Timestamps are in GMT/BST.
[0:58] * JoshPara1roid is now known as JoshParadroid
[2:05] * Disconnected.
[2:06] [freenode-connect VERSION]
[10:47] * FreenetLogBot (n=PircBot@emu.freenetproject.org) has joined #freenet
[10:47] * Topic is 'http://freenetproject.org | Stable: Upgrade to 5106 | Unstable:60275 |Channel logs: http://emu.freenetproject.org/irc/ | #freenet-alphatest for anyone who wants to help test 0.7 | DO NOT use the freenet packages supplied by debian! | A fresh install of freenet will take some time to become useful, be patient'
[10:47] * Set by NullAcht15 on Wed Nov 23 22:53:27 UTC 2005
[10:47] [freenode-connect VERSION]
[10:47] <TheSeeker> hello FreenetLogBot @_@
[10:48] <TheSeeker> stupid thing making me think I'd disconnected...
[11:26] <freenet> thanks
[11:26] * freenet (n=freenet@cpc1-nott8-4-1-cust246.nott.cable.ntl.com) Quit ("leaving")
[13:22] * mazzanet_ (n=irc@unaffiliated/mazzanet) has joined #freenet
[13:23] * mazzanet (n=irc@unaffiliated/mazzanet) Quit (Read error: 104 (Connection reset by peer))
[13:37] * FallingBuzzard (n=FallingB@66.151.22.70) has joined #freenet
[13:52] <sanity> the new gui designer in NetBeans is fscking cool
[13:53] <sanity> its basically drag and drop, and it infers alignment of objects (rather than mucking around with layout managers)
[14:09] <nextgens> who is Roman Glebov ? :)
[14:46] <sanity> i don't know - but so far, I like him
[14:53] * greycat (i=rfc1413@wooledge.org) has joined #freenet
[15:16] <nextgens> here is the problem :
[15:16] <nextgens> it seems that .7 runs or will run on free jvms in the near future,
[15:17] <nextgens> however it works, there is still at least 3 issues :
[15:17] <nextgens> 1) the most important : Load related :
[15:17] * NullAcht15 (n=NullAcht@dslb-082-083-253-238.pools.arcor-ip.net) has joined #freenet
[15:18] <nextgens> indeed, running .7 on a free jvm is feasible but will be out of interest as long as load limiting (Backoff) is inadapted
[15:18] <sanity> why?
[15:19] <nextgens> because every link is "backoffed" and the node is always RNFing because no link is active
[15:19] <sanity> what does that have to do with free jvms?
[15:19] <nextgens> they can't use optimized libraries atm
[15:19] <sanity> isn't that just a general problem with load balancing right now?
[15:20] <nextgens> well, if you do use, x86, sun's jvm, it's working ;)
[15:20] <sanity> oh, so basically you are saying that freenet runs too slowly on free jvms?
[15:20] <nextgens> yes
[15:20] <nextgens> at least to slowly regarding how load balancing is implemented atm
[15:20] <nextgens> that's the first issue
[15:21] <sanity> ok
[15:21] <nextgens> 2)
[15:21] <nextgens> it requires a bit of patching as some classes are missing from the main classpath ...
[15:22] <nextgens> and especialy Digest (SHA256) things ...
[15:23] <sanity> it sounds like there is a general problem with backoff if it is causing nodes to reject requests when they don't need to
[15:24] <nextgens> so, either we do bundle gnu-crypto/cryptix/bouncycastle in the main tree, use the lib. from the jvm if aviable and fallback on them if needed
[15:25] <nextgens> or we will have to maintain a different tree (or a special thing/flag in the node's config) to detect/force that
[15:25] <nextgens> for free jvms
[15:25] <nextgens> (might be a branch in svn as well)
[15:25] <sanity> well, the value of supporting free vms is mainly for those that want to run freenet on an entirely free software stack. such people shouldn't be afraid to hack around a bit if necessary
[15:25] <sanity> the problem wiht branching svn is that we then need to keep the branches in sync
[15:26] <nextgens> but we will have to decide if we'd like to support it or not
[15:26] <nextgens> we can release tags of branches too ;)
[15:26] <sanity> well, it sounds like these problems are caused by shortcomings in the freevms - right?
