Timestamps are in GMT/BST.
[21:16] [freenode-connect VERSION]
[22:03] [freenode-connect VERSION]
[22:39] [freenode-connect VERSION]
[23:05] * Disconnected.
[23:17] [freenode-connect VERSION]
!
[0:04] <hobx> online iq tests are crap
[0:04] <hobx> sorry to tell ya
[0:04] <Elly> I didn't take an online one
[0:04] <Elly> I was given one by a child psychologist and professional counsellor
[0:04] <Elly> it took 5 hours >=(
[0:05] <hobx> you a kid?
[0:05] <Elly> I'm 16 now
[0:05] <Elly> I was 12 when I took it
[0:06] <Elly> no, 11, my mistake
[0:06] <hobx> should be careful with that. They scale up child IQs.
[0:06] <hobx> You should take another one now.
[0:07] <Elly> I will
[0:07] * Elly shrugs
[0:07] <Elly> the counsellor said that given the opportunity I should skip 6th grade and go straight to 7th >=\
[0:07] <hobx> also, don't show off about your IQ. It puts you in bad company - most people who do are idiots.
[0:08] <hobx> Incidentally, I did skip 6th and go directly to 7th :-)
[0:08] <Elly> How ironic
[0:08] <Elly> It was NOT fun
[0:08] <Elly> because I was coming from a french private school to an american public school
[0:09] <hobx> and the french school was better?
[0:09] <Elly> no
[0:09] <Elly> it was just different
[0:09] <Elly> american and french school cultures are different beyond words
[0:09] <hobx> I went from an american public school to a swedish one.
[0:10] <hobx> turns out in Sweden, they aren't big on actually making sure kids learn stuff in school.
[0:10] <Elly> Haha
[0:11] <hobx> funny for you - you don't have to teach them once they reach university...
[0:11] <linyos> they don't segregate the smart students there, right?
[0:12] <hobx> linyos: Right. ANY advanced groups for anything (other than gym and music and other dumb stuff) is completely unallowed.
[0:12] <Elly> heh
[0:12] <Elly> I was in our gifted program until it terminated in 9th grade
[0:12] <hobx> I know mathematicians who got in trouble when they were in school because they worked ahead in the book.
[0:15] <linyos> here (chicago north shore) they formally start at 4th grade.
[0:15] <linyos> and informally they do it even earlier.
[0:15] <linyos> (or so my parents were told.)
[0:16] <Elly> actually I'm browsing a couple chapters ahead in my calc book right now
[0:16] <Elly> rectangular approximation methods
[0:21] <hobx> approximating integrals?
[0:21] * Ash-Fox (i=UNKNOWN@edm70.neoplus.adsl.tpnet.pl) Quit (Connection timed out)
[0:26] <Elly> yes
[0:26] <hobx> use monte-carlo integration
[0:26] <hobx> it doesn't work very well, but it is much more fun.
[0:31] <Elly> hmm
[0:35] <hobx> stupid peace of shit dist-upgrade
[0:35] <hobx> piece even
[0:36] <hobx> I can't go to bed until the fucking thing has halted, but it just won't stop a config question.
[1:04] <linyos> Elly: what do you think of the "last mile application-protocol filtering problem?"
[1:04] <linyos> Elly: meaning, if internet users use a small set of applications, it is easy to lock down (filter, control) each one.
[1:06] <linyos> the worst-case is that the only application people use is a web browser. hence, the state needs only to install a filtering web proxy, and it has as much control as it cares to use.
[1:07] <linyos> the best case would be if operating systems required complex, diverse network interactions to function
[1:08] <Elly> not gonna happen
[1:08] <Elly> https :O
[1:09] <linyos> i've been thinking about distributed network-wide computing as a way to create this incomprehensible flurry of data
[1:10] <Elly> creating incomprehensible data is easily
[1:10] <Elly> it's a quick hack to the TCP/IP stack - accept connections on any port
[1:10] <Elly> and then dump all data to /dev/null
[1:10] <Elly> no stack hack needed even, just an iptables change
[1:11] <linyos> the point is: suppose you're a state and you want to lock down the internet. how much collateral damage must you cause?
[1:11] <linyos> i am trying to increase that damage.
[1:11] <Elly> uhh
[1:11] <Elly> 'the internet' and 'lock down' are pretty much incompatible
[1:11] <Elly> as it stands currently
[1:11] <Elly> you have to ban all encryption
[1:11] <Elly> and especially all steganography
[1:12] <linyos> Elly: http://emu.freenetproject.org/pipermail/tech/2005-October/000154.html
[1:12] <linyos> for one way it could happen.
[1:13] <linyos> that's the worst-case scenario. unfortunately it also seems to be what we have today.
[1:13] <linyos> (for the most part)
[1:13] <Elly> as I said
[1:13] <Elly> they'd have to ban all crypto for one
[1:13] <linyos> no problem.
[1:13] <Elly> actually
[1:13] <linyos> ban all applications except web browsing.
[1:13] <Elly> huge problem :)
[1:13] <Elly> uhh
[1:14] <Elly> that would destroy the internet
[1:14] <Elly> also
[1:14] <linyos> read my post.
[1:14] <Elly> the cat is already out of the bag
[1:14] <Elly> yes, you said
[1:14] <Elly> but you can't filter shit you can't read
[1:15] <linyos> if people are encoding encrypted data in web pages, they will be found out sooner or later by the police.
[1:15] <Elly> and the crypto cat is already out of the bag, as I just said - RSA and Rijndael and IDEA and GPG and ...
[1:15] <Elly> wrong. certain types of data file are indistinguishable from encrypted data
[1:15] <Elly> especially compiled executables
[1:15] <Elly> and steganography software lets you hide encrypted data pretty much anywhere
[1:15] <linyos> sure, but try communicating on a large scale over such channels.
[1:16] <Elly> if I'm sending you a 5MB executable I can safely stash 500KB of data in it
[1:16] <linyos> when the police are running statistical analyses on your browsing habits.
[1:16] <Elly> that's computationally infeasible
[1:16] <linyos> if you're sending me an executable you're already doing something abnormal.
[1:16] <linyos> they would then check to see, what program is this?
