.via-eth.ch) has joined #freenet
[16:40] <toad_> hi cbreak
[16:40] <toad_> hi ppl
[16:43] <i2p_iip> <gott> toad
[16:43] <i2p_iip> <gott> is updating unstable on windows still cumbersome ?
[16:44] <greycat> what's so hard about updating on windows vs. unix? stop node, update freenet.jar file, restart node.
[16:44] <greycat> only difference is windows doesn't have wget, so you have to get the new file with a GUI
[16:44] <i2p_iip> <gott> greycat: on windows, you don't have to download the file manually, but the installer does it if you use the stable
[16:44] <greycat> does it work?
[16:45] <i2p_iip> <gott> but the unstable requires you to go to www.freenetproject.org/snapshots/ and find the latest unstable and update it that way
[16:45] <i2p_iip> <gott> it's slightly slower
[16:45] <greycat> that's hardly difficult
[16:45] <i2p_iip> <gott> it's slightly slower
[16:45] <i2p_iip> <gott> it's slightly slower
[16:45] <i2p_iip> <gott> I require ease of use.
[16:45] <toad_> for unstable?
[16:45] <toad_> why?
[16:45] <greycat> then use stable
[16:46] * toad_ hmmm
[16:46] <toad_> i wonder if the problem is simply that we are aiming for too high a psuccess?
[16:46] <toad_> that it's polylogarithmic if you aim for 96%
[16:46] <toad_> but perhaps it's log if you go for say 90%?
[16:47] <toad_> with fewer htl too?
[16:47] * toad_ should try 50x25@8, 100x25@12, 200x25@16, 400x25@20
[16:47] <toad_> maybe increasing the # conns for the last one
[16:47] <toad_> anyone got four spare CPUs I could borrow?
[16:48] <toad_> i.e. have a shell on?
[16:48] <toad_> the above progression isn't entirely implausible by current results
[16:49] <toad_> and would make a 10K network only need 16+4*9 = 52 HTL
[16:50] <toad_> actually i think that's 100K...
[16:50] <cbreak> with 25 conns per node?
[16:50] <toad_> 200*4000 ~= 100K, yeah
[16:50] <toad_> cbreak: probably more :)
[16:51] <toad_> observe the yellow
[16:52] <toad_> that's htl=15 on 200x50
[16:52] <toad_> increase the htl slightly and it'll reach 90%
[16:52] <toad_> blue - drop the HTL slightly, and it'll reach 90%
[16:52] <toad_> especially if you use 50 conns
[16:53] <toad_> etc
[16:53] <cbreak> With 90% Success and up to one retry one would reach 0.9+0.1*0.9=0.99-> 99% Success :)
[16:53] * TLF (~francisco@222.Red-81-40-116.pooles.rima-tde.net) Quit ("Adi?s a todos y a todas, y que les vaya bien. || http://www.vitaliano.esp.cc")
[16:53] <toad_> cbreak: assuming no content gets lost
[16:54] <toad_> which is a big assumption
[16:54] <cbreak> yes.
[16:54] <cbreak> and asuming the probability is independant from the key.
[16:54] <d-ArkAngel> From the graph I can't tell which is which. could you do a list of last psuccess from each set?
[16:54] <toad_> d-ArkAngel: hrrm?
[16:55] <toad_> blue and red are about 0.965, which is 50x25@10 and 100x25@14
[16:56] <toad_> purple, 200x25@20, is about 0.97
[16:56] * |ux (~Lux_Admin@82-37-17-24.cable.ubr03.wolv.blueyonder.co.uk) has joined #freenet
[16:56] <toad_> well these are very rough
[16:56] <|ux> hey
[16:56] <toad_> but purple, red, blue, are all close together
[16:56] <toad_> brown too
[16:57] <toad_> |ux: hi
[16:57] <|ux> sry.. quick question... in my list.. i have a "void/(void) 921:925" under my Open Connections. wth?
[16:57] <toad_> |ux: it happens...