[15:26] <sanity> if so, we should file bugs in the free vms
[15:26] <nextgens> not shortcomings : patent related afaik
[15:27] <sanity> in the past we have had a good working relationship with them
[15:27] <sanity> oh, ouch
[15:27] <sanity> sha256 is patent encumbered?
[15:27] <nextgens> for crypto at least ... or maybe not patents for gnu-crypto, but US exports law
[15:28] <nextgens> sanity> I don't think so, but those libs. aren't providing only sha
[15:28] <nextgens> and finally, 3)
[15:28] <nextgens> the main point imho :
[15:30] <sanity> i thought they relaxed us export laws?
[15:30] <nextgens> well, not really a point, but an open ask ; should "we" officially support free-jvms, will we be able to deal with jvm-specific bugs, ... won't it delay the release of .7
[15:30] <nextgens> ?
[15:30] <nextgens> or should we plan to do it .... later ^-^
[15:30] <sanity> i think post-0.7
[15:31] <sanity> we need to get 0.7 out ASAP
[15:31] <nextgens> agreed
[15:31] <sanity> people will hate me for saying this, but the main purpose of supporting free VMs is to satisfy free software advocates
[15:31] <nextgens> though, it would have been great to announce that .7 is free-jvm compliant when slashdotting ;)
[15:31] <nextgens> sanity> no :)
[15:32] <sanity> and advocating free software, while a nobel cause, isn't our cause (although if we can do it without compromising our goals we should)
[15:32] <nextgens> sanity> to allow them to run java things on "uncommon" hardware
[15:32] <sanity> that too, but the reality is that it is only of interest to a tiny minority of our users
[15:32] <nextgens> that's more arguable
[15:33] <sanity> so, yes, we should support free vms if we can, but it shouldn't delay the 0.7 release
[15:33] <sanity> if there are changes that can be made to the code that makes it easier to run on free vms, without making anything else worse, then we should make those changes
[15:34] <sanity> but otherwise, we should leave it until after 0.7
[15:34] <nextgens> I don't developp software to be run by masses : especially not for windows users :p if they can use the software I'm working on, cool ; otherwise, I wouldn't take any decision based on the fact that the majority is using some crappy software
[15:34] <sanity> the more people that use freenet, the better for freenet
[15:34] <nextgens> (windows is the "example" there)
[15:34] <nextgens> agreed
[15:35] <sanity> even if they are windows users
[15:35] <nextgens> :-D
[15:35] <sanity> there is no shame in focussing on ease of use, and there is no shame in asking what will make freenet most useful to the largest number of people. indeed, we have a responsibility to do that
[15:36] * smains (n=smains@tor/session/x-a361a473ea46d359) has joined #freenet
[15:36] <nextgens> agreed too
[15:36] <nextgens> however, I make a nuance :
[15:36] <sanity> so, do you think it is reasonable to say that pre-0.7 we can look at how to make freenet run on free VMs, but we don't let it delay 0.7's release?
[15:36] <nextgens> as long as " ease of use" doesn't interferes with technical and architectural choices
[15:37] <sanity> well, unfortunately "technical and architectural choices" could potentially be used to veto almost anything
[15:37] <sanity> it doesn't matter how technically elegant a piece of software is if nobody uses it :-)
[15:37] <nextgens> that's a part on what I'd like to get matthew's point as he is the main^wonly devl ;)
[15:38] <sanity> just look at Dijjer :-)
[15:38] <nextgens> hehe
[15:39] <nextgens> I will do some benchmarks with various crypto libs/jvms ... to evaluate wich one seems the most promessing
[15:39] <sanity> ok
[15:39] <nextgens> and if he aggrees, I'll create a branch in subversion
[15:39] <sanity> boy do i get a lot of spam, even with greylisting
[15:40] <nextgens> sanity> ? you shouldn't anymore ...
[15:40] <nextgens> are they flagged as spam by spamassassin ?
[15:41] <sanity> i don't see any spam header
[15:41] <nextgens> atm, sablevm is hammering the cpu while running
[15:41] <nextgens> sanity> huh ?
[15:41] <nextgens> you should :)
[15:42] <sanity> what is the name of the header spam assassin adds?