[1:16] <Elly> detecting steganography is time-intensive unless you already know the key
[1:17] <linyos> they would find: oh wow, this is a program we haven't seen before. wonder what it is?
[1:17] <Elly> or take for example an mpeg
[1:17] <linyos> we better give this web session to one of our policemen to take a look at.
[1:17] <Elly> if I were to send you a 30MB music video mpeg
[1:17] <Elly> linyos, it's not feasible
[1:17] <Elly> the whole idea
[1:17] <Elly> the chinese can't lock down the internet; neither can anyone else
[1:18] <linyos> it is completely possible, if you are willing to seek out unusual patterns of activity and hire people to investigate them.
[1:18] <Elly> I suspect that any government that tried to block all protocols except HTTP
[1:18] <Elly> would find itself extremely unpopular extremely quickly
[1:18] <Elly> furthermore, the US government lacks the authority to do so
[1:19] <linyos> why unpopular? most people would not notice.
[1:19] <Elly> uhh
[1:19] <Elly> I beg to differ
[1:19] <Elly> millions of people use, say, instant-messaging software
[1:19] <Elly> and HTTPS for internet banking
[1:19] <Elly> and peer-to-peer apps
[1:19] <Elly> and IRC, for that matter
[1:19] * Bombe (n=bombe@p5483E99C.dip.t-dialin.net) has joined #freenet
[1:20] <linyos> it would not be the first time a repressive government inconvenienced people.
[1:20] <Elly> linyos: I can't think of such a government off the top of my head
[1:20] <Elly> the U.S. is unlikely to do such a thing anytime soon
[1:20] <linyos> i agree.
[1:20] <Elly> and if the U.S. doesn't do it, other countries will find it exceedingly hard to erect barriers
[1:20] <Elly> witness the chinese web proxies and triangle boy
[1:20] <Elly> and for that matter freenet
[1:21] <linyos> china's blocking efforts are pretty effective, right?
[1:21] <Elly> not at all
[1:21] <Elly> they're completely ineffective against a user with even a modicum of technical skill
[1:21] <Elly> heh
[1:21] <linyos> sure, but how many users are like that?
[1:22] <Elly> the ones that matter :)
[1:22] <linyos> no, they don't matter.
[1:22] <linyos> if people could casually speak their minds on the net, it would be a big problem.
[1:23] <Elly> to quote from the article on the subject: The government does not appear to be systematically examining Internet content, as this appears to be technically impractical.
[1:23] <linyos> don't they block keywords at the router level?
[1:24] <Elly> N
[1:24] <Elly> they have web proxies that do that
[1:24] <Elly> quote again
[1:24] <Elly> The firewall is largely ineffective at preventing the flow of information and is rather easily circumvented by determined parties by using proxy servers outside the firewall. VPN and ssh connections to outside mainland China are not blocked, so circumventing all of the censorship and monitoring features of the Great Firewall of China is trivial for those who have these secure connection methods to computers outside mainland China available to
[1:24] <Elly> them.
[1:25] <linyos> but it's effective enough for the government's purposes
[1:25] <Elly> no, it's not:
[1:25] <Elly> One notable case in which this occurred was in response to a school explosion in 2001, when local officials tried to suppress the fact the explosion resulted from children illegally producing fireworks. By the time local officials forced the story to be removed from the Internet, the news had already been widely disseminated.
[1:25] <linyos> it is not much threatened by a few people who are determined to break through the firewall
[1:26] <linyos> that's something different, that has to do with press freedom in general and making concessions, etc
[1:26] <Elly> quote: The Chinese government does block access to certain sites by IP address in what has been called the Great Firewall of China. These blocks are easily circumvented and are generally ineffective at preventing external news from entering the country.
[1:27] <linyos> "from entering the country", what does that matter? the point for them is that the average internet user can't google up "democracy" and form his own opinions.
[1:27] * Elly shrugs
[1:28] <Elly> it's not working
[1:28] <Elly> if I were in China, I'd be able to google up democracy and find it
[1:29] <Elly> if you're going to assume the average user is clueless, then pretty much anything would stop them from getting any information you want to stop them from getting
[1:29] <Elly> and there's nothing we can do about it
[1:29] <Elly> the average user does not have freenet installed or any other anonymizer
[1:29] <Elly> they have Internet Explorer on Windows XP with AOL dialup
[1:30] <Elly> I think it's fairly safe to assume that 10% of all users are saavy enough to bypass such blocking
[1:31] <Elly> (that number is drawn from the market-share of alternate browsers)
[1:31] <linyos> very few users are (a) informed enough to circumvent blocking, (b) motivated enough to look for open proxies, etc
[1:31] <linyos> that, for the regime, is success enough.
[1:31] <Elly> anyone determined enough to install firefox
[1:31] <Elly> can be considered determined enough to install freenet
[1:31] <Elly> 10% is plenty
[1:31] <Elly> 10% of the populace is more than enough for a revolution
[1:32] <linyos> it's not enough to change popular feeling against the government's policies.
[1:32] <Elly> let's see
[1:32] <linyos> with ten percent, you get a small minority. (that anyway cannot express its views in public)
[1:33] <Elly> 1991
[1:33] <Elly> err, 1989 rather
[1:33] <Elly> they managed the fall of the berlin wall sans internet
[1:33] <Elly> and amid a repressive government
[1:33] <linyos> sure, news spreads pretty well.
[1:34] <Elly> I'm going to say that short of banning all encryption, it's not possible that such a thing can come to pass
[1:34] <Elly> and banning crypto makes e-commerce very hard
[1:34] <linyos> i'm talking about a popular culture of freely criticizing the government. that is really threatening, and that is what they now suppress very well.
[1:34] <Elly> the chinese?
[1:34] <linyos> nah, you just proxy the ssl sessions through a government proxy.
[1:35] <linyos> yeah.
[1:35] <Elly> the great firewall can't last indefinitely
[1:35] <Elly> and uhh, how would that help? you don't know the SSL keys
[1:35] <linyos> the proxy connects to the site for you.
[1:35] <linyos> your crypto session is with the proxy.