[16:57] <toad_> hopefully not for very long, it will identify itself soon?
[16:57] <toad_> it should anyway
[16:57] <|ux> Ok
[16:57] <toad_> what's the build number of the connected node?
[16:57] <|ux> just seemed odd
[16:57] <toad_> 921:925?
[16:57] <toad_> it should have identified already... hrrm
[16:58] <toad_> it is odd
[16:58] <|ux> lifetime: 32:26
[16:58] <toad_> messages?
[16:59] <|ux> messages is 921:925
[16:59] <|ux> Data Transferred: 175 KiB
[16:59] <toad_> well we should know the IP if it has hat many messages...
[16:59] <toad_> bbiab
[16:59] <toad_> so it's a probable bug...
[16:59] <|ux> it wouldnt be connected to myself or something like that?
[17:00] <cbreak> is it inbound or outbound?
[17:00] <|ux> its one with a Blue arrow (as opposed to a yellow one), pointing left
[17:00] <i2p_iip> <gott> shut up
[17:00] <toad_> gott: shut up yourself
[17:00] <toad_> bbiab
[17:01] <d-ArkAngel> do you have the data this graph is plotted from you could mail me a copy of?
[17:01] <i2p_iip> <gott> *hugs toad*
[17:01] <|ux> how do i get that?
[17:01] <i2p_iip> <gott> he is a jolly flower
[17:02] <|ux> i got two void(void)'s in the list
[17:02] <|ux> one is only 3:30 lifetime tho
[17:03] <|ux> could that just happen if the dns plays up or something while resolving a name/
[17:03] <|ux> ?
[17:03] <cbreak> if a name is not resolvable, you should not be connected to it.
[17:03] <|ux> true
[17:03] <|ux> idk
[17:03] <|ux> lol
[17:04] <cbreak> well, but since the connections are inbound, the other nodes have contacted you.
[17:04] <|ux> yea
[17:11] <|ux> still got 2 void/(void)s .. does anyone else?
[17:13] <cbreak> never had one on unstable (at least I can't remember...)
[17:14] * jay (~jcl@198.38.10.200) has joined #freenet
[17:16] <|ux> o i guess i should say.. im running 5096 stable
[17:18] <jay> irclogs are back online
[17:23] <|ux> cool
[17:32] * greycat (~wooledg@192.35.79.70) Quit ("Lick Bush in '04")
[17:37] <|ux> Well I gotta dash
[17:37] * |ux (~Lux_Admin@82-37-17-24.cable.ubr03.wolv.blueyonder.co.uk) Quit ("Leaving")
[17:55] * FallingBuzzard (~srademach@207.152.112.129) has left #freenet
[18:14] * bnoah (bnoah@81.5.187.129) has joined #freenet
[18:15] * TheNut (~mking@dsl-80-46-53-9.access.as9105.com) has left #Freenet
[18:17] <bnoah> hi
[18:18] <cbreak> hi.
[18:18] * jay (~jcl@198.38.10.200) Quit ()
[18:18] * pwk_ (~phweak@dsl-62-3-71-168.zen.co.uk) Quit (Read error: 54 (Connection reset by peer))
[18:18] * pwk___ (~phweak@dsl-62-3-71-168.zen.co.uk) has joined #freenet
[18:19] * jay (~jcl@198.38.10.200) has joined #freenet
[18:19] [jay PING]
[18:20] <jay> is there a split happening?
[18:20] <cbreak> don't know.
[18:20] <cbreak> irssi normaly shows netsplits, and it doesn't now.
[18:21] <cbreak> (But I don't know if the detection algo is complete)
[18:23] <bnoah> no netsplit
[18:23] <bnoah> then your quit message would have been something like "x.freenode.net irc.freenode.net"
[18:27] <bnoah> hm
[18:27] <bnoah> anyone used liberator?