[15:42] <nextgens> maybe I forgot to add your domain on the spamfilter :$
[15:42] <nextgens> X-SPAM-STATUS or something like that
[15:42] <sanity> i'm looking at a spam that went to ian@freenetproject.org
[15:42] <sanity> oh, no
[15:42] <sanity> it was ian@locut.us
[15:42] <nextgens> ;)
[15:42] <nextgens> ok, I'll add it for this domain too
[15:42] <sanity> thanks
[15:42] <sanity> and @cematics.com too
[15:43] <nextgens> ok
[15:43] <sanity> in fact, why not just add it for all domains?
[15:43] <sanity> hopefully gmail respects that header
[15:43] <nextgens> if you'd like ... but there is always a small probability to missflag/drop legit. mail
[15:43] <sanity> i can accept that risk
[15:43] <nextgens> ok
[16:43] <toad_> hi
[16:50] <toad_> hi
[16:56] <sanity> turns out that newsbyte spammed the Freenet article on Wikipedia *after* i exposed the fact that he had spammed the software patents article. what a shameless asshole.
[17:04] <toad_> rah: see you then
[17:06] <toad_> nextgens: THEY DONT NEED TO USE OPTIMIZED LIBRARIES
[17:06] <toad_> because they all use libgmp!
[17:06] <toad_> or is it a hashing problem?
[17:06] <toad_> [15:23] <sanity> it sounds like there is a general problem with backoff if it is causing nodes to reject requests when they don't need to
[17:06] <toad_> that's the whole point of backoff
[17:06] <nextgens> hi toad_ :)
[17:07] <toad_> they reject requests in order to avoid timing out on requests
[17:07] <nextgens> we agree :)
[17:08] <nextgens> but here, my Kaffe node needs at least 3 sec. to reply to a ping request with only one active link and beeing idle
[17:08] <toad_> rijndael isn't patented, nor is twofish, both were AES finalists, so _we are allowed to use them_, and so is GNU Classpath
[17:08] <toad_> [15:31] <sanity> people will hate me for saying this, but the main purpose of supporting free VMs is to satisfy free software advocates
[17:08] <toad_> THIS IS NOT TRUE
[17:08] <toad_> the main purpose of supporting free VMs is to be able to package freenet properly
[17:08] <toad_> so we can get it back into debian
[17:09] <toad_> which around 10% of our target audience runs
[17:09] <toad_> at least
[17:09] <toad_> the second purpose is not to be dependant on the Evil Corporation Sun
[17:09] <toad_> and the third purpose is to satisfy free software advocates
[17:09] <nextgens> :p
[17:09] <nextgens> :)
[17:10] <nextgens> toad_> whirlpool or something like that is subject to patents
[17:11] <nextgens> toad_> I tried gnu-crypto, it works but require a bit of patching
[17:11] <nextgens> and I would say it doesn't perform "fast" to be nice ;)
[17:12] <toad_> nextgens: we don't use whirlpool
[17:14] <Sugadude> toad_: Sun isn't evil, it's just misunderstood. Like Microsoft. :)
[17:14] <nextgens> I know
[17:15] <toad_> heh
[17:16] <toad_> Sun is an amoral corporation legally obliged to do whatever it can get away with. it's a single point of failure, and it refuses to open source the JVM. we can trust Sun as far as we can throw it.
[17:17] <toad_> but the main reason is packaging
[17:17] <Sugadude> toad_: Don't take this the wrong way, but you just turned me on beyond belief. :)
[17:17] * Sugadude was kicked from #freenet by toad_
[17:17] * toad_ takes it the wrong way ;)
[18:02] * NullAcht15 (n=NullAcht@dslb-082-083-253-238.pools.arcor-ip.net) Quit (Connection timed out)
[18:02] * NullAcht15 (n=NullAcht@dslb-082-083-177-086.pools.arcor-ip.net) has joined #freenet
[18:04] * danderso1 (n=dave@natulte.net) has joined #freenet
[18:08] * danderso1 (n=dave@natulte.net) has left #freenet
[18:29] <nextgens> newsbyte's answer on @tech worth even less than his bugreport :XD
[18:29] <nextgens> sleon> you'll be happy :)
[18:30] <sleon> hr
[18:30] <sleon> :)))
[18:30] <nextgens> worth+s
[18:30] * sleon reads
[18:31] <sleon> rofl
[18:37] * danderso1 (n=dave@natulte.net) has joined #freenet
[18:38] * danderso1 (n=dave@natulte.net) Quit (Client Quit)
[18:49] <sanity> is it just me, or do his emails look like they have been passed through babelfish, english to swahili, and back again
[19:25] * Eol (n=Eol@tor/session/x-bc5f8a848974fba5) has joined #freenet
[19:32] * Tubbie (n=tubbie@ipd50a5ac4.speed.planet.nl) Quit ()
[19:47] <sandos> ;)
[19:49] * FallingBuzzard (n=FallingB@66.151.22.70) Quit (Nick collision from services.)