[1:36] <Elly> then the proxy is a single point of failure
[1:36] <Elly> hacks0r the proxy and you own SSL, basically
[1:36] <linyos> duh, so have lots of them
[1:36] <linyos> big deal
[1:36] <Elly> also, that would be trivial to circumvent
[1:36] <Elly> how do you spot an SSL session?
[1:36] <linyos> it's easy, the protocol is identifiable
[1:37] <linyos> your web proxies would block them
[1:37] <Elly> that'd be a seriously grunty proxy
[1:37] <linyos> different port, etc
[1:37] <Elly> how about a cryptographic protocol with no header?
[1:37] <Elly> ie, fetch the key over normal HTTP, then just open a session and send the crypted session key
[1:38] <linyos> the point is, if people ever started, in mass, to exploit any such thing to freely express themselves online, it would be intolerable by the state.
[1:39] <Elly> right.
[1:39] <Elly> Which is why it's necessary.
[1:39] <Elly> We should make it.
[1:39] <linyos> and why once it happens, it will be suppressed at any price.
[1:39] <Elly> 3072-bit RSA keys...you just get the key over HTTP, and connect to a specific random port, and the first thing you send is the encrypted key
[1:40] <Elly> indistinguishable from binary data without (technically infeasible) analysis
[1:41] <Elly> you could even if you want have it pretend to be a zip file or a gzip file
[1:41] <Elly> which makes analysis even MORE infeasible
[1:42] <Elly> I don't see how you could reasonably suppress that
[1:42] <linyos> my point is that if you succeed in doing that, an intolerable situation will be created.
[1:42] <linyos> tyrants cannot tolerate freedom of expression.
[1:42] <Elly> right
[1:42] <linyos> it leads directly to challenges to their power.
[1:42] <Elly> which is why we should do it now
[1:43] <Elly> and start air-dropping CD-ROMs of it on china.
[1:43] <linyos> no, which is why they will pull the plug on the internet before they let it spread
[1:43] <Elly> heh
[1:43] <Elly> then we go to ad-hoc wifi nets
[1:43] <Elly> super-fun for the government
[1:43] <linyos> they will ban wifi hardware.
[1:43] <Elly> too many people already have it
[1:43] <linyos> they have power, they can do just about anything.
[1:43] <Elly> I myself have enough hardware on my desk to give wifi access to a small city
[1:44] <linyos> i sympathize with wifi and dvds and so on, but i can't imagine such a thing lasting.
[1:44] <Elly> what about SneakerNet?
[1:44] <linyos> or being used by the masses.
[1:44] <Elly> I'm still of the opinion
[1:44] <Elly> that the government lacks the ability
[1:45] <linyos> lots of work passing around dvds.
[1:45] <Elly> not to say the authority
[1:45] <Elly> but anyway
[1:45] <Elly> it's food time
[1:45] <Elly> ta!
[1:45] <linyos> later.
[1:53] <linyos> anyway, i could imagine the state of the art in operating systems / modern software being made intimately dependent on networking in a new way.
[1:54] <linyos> so any regime that wants to lock down networking would have to design its own software and cope with compatibility problems with the rest of the world.
[1:54] <linyos> that might provide some deterrence in a few marginal cases.
[1:56] <linyos> maybe that is not possible, perhaps it will always be easy to mandate monitoring. i don't know. it is certainly the best hope.
[1:58] <linyos> (TCPA would let you watch monitors)
[2:11] <Elly> you have to take into account, though, that power corrupts :)
[2:11] * Bombe (n=bombe@p5483E99C.dip.t-dialin.net) Quit (Read error: 110 (Connection timed out))
[2:11] * QshelTier (n=bombe@p5483C81F.dip.t-dialin.net) has joined #freenet
[2:13] * sanity (n=ian@81-178-94-218.dsl.pipex.com) Quit ()
[2:27] * Hory (n=Miranda@82.78.27.85) Quit ("http://CyberLore.net - Recommendations on the best games, freeware and websites.")
[2:31] * QshelTier is now known as Bombe
[2:37] * gathers (n=gathers@c-8e40e255.434-1-64736c11.cust.bredbandsbolaget.se) Quit ("L??mnar")
[3:10] * Romster2 (n=romster@tor/session/x-a8166f6ba71af843) has joined #freenet
[3:16] * Romster2 is now known as Romster
[4:03] * NullAcht15 (n=NullAcht@dslb-082-083-205-144.pools.arcor-ip.net) Quit (Remote closed the connection)
[4:12] * sandos_ (n=sandos@tor/session/x-ac461066a20479e3) has joined #freenet
[4:13] * hobx (n=hobx@c83-248-105-188.bredband.comhem.se) Quit (zelazny.freenode.net irc.freenode.net)
[4:13] * TheSeeker (i=Fridlekh@67-40-225-97.phnx.qwest.net) Quit (zelazny.freenode.net irc.freenode.net)
[4:13] * sandos (n=sandos@tor/session/x-247cec9d5d2ec0be) Quit (zelazny.freenode.net irc.freenode.net)
[4:13] * hobx_ (n=ag@ankh.math.chalmers.se) Quit (zelazny.freenode.net irc.freenode.net)
[4:13] * Hadaka (i=naked@naked.iki.fi) Quit (zelazny.freenode.net irc.freenode.net)
[4:13] * pwk_ (n=ben@dsl-62-3-71-168.zen.co.uk) Quit (zelazny.freenode.net irc.freenode.net)
[4:13] * nextime (n=nextime@213-140-6-96.ip.fastwebnet.it) Quit (zelazny.freenode.net irc.freenode.net)
[4:13] * sandos_ is now known as sandos
[4:16] * hobx (n=hobx@c83-248-105-188.bredband.comhem.se) has joined #freenet
[4:16] * TheSeeker (i=Fridlekh@67-40-225-97.phnx.qwest.net) has joined #freenet