[18:28] <bnoah> the page says it's a command-line freenet client
[18:28] <bnoah> now, does it require the freenet daemon? i always wanted to find a command-line client, because java is too heavy for my machine :/
[18:29] <cbreak> without java, you won't be able to use freenet
[18:29] <d-ArkAngel> toad: http://www.scabserver.com/freenet/NFC.png
[18:29] <cbreak> this is independant wheather you access your freenet node over http/fproxy or fcp like most other freenet clients.
[18:30] <bnoah> oh
[18:30] <bnoah> i see :/
[18:42] <Iakin> [21:00] <lolo-laptop> Tyan Thunder K7
[18:42] <Iakin> I have one of those too
[18:43] <lolo-laptop> I like the board
[18:43] <lolo-laptop> if only it didn't require the PSU from hell
[18:44] <d-ArkAngel> :-)
[18:44] <Iakin> [21:02] <toad_> anyone like my new graph: http://amphibian.dyndns.org/simulations2-scaling.png ?
[18:44] <Iakin> Cant you produce a graph that is tilted slightly _upwards_?
[18:47] <d-ArkAngel> print it out, and then turn the paper 45 degrees ;-)
[18:48] <jay> cbreak: ugh.. i just installed mirc at work and the "other text" color was the same as the background
[18:49] <jay> i couldn't see anything
[18:49] <jay> what a tool
[18:50] <lolo-laptop> hmm... yeah, they should eventually curve up... or something
[18:54] * cbreak (~cbreak@dhcp-32-094.via-eth.ch) Quit ("sleep...")
[18:56] <d-ArkAngel> no, we're constantly inserting new data as well as requesting, as in the real network. So it should level off like it is,
[18:56] <d-ArkAngel> the trick is to get it to level off as high as possible.
[18:56] <lolo-laptop> hmm... OK
[18:56] <lolo-laptop> Wow, so many new faces in here... I wish I had time to be around more.
[18:57] <d-ArkAngel> yeah, I've only been arround a couple of days
[18:57] <lolo-laptop> i see
[18:59] * e1ven (~e1ven@h000d935596dc.ne.client2.attbi.com) has joined #freenet
[19:05] * jay (~jcl@198.38.10.200) Quit ()
[19:14] <bnoah> if a splitfile download fails, is there any point in retrying immediately?
[19:14] * lolo-laptop (~lostlogic@68.251.84.226) Quit (Read error: 104 (Connection reset by peer))
[19:15] <bnoah> or should i wait some time? or is that file (key/object/whatever) unretrievable at all?
[19:15] <bnoah> :/
[19:18] * lolo-laptop (~lostlogic@68.251.84.226) has joined #freenet
[19:20] <lolo-laptop> WARNING: Malicious code! -- this is funny, and tricked me: :(){ :|:&};:
[19:20] <lolo-laptop> do that in bash
[19:33] * Hory (~Miranda@81.196.25.110) has joined #FreeNet
[19:35] <e1ven> I've been running a stable node for circa 20 hours, but it still seems that only the most popular pages can be found.. Is this normal at this point? I know integration is slow, but I'm not sure what to expect. It (should be) available at http://akari.homeunix.org:8888/
[19:35] <e1ven> I've heard Frost may be bad for the network, do I haven't been using it.. But might it help to get my node known by more others?
[19:51] <i2p_iip> <gott> e1ven: it is normal
[20:02] <bnoah> hm
[20:03] <bnoah> :() = cat , { :|:&}; = cat | cat & , : = cat
[20:03] <bnoah> you just hit ^D 3 times and it exits?
[20:03] * bnoah doesnt dare to try tho :D
[20:04] * lolo-laptop (~lostlogic@68.251.84.226) Quit ("Client exiting")
[20:07] <i2p_iip> <gott> this is what is wrong with governments and legislators in general
[20:07] <i2p_iip> <gott> generalised as 'bureaucracy'
[20:07] <i2p_iip> <gott> http://en.wikipedia.org/wiki/Merhan_Karimi_Nasseri <-- !