[19:49] * FallingBuzzar1 (n=FallingB@66.151.22.70) has joined #freenet
[21:00] * uluru (n=chatzill@213-140-21-235.fastres.net) has joined #freenet
[21:01] <uluru> hi everybody
[21:01] <uluru> is a freenet mac version available?
[21:04] <sbc> uluru: If you have java you should be set to go. If you have a newish mac it should be ready to go.
[21:04] <sbc> just follow the instructions given for unix on the webpage.
[21:12] <uluru> ok, thanks
[21:12] <uluru> yes, i have the latest mac os X,10.4
[21:13] <sbc> that's tiger right? Then you should have acces to java 1.5. You should be good to go.
[21:14] <sbc> uluru: if you do run into trouble, fee free to ask.
[21:14] <uluru> so with mac ther's no gui.. only shell
[21:14] <uluru> yes, tiger
[21:14] <uluru> thanks for your help
[21:14] <uluru> do you know where i can find some detailed paper about the 0.7 release?
[21:15] <uluru> i'm curious about it
[21:15] <uluru> i have an idea, i need to know if can be implemented
[21:16] <uluru> i go away few minutes
[21:20] * NullAcht15 (n=NullAcht@dslb-082-083-177-086.pools.arcor-ip.net) Quit (Success)
[21:21] * NullAcht15 (n=NullAcht@dslb-082-083-218-205.pools.arcor-ip.net) has joined #freenet
[21:23] <sbc> uluru: You still here? You don't loos much gui wise on the mac. The primary freenet interface is web based anyway, so if you use firefox on win, linux or mac is no difference, and frost runs fine on mac to.
[21:24] <sbc> uluru: Besides the papers linked from the webpage I can't point you to any detailed papers about 0.7. But if you have specific questions, do ask them here, I'm sure someone can answer them.
[21:27] <uluru> i'm back
[21:30] * FallingBuzzar1 (n=FallingB@66.151.22.70) has left #freenet
[21:42] * greycat (i=rfc1413@wooledge.org) Quit ("This time the bullet cold rocked ya / A yellow ribbon instead of a swastika")
[21:43] * smains (n=smains@tor/session/x-a361a473ea46d359) Quit ("Leaving")
[21:49] * Maeglin (n=Eol@tor/session/x-101502c3e5cd91f4) has joined #freenet
[21:58] <uluru> ..
[22:10] * Bombe (n=bombe@213-102-97-27.cust.tele2.de) has joined #freenet
[22:13] * Lomion (n=Eol@tor/session/x-1d7d283c858941ba) has joined #freenet
[22:23] * Eol (n=Eol@tor/session/x-bc5f8a848974fba5) Quit (Nick collision from services.)
[22:26] * uluru (n=chatzill@213-140-21-235.fastres.net) Quit (Remote closed the connection)
[23:21] * dannycool (n=hotdanny@84.54.10.131) has joined #freenet
[23:21] <dannycool> hi room
[23:22] * sbc (n=sbc@83.72.201.52.ip.tele2adsl.dk) Quit ("Leaving")
[23:25] * Ash-Fox (i=UNKNOWN@efb3.neoplus.adsl.tpnet.pl) Quit (Nick collision from services.)
[23:25] * Ash-Fox (i=UNKNOWN@edx239.neoplus.adsl.tpnet.pl) has joined #FreeNET
[23:26] * sleon (n=sleon@e180018147.adsl.alicedsl.de) Quit (Remote closed the connection)
[23:31] * dannycool (n=hotdanny@84.54.10.131) Quit (Read error: 104 (Connection reset by peer))
[23:56] * verl (i=verl@h56n2fls31o877.telia.com) Quit ()
Irc logs of #freenet : 2008 2007 2006 2005
These logs were automatically created by FreenetLogBot on chat.freenode.net using the Java IRC LogBot.