[4:16] * hobx_ (n=ag@ankh.math.chalmers.se) has joined #freenet
[4:16] * Hadaka (i=naked@naked.iki.fi) has joined #freenet
[4:16] * nextime (n=nextime@213-140-6-96.ip.fastwebnet.it) has joined #freenet
[4:16] * pwk_ (n=ben@dsl-62-3-71-168.zen.co.uk) has joined #freenet
[4:20] * pwk_ (n=ben@dsl-62-3-71-168.zen.co.uk) Quit (zelazny.freenode.net irc.freenode.net)
[4:20] * hobx_ (n=ag@ankh.math.chalmers.se) Quit (zelazny.freenode.net irc.freenode.net)
[4:20] * TheSeeker (i=Fridlekh@67-40-225-97.phnx.qwest.net) Quit (zelazny.freenode.net irc.freenode.net)
[4:20] * Hadaka (i=naked@naked.iki.fi) Quit (zelazny.freenode.net irc.freenode.net)
[4:20] * hobx (n=hobx@c83-248-105-188.bredband.comhem.se) Quit (zelazny.freenode.net irc.freenode.net)
[4:20] * nextime (n=nextime@213-140-6-96.ip.fastwebnet.it) Quit (zelazny.freenode.net irc.freenode.net)
[4:20] * sandos (n=sandos@tor/session/x-ac461066a20479e3) Quit (zelazny.freenode.net irc.freenode.net)
[4:20] * MrNaughty (n=mrnaught@S0106001125372935.ed.shawcable.net) Quit (zelazny.freenode.net irc.freenode.net)
[4:20] * mazzanet (n=irc@unaffiliated/mazzanet) Quit (zelazny.freenode.net irc.freenode.net)
[4:20] * PazO (i=pazo@fuck.the.gov.and.legalize.piracy.nl) Quit (zelazny.freenode.net irc.freenode.net)
[4:20] * locksy (n=nlocksy@mrtg.sisgroup.com.au) Quit (zelazny.freenode.net irc.freenode.net)
[4:36] -dmwaters- {global notice} Hi all! one of our sponsors is doing some router maintenence, this is scheduled maintenence. In the process, I'm going to upgrade one of the main rotation servers. this won't effect anyone, since the server has already dropped it's users. Thank you for your patience, and thank you for using freenode!
[4:38] * pwk_ (n=ben@dsl-62-3-71-168.zen.co.uk) has joined #freenet
[4:38] * sandos (n=sandos@tor/session/x-ac461066a20479e3) has joined #freenet
[4:38] * MrNaughty (n=mrnaught@S0106001125372935.ed.shawcable.net) has joined #freenet
[4:38] * mazzanet (n=irc@unaffiliated/mazzanet) has joined #freenet
[4:38] * PazO (i=pazo@fuck.the.gov.and.legalize.piracy.nl) has joined #freenet
[4:38] * locksy (n=nlocksy@mrtg.sisgroup.com.au) has joined #freenet
[4:45] * hobx_ (n=ag@129.16.168.123) has joined #freenet
[4:45] * hobx (n=hobx@c83-248-105-188.bredband.comhem.se) has joined #freenet
[4:45] * TheSeeker (i=Fridlekh@67-40-225-97.phnx.qwest.net) has joined #freenet
[4:45] * Hadaka (i=naked@naked.iki.fi) has joined #freenet
[4:45] * nextime (n=nextime@213-140-6-96.ip.fastwebnet.it) has joined #freenet
[4:46] * Naked (i=naked@naked.iki.fi) has joined #freenet
[4:51] * Hadaka (i=naked@naked.iki.fi) Quit (Connection refused)
[4:51] * Naked is now known as Hadaka
[4:56] * Hadaka (i=naked@naked.iki.fi) Quit (zelazny.freenode.net irc.freenode.net)
[4:56] * TheSeeker (i=Fridlekh@67-40-225-97.phnx.qwest.net) Quit (zelazny.freenode.net irc.freenode.net)
[4:56] * hobx_ (n=ag@129.16.168.123) Quit (zelazny.freenode.net irc.freenode.net)
[4:56] * hobx (n=hobx@c83-248-105-188.bredband.comhem.se) Quit (zelazny.freenode.net irc.freenode.net)
[4:56] * nextime (n=nextime@213-140-6-96.ip.fastwebnet.it) Quit (zelazny.freenode.net irc.freenode.net)
[4:56] * hobx (n=hobx@c83-248-105-188.bredband.comhem.se) has joined #freenet
[4:56] * hobx_ (n=ag@ankh.math.chalmers.se) has joined #freenet
[4:56] * TheSeeker (i=Fridlekh@67-40-225-97.phnx.qwest.net) has joined #freenet
[4:57] * nextime (n=nextime@213-140-6-96.ip.fastwebnet.it) has joined #freenet
[5:01] * Bombe (n=bombe@p5483C81F.dip.t-dialin.net) Quit (Read error: 110 (Connection timed out))
[5:01] * Naked (i=naked@naked.iki.fi) has joined #freenet
[5:01] * Naked is now known as Hadaka
[5:02] * Bombe (n=bombe@p5483CEF8.dip.t-dialin.net) has joined #freenet
[6:31] * smains (n=smains@tor/session/x-293dbde9fe355748) has joined #freenet
[6:33] * smains (n=smains@tor/session/x-293dbde9fe355748) Quit (Remote closed the connection)
[6:41] * smains (n=smains@tor/session/x-dd6c522c678d2530) has joined #freenet
[7:26] * Ash-Fox (i=UNKNOWN@edj117.neoplus.adsl.tpnet.pl) has joined #FreeNET
[8:01] * QshelTier (n=bombe@p5483E25B.dip.t-dialin.net) has joined #freenet
[8:05] * Bombe (n=bombe@p5483CEF8.dip.t-dialin.net) Quit (Read error: 110 (Connection timed out))
[8:58] * QshelTier (n=bombe@p5483E25B.dip.t-dialin.net) Quit (Read error: 110 (Connection timed out))
[8:59] * QshelTier (n=bombe@p5483C839.dip.t-dialin.net) has joined #freenet
[9:04] * sanity (n=ian@81-178-94-218.dsl.pipex.com) has joined #freenet
[9:35] * sbc (n=sbc@83.72.201.52.ip.tele2adsl.dk) has joined #freenet
[10:29] * sleon|tuX (n=sleon|tu@e180018060.adsl.alicedsl.de) has joined #freenet
[10:38] * blank_ (n=smains@tor/session/x-20bb7277530ca915) has joined #freenet
[10:52] <TheSeeker> hmm, how hard would it be to make a spider that parsed all the pages it ran across and inserted a page of links for each unique word? ... so you'd have something like SSK@KEY/SearchSpider/privacy/ which would have a list of all pages containing the word "privacy"
[10:56] <linyos> the lists would be too long.