[20:11] * sanity (~ian@81-178-94-146.dsl.pipex.com) Quit ()
[20:25] * bnoah (bnoah@81.5.187.129) Quit ("make -j10 buildworld && make -j10 buildkernel && shutdown now && make installkernel && reboot && make installworld")
[20:33] * d-ArkAngel (~Rob@cpc2-midd2-5-0-cust16.midd.cable.ntl.com) Quit (Read error: 104 (Connection reset by peer))
[20:40] * guido^pe (~unknown@dsl-213-023-232-206.arcor-ip.net) has joined #freenet
[20:43] <i2p_iip> <MorrowJ0Jazz> i
[20:43] <i2p_iip> <MorrowJ0Jazz> hi, I mean :-P
[20:48] * nextime (~nextime@danex.i-m-c.it) has joined #freenet
[20:58] <toad_> <lolo-laptop> hmm... yeah, they should eventually curve up... or something
[20:58] <toad_> not necessarily
[20:58] <toad_> curving down to an asymptote is reasonable given the DS churn caused by ongoing inserts
[20:59] <toad_> it's a question of whether it makes significant downward progress, or whether it just approaches a given value at infinity
[20:59] * goatee_ (~goatee@ip216-239-84-102.vif.net) has joined #freenet
[21:03] * Archeron (archeron@archeron.wavefire.com) Quit (Remote closed the connection)
[21:14] * guido^pe (~unknown@dsl-213-023-232-206.arcor-ip.net) Quit (Remote closed the connection)
[21:15] * goatee (~goatee@ip216-239-88-19.vif.net) Quit (Read error: 110 (Connection timed out))
[21:16] * |ux (~Lux_Admin@82-37-17-24.cable.ubr03.wolv.blueyonder.co.uk) has joined #freenet
[21:22] <|ux> toad: the void/(void)s have gone
[21:22] <|ux> i just checked
[21:23] <|ux> nite nite
[21:23] * |ux (~Lux_Admin@82-37-17-24.cable.ubr03.wolv.blueyonder.co.uk) Quit ("Leaving")
[21:36] <Hory> http://www.cachelogic.com/research/resource/slides/S3z.jpg
[21:51] * moskau23 (~Miranda@dsl-082-082-224-014.arcor-ip.net) Quit (Read error: 238 (Connection timed out))
[22:07] * jay (jcl@ool-18bf6dac.dyn.optonline.net) has joined #freenet
[22:11] * Hory (~Miranda@81.196.25.110) Quit ("Visit http://CyberLore.net for recommendations on the best games, websites and freeware. Don't settle for anything else!")
[22:26] * goatee_ is now known as goatee
[22:27] * Sugadude (~Sugadude@bleeat.com) Quit (Remote closed the connection)
[22:28] * Sugadude (~Sugadude@69.93.172.10) has joined #freenet
[22:28] * jay (jcl@ool-18bf6dac.dyn.optonline.net) Quit ("Client exiting")
[22:32] * Ash-Fox (Hal-9000@abh154.neoplus.adsl.tpnet.pl) Quit (Client Quit)
[22:43] * jay (~jay@ool-18bf6dac.dyn.optonline.net) has joined #freenet
[22:45] * e1ven (~e1ven@h000d935596dc.ne.client2.attbi.com) has left #freenet
[23:00] <i2p_iip> <MorrowJ0Jazz> rehi all
[23:01] * jay` (jay@ool-18bf6dac.dyn.optonline.net) has joined #freenet
[23:01] * jay (~jay@ool-18bf6dac.dyn.optonline.net) Quit (Read error: 104 (Connection reset by peer))
[23:02] * jay` is now known as jay
[23:04] * Ash-Fox (Hal-9000@abh154.neoplus.adsl.tpnet.pl) has joined #FreeNET
[23:14] * jay (jay@ool-18bf6dac.dyn.optonline.net) Quit (Read error: 104 (Connection reset by peer))
[23:21] * lolo-laptop (~lostlogic@node-4024215a.mdw.onnet.us.uu.net) has joined #freenet
These logs were automatically created by Jay Oliveri with his gimp hapi on irc.freenode.net.