[10:56] <TheSeeker> then an external application could be written to do simple boolean searches based on multiple querries to this site... request two or more pages, then return the cross section of the liks that exist on each
[10:57] <TheSeeker> too long for what?
[10:57] <linyos> countless millions of pages would match the word "dog" for instance.
[10:57] <linyos> the files would be huge.
[10:57] <TheSeeker> There aren't that many freesites which contain the word "dog" ... :P
[10:58] <linyos> sure, on a small scale it would be acceptable.
[10:59] <linyos> p2p search is a hard problem.
[10:59] <TheSeeker> if any word is so popular that it creates a page that when compressed in a container, it's too big, then that page could just not be inserted... things like /spidersearch/and/ have no reason to exist :P
[11:00] <TheSeeker> it would be a very slow search anyway, so files as large as 1 MB per term seem acceptable to me.
[11:00] <linyos> "Results 1 - 10 of about 150,000,000 for dog"
[11:01] <TheSeeker> do you know what the planned max size before splitting is for 0.7?
[11:01] <TheSeeker> freenet will never have anywhere near the the content of the regular web
[11:01] <linyos> 32kbytes, i think.
[11:02] <linyos> nah, you'd probably want to mirror the whole web on freenet if it ever really worked.
[11:02] <linyos> i mean, why not.
[11:03] <TheSeeker> because a million personal computers don't have the kind of storage that it takes to hold the contents of the web.
[11:03] <linyos> sure they do.
[11:04] <linyos> million * 100GB (or whatever) == huge
[11:04] <TheSeeker> who allocates 100 GB for freenet?
[11:04] <TheSeeker> I can only spare 6.
[11:05] <linyos> million * 6gb is also huge.
[11:05] <TheSeeker> 33% of datastores are full of heal blocks. (assumung 50% heal)
[11:06] <TheSeeker> what are the odds of caching data as you pass it? that redundancy will cut the total network storage down further.
[11:06] <linyos> well, you see my point. it does work all right for smaller networks.
[11:07] <linyos> yeah.
[11:07] <linyos> but you'd still have room for billions of pages.
[11:07] <TheSeeker> as long as those pages contained no images, or sounds, or other multimedia
[11:09] <TheSeeker> Alexa gathers 1.6 TB of data per day from the web...
[11:10] <TheSeeker> "After each snapshot of the Web, which takes approximately two months to complete, Alexa has gathered 4.5 Billion pages from over 16 million sites.
[11:10] <TheSeeker> "
[11:11] * sbc (n=sbc@83.72.201.52.ip.tele2adsl.dk) Quit ("Leaving")
[11:12] <linyos> anyway, downloading all the lists is very inefficient, if you wanted to write a p2p search network.
[11:13] <linyos> you want to sort the keywords from least-matches to most-matches
[11:13] <TheSeeker> you wouldn't download al the lists. just ones that are for your query.
[11:13] <linyos> i know
[11:13] <linyos> so, say you search for "foo bar baz"
[11:13] <linyos> foo: 100 hits
[11:13] <linyos> bar: 1000 hits
[11:13] <linyos> baz: 100000 hits
[11:13] <linyos> host A has the list for "foo"
[11:14] <linyos> host B has the list for "bar"
[11:14] <linyos> host C has the list for "baz"
[11:14] <linyos> the naive way is just to download each host's list and do a set intersection locally.
[11:14] <linyos> but that is very bandwidth inefficient.
[11:14] <linyos> the efficient way is:
[11:15] <linyos> you -> host A: "send your list to host B"
[11:15] <linyos> host B does a set intersection on the two lists
[11:15] <linyos> yielding a set of hits for "foo" && "bar"
[11:15] <linyos> then
[11:15] <linyos> host B sends that to host C
[11:15] <linyos> host C does another set intersection
[11:16] <linyos> and finally host C sends the final set, "foo" && "bar" && "baz", back to you.
[11:16] <linyos> the point is, the sets get a lot smaller at each phase.
[11:16] <linyos> and they start small.
[11:16] <TheSeeker> that might be better on bandwidth, but the lists are not assigned to specific hosts, they are just keys... the people hosting them have no idea that the keys are lists of freesites... I'm making an argument for a feasible means of 'search. using freenet, not some ideal non-anonymous search mechanism.
[11:17] <linyos> yeah. i'm just showing off how i thought of all that two years ago.
[11:17] <linyos> it _is_ much more efficient when you have a search query like "blahblahbuncommonkeyword dog"
[11:18] <linyos> the first word might have ten matches, but "dog" might have 100 million.
[11:18] <linyos> way better to send a set of ten matches to the "dog" index host...
[11:18] <linyos> than to download all 100 million to do the intersection yourself.
[11:19] <TheSeeker> or you never get a reply for "dog" because the spider is set up not to not insert lists of freesites for common terms.
[11:21] <TheSeeker> and, I like the idea of making the person who is doing the request perform the computationally intesive task of intersecting the data... otherwise you'd have people spamming searches with lots of common terms just to make as many other nodes as possible do a lot of work for nothing.
[11:24] <linyos> it's not actually that bad. you have a hashtable, you do N lookups.
[11:27] * MrNaught (n=mrnaught@S0106001125372935.ed.shawcable.net) has joined #freenet
[11:28] <TheSeeker> I should look up how frost works...
[11:39] <TheSeeker> I could see an application with the features you describe working like: requestor starts request, gets headers for keys related to search terms -> makes post to 'search' board with keys for the terms and the expressions for how they should be matched (smallest file size = higher significance) ->
[11:40] <TheSeeker> while no replies exist with a subject indicating that the term has been matched and calculated, nodes also running this app check their stores for the key in question and if they have it, perform a cross section of the previous term and -> make a post declaring that term finished, with a key attached with the cross section results
[11:40] <TheSeeker> ... there's a trust issue here though, someone could generate fake or filtered results. There's also the issue that even if a key for a term is known to exist, it's not guaranteed to be on any node running this app.
[11:41] <linyos> you could make the nodes programmable.
[11:41] <linyos> so you could send an intersection-sort program to another node.
[11:41] <linyos> (wishful thinking...)
[11:42] <TheSeeker> sounds like wishing you had a virus to me.
[11:43] <linyos> nah, while you are using someone's bandwidth and storage space, you might as well use some of their processor cycles too, right?
[11:44] <linyos> so you could program nodes in the network to do all sorts of applications, such as searching.
[11:45] <TheSeeker> you seem to be talking about turning every node into a spider... that's just not feasible
[11:45] <linyos> (05:11:59) linyos: you -> host A: "send your list to host B"
[11:45] <linyos> (05:12:11) linyos: host B does a set intersection on the two lists
[11:46] <linyos> imagine writing a program for host A
[11:46] <linyos> you know: send_file(list_key, host_b) ...
[11:46] <TheSeeker> "Host A, log all your keystrokes and upload it to the following key, kthxbye"
[11:46] <linyos> and then writing a program for host B
[11:46] <linyos> they'd be sandboxed obviously, run on a virtual machine.
[11:48] <linyos> then you could run an infinite number of p2p protocols over the same darknet.
[11:48] <linyos> limited only by resources and the users' creativity.
[11:49] <linyos> you could even implement freenet routing as one of those programs.
[11:49] <TheSeeker> I use freenet, and agree to share my badwidth and hdd space. I'm not agreeing to allowing arbitrary code to be run on my computer.
[11:50] <linyos> why would you make the distinction?
[11:51] <TheSeeker> I don't care if it's "sandboxed" ... I don't want to be forced into having my CPU bogged down by a dozen other people telling ym computer to start breaking encryption, or doing seti crap.
[11:51] <TheSeeker> my
[11:51] <linyos> obviously the programs would have no interaction with your operating system. they'd be written exclusively for the freenet virtual machine.
[11:51] <linyos> so they'd have the lowest priority.
[11:51] <TheSeeker> shared storage and bandwidth is needed for anonymity on the network. shared CPU is not.
[11:52] <linyos> i agree.
[11:52] <TheSeeker> the only reason to use freenet is anonymity.
[11:53] <TheSeeker> I'm not giving away my computer's resources for nothing in return.
[11:53] <linyos> you give away your bandwidth and disk space.
[11:54] <TheSeeker> no, I share it. I'm constantly uploading and downloading from freenet
[11:55] <linyos> well, programmable nodes would work the same way. you program somebody's node to do p2p searches, and in return, somebody else programs your node to do his favorite p2p application.
[11:55] <linyos> the trade is mutually advantageous.
[11:55] <TheSeeker> The only way I would accept that, is if I could only have my node be programmed to do something I've programmed someone else's node to do.
[11:56] <linyos> i guess there could be an opt-in.
[11:56] <TheSeeker> and in that case, there's no reason to have any of that running over freenet
[11:57] <linyos> the reason would be that you have the darknet already, why not use it for as many applications as possible? such as (more efficient) searching.
[11:57] <linyos> i mean, the hard part is setting up all the links to your friends.
[11:58] <TheSeeker> because I have no iterest in those other applications, and if I'm giving up bandwidth and HDD space, I want it going to freenet, not partly to freenet, and partly to a dozen other things I'm not interested in
[11:58] <TheSeeker> I have little to no intention of using the darknet version of 0.7
[11:58] <linyos> oh, ok.
[11:58] * QshelTier (n=bombe@p5483C839.dip.t-dialin.net) Quit (Read error: 110 (Connection timed out))
[11:58] <linyos> that is fair enough.
[11:59] * QshelTier (n=bombe@p5483E502.dip.t-dialin.net) has joined #freenet
[12:01] <TheSeeker> and since the darknet and the open net will be the same, with the exception of connections, I don't really see any of your suggested features being implemented :P
[12:01] <TheSeeker> or maybe it's just that I think they're horrible and would never want to see them implemented...
[12:04] <TheSeeker> good idea: being able to run arbitrary internet applications over a darknet. bad idea: being able to force other computers to run arbitrary internet applications because they are part of the darknet
[12:05] <linyos> trading computer resources makes a lot of sense in theory.
[12:05] <linyos> whether it reduces to practice in this case, i doubt, yeah.
[12:08] <linyos> oh, you can tunnel IP over a darknet already, in an end-to-end way. my point is that end-to-end communication is often inferior to smarter protocols where the intermediate hosts play a more active role.
[12:08] <linyos> and that in theory requires them to be programmable.
[12:09] <linyos> sort of the cliched difference between peer-to-peer and client-server.
[12:22] <TheSeeker> my node status page says I have about 5600 more keys in my datastore than I have files in my datastore... :/ I had hoped that bug was fixed...
[12:33] * gathers (n=gathers@c-9746e255.434-1-64736c11.cust.bredbandsbolaget.se) has joined #freenet
[12:45] * hobx (n=hobx@c83-248-105-188.bredband.comhem.se) Quit ("Client exiting")
[13:00] * rah (n=rah@cpc1-nott8-4-1-cust246.nott.cable.ntl.com) has joined #freenet
[13:01] * rah (n=rah@cpc1-nott8-4-1-cust246.nott.cable.ntl.com) Quit (Client Quit)
[13:02] * rah (n=rah@cpc1-nott8-4-1-cust246.nott.cable.ntl.com) has joined #freenet
[13:06] * toad_ (i=toad@pdpc/supporter/active/toad-with-underline) has joined #freenet
[13:06] * ChanServ sets mode +o toad_
[13:12] * hobx_ (n=ag@ankh.math.chalmers.se) Quit (Remote closed the connection)
[13:16] * linyos (n=linyos@109-17.73-24.tampabay.res.rr.com) has left #freenet
[13:35] * QshelTier is now known as Bombe
[13:35] * Bombe (n=bombe@p5483E502.dip.t-dialin.net) Quit ("Leaving")
[13:58] * hobx_ (n=ag@ankh.math.chalmers.se) has joined #freenet
[14:05] * NullAcht15 (n=NullAcht@dslb-082-083-205-144.pools.arcor-ip.net) has joined #freenet
[14:07] * greycat (i=rfc1413@wooledge.org) has joined #freenet
[14:09] * Ash-Fox (i=UNKNOWN@edj117.neoplus.adsl.tpnet.pl) Quit ()
[14:19] * Ash-Fox (i=UNKNOWN@edj117.neoplus.adsl.tpnet.pl) has joined #FreeNET
[14:21] * FallingBuzzard (n=FallingB@c-24-12-230-255.hsd1.il.comcast.net) has joined #freenet
[14:33] * unknown_ (n=NullAcht@dslb-082-083-229-243.pools.arcor-ip.net) has joined #freenet
[14:37] * rah_ (n=rah@cpc1-nott8-4-1-cust246.nott.cable.ntl.com) has joined #freenet
[14:37] * rah_ (n=rah@cpc1-nott8-4-1-cust246.nott.cable.ntl.com) Quit (Read error: 104 (Connection reset by peer))
[14:49] * NullAcht15 (n=NullAcht@dslb-082-083-205-144.pools.arcor-ip.net) Quit (Read error: 110 (Connection timed out))
[15:06] * MrNaught (n=mrnaught@S0106001125372935.ed.shawcable.net) Quit ("\(^_^)/'")
[15:22] * sandos_ (n=sandos@83.233.112.133) has joined #freenet
[15:32] * sandos (n=sandos@tor/session/x-ac461066a20479e3) Quit (Read error: 110 (Connection timed out))
[15:32] * sandos_ is now known as sandos
[15:39] * sbc (n=sbc@83.72.201.52.ip.tele2adsl.dk) has joined #freenet
[16:00] * thong (n=thong@warp-32.cl.cam.ac.uk) Quit ("Leaving")
[16:43] * mazzanet (n=irc@unaffiliated/mazzanet) Quit (Read error: 110 (Connection timed out))
[17:10] * toad_ (i=toad@pdpc/supporter/active/toad-with-underline) Quit (Remote closed the connection)
[17:17] * mazzanet (n=irc@unaffiliated/mazzanet) has joined #freenet
[17:54] * sanity (n=ian@81-178-94-218.dsl.pipex.com) Quit ()
[18:00] * sanity (n=ian@81-178-94-218.dsl.pipex.com) has joined #freenet
[18:10] * blank_ (n=smains@tor/session/x-20bb7277530ca915) Quit ("Leaving")
[18:12] * toad_ (i=toad@pdpc/supporter/active/toad-with-underline) has joined #freenet
[18:13] * ChanServ sets mode +o toad_
[18:17] <toad_> hi folks
[18:21] <toad_> http://127.0.0.1:8888/SSK@p4vgLyAEFsmDQQCDQhN~0-c2iG4PAgM,RutkWkbmfmuqVvUJ-1-DLg/ChechnyaFilmFestival//
[18:22] <trams> hey toad.
[18:22] <trams> sup
[18:22] <toad_> not a lot
[18:22] <toad_> haven't had time to do much work this week
[18:22] <toad_> but i'm back now
[18:22] <trams> ::/
[18:23] <trams> where have you been?
[18:23] <toad_> went to the Mass Lobby For Trade Justice on tuesday
[18:23] <toad_> and met my MP, who was very nice
[18:24] <toad_> just had lunch and watched the first video from the above freesite
[18:26] <toad_> somebody once said "when the news is good, we don't believe it, because it is so difficult to believe... but when the news is bad, we believe it because it is so easy to believe". it is disturbingly easy to believe the conspiracy theory perpetuated by that film. especially as the evidence presented is pretty solid, as these things go...
[18:26] <toad_> probably it is easy to believe mainly because the country involved is russia
[18:29] <nextgens> hi
[18:29] <toad_> anyway it's probably worth watching
[18:29] <toad_> nextgens: hi
[18:30] <toad_> nextgens: genuinely politically interesting content on freenet: http://127.0.0.1:8888/SSK@p4vgLyAEFsmDQQCDQhN~0-c2iG4PAgM,RutkWkbmfmuqVvUJ-1-DLg/ChechnyaFilmFestival//
[18:31] <toad_> nextgens: how's things?
[18:31] <nextgens> not to bad ^-^
[18:32] <nextgens> I have to resume studying ^-^ ... and as usual it's a pain for me
[18:32] <toad_> :(
[18:32] <nextgens> toad_> what do you think about http://lurker.sourceforge.net/ ?
[18:33] <nextgens> as a mailling list archiver/browser/searchengine
[18:33] <toad_> looks nice
[18:34] <toad_> we have many thousands of messages to import from various sources
[18:34] <toad_> if you want to set one up, it'd be really nice...
[18:34] <toad_> there may be some complications with importing stuff
[18:35] <nextgens> ok, maybe I will this WE
[18:40] <nextgens> toad_> I was aware that they are some Chinese dissidents on Freenet ... but not Russians^wChechnyans one :)
[18:41] <toad_> well, if the documentary is true...
[18:42] <toad_> ... then russia is rather closer to china than the west customarily thinks
[18:45] <toad_> i hope we can get some serious i18n by the time 1.0 comes out
[18:46] <toad_> 0.7 should solve the content charset problem anyway
[18:47] <nextgens> toad_> I can do the french translation of fproxy if needed
[18:47] <nextgens> but I think that the french user community is quite active :)
[18:48] <nextgens> there is a lot of french documentation on "how to connect to freenet"
[18:48] <toad_> hmmm
[18:48] <toad_> thetrainline.com thinks that bristol is in somerset...
[18:48] <toad_> i hope this doesn't result in my tickets getting lost in the post...
[18:48] <toad_> nextgens: indeed, it's not a problem for iso-8859-1 languages
[18:49] * mazzanet (n=irc@unaffiliated/mazzanet) Quit (Read error: 104 (Connection reset by peer))
[18:49] * mazzanet (n=irc@unaffiliated/mazzanet) has joined #freenet
[18:52] <toad_> nextgens: technically, there are two problems. one is that the anon filter needs to know the charset IN ADVANCE
[18:52] <toad_> the second is that we need to support i18n text substitution on the web interface, and we need to get some actual humans to do translations for us
[18:53] <toad_> the answer to the first is to detect it during insert, and give the FCP insert code enough information to detect the charset, so for example, we don't pass ZIP manifests directly in, we let the node construct them from the individual files (this has numerous other advantages)
[18:53] <toad_> and then we either encode the charset into the MIME metadata, or we just convert everything to UTF8
[18:54] <toad_> the second point is pretty easy
[18:54] <nextgens> agreed
[18:55] <toad_> nextgens: i'd appreciate your input on updating mechanisms, in a minute...
[18:55] <toad_> brb
[18:55] <nextgens> what about setting up a configuration option for 1) ?
[18:57] * sbc (n=sbc@83.72.201.52.ip.tele2adsl.dk) Quit ("Leaving")
[19:01] <toad_> nextgens: hmm?
[19:01] <toad_> that's possible, but i'm not sure it's worthwhile
[19:10] <nextgens> well, usually people are only using their current locale; aren't they ?
[19:12] <nextgens> I mean, if I was using a special charset, I would probably want my browser/fproxy to display things using it
[19:12] <nextgens> even if sometimes I have to change the setting in order to browse a freesite with a different charset
[19:13] <toad_> usually on the web people specify the charset in META HTTP-EQUIV name="content-type" content="text/html; charset=<whatever>"
[19:13] <toad_> in the HTML
[19:13] <toad_> we can't do that on freenet
[19:13] <toad_> because the anonymity filter needs to know with absolute certainty what the charset is before it filters it
[19:13] <nextgens> they should specify it ^-^
[19:13] <toad_> e.g. UTF16 can go straight past the filter, unless we explicitly check it
[19:13] <toad_> they don't know how to specify it, in general
[19:14] <toad_> this is an unnecessary barrier
[19:14] <toad_> anyway their HTML tools *DO* specify it
[19:14] <toad_> we ought to pick it up
[19:14] <toad_> but on insert, so the retrieve code is simple
[19:15] <nextgens> [18:52] < toad_> | nextgens: i'd appreciate your input on updating mechanisms, in a minute...
[19:15] <nextgens> what were you planning to do ? :)
[19:15] <nextgens> some kind of "svn over freenet pooling" thing ?
[19:15] <toad_> Subject: [Tech] High-latency functionality and the download manager backend
[19:15] <toad_> in 0.7 was Re: [freenet-dev] Background downloading of large
[19:15] <toad_> container
[19:16] <toad_> no not that one
[19:16] * nextgens is checking his mails
[19:16] <toad_> although that one is relevant too
[19:16] <toad_> originally this one:
[19:16] <toad_> Subject: [Tech] Unified updating system
[19:16] <nextgens> fetchmail: POP3< +OK 43 messages
[19:16] <toad_> :|
[19:16] <toad_> you have 1 real message and 42 spams, unfortunately i have forgotten which one is the real message...
[19:17] <nextgens> :)
[19:19] <nextgens> deux informations pour MI41
[19:19] <nextgens> pas de TP le lundi 7 novembre
[19:19] <nextgens> pas de TD le mardi 8 novembre (il y a en revanche TP)
[19:19] <nextgens> NL
[19:19] <nextgens> raah
[19:19] <nextgens> sorry
[19:23] <toad_> gah
[19:23] <nextgens> toad_> I don't understand why we should get rid of double /
[19:23] <toad_> the wiki is down, almost certainly because of sf's migration from mysql 3.23 to mysql 4.1
[19:24] <nextgens> erf
[19:24] <toad_> nextgens: because it's perverse, unix systems assume // == /, and no major uri scheme uses double slashes except at the beginning
[19:24] <nextgens> it will break with the password() thing :(
[19:24] <nextgens> true ...
[19:24] <nextgens> some webrowsers are even sanitizing it :(
[19:25] <toad_> well, if we want to get rid of // before 1.0, we really have to do it now
[19:25] <toad_> but that's not the key issue here imho
[19:26] <toad_> the key issue is that taking the responsibility for updating mechanisms out of the user's and client's hands is a good thing
[19:26] <toad_> especially as we will undoubtedly have many new authors after 0.7 comes out
[19:26] <toad_> hence USKs
[19:27] <toad_> also, USKs seem to gel reasonably well with the download manager
[19:27] <toad_> as is explained in the other mail
[19:27] * nextgens is reading the whole thread
[19:27] <nextgens> I'm a bit slow at reading technical english :$
[19:29] <nextgens> well, I agree that if something needs to be done, it should be ASAP
[19:30] <nextgens> before you release an 'alpha' with splitfiles ;)
[19:30] <nextgens> but the // thing wasn't shocking me ...
[19:31] <nextgens> ...
[19:32] <nextgens> but it could be usefull if we do want "existing" software to handle freenet:// URIs, definitely this could help
[19:33] <toad_> not just that, http://127.0.0.1:8888/ URIs are affected with some software, and it's just plain ugly imho
[19:33] <toad_> i don't know if it's illegal in the uri spec, it might be
[19:33] <nextgens> and I do agree that the more thing we do in the node, the less clients complicated will be
[19:33] <nextgens> erf
[19:34] <toad_> hmmm?
[19:34] <nextgens> not sure my latest sentence is understandable by an english person :$
[19:34] <toad_> it is by me :)
[20:05] <FallingBuzzard> What is an USK?
[20:08] <toad_> my new proposed updating scheme
[20:08] <toad_> Subject: [Tech] Unified updating system
[20:08] <FallingBuzzard> Does it replace TUK's
[20:09] <toad_> initially it'd essentially be an automated way of doing editions, but later we could make it use TUKs, DBRs, passive requests and whatever else
[20:09] <FallingBuzzard> k
[20:09] <toad_> the key point is the user no longer has to worry about DBR versus edition etc
[20:09] * FallingBuzzard goes to read up on USK's
[20:18] * TheSeeker (i=Fridlekh@67-40-225-97.phnx.qwest.net) Quit (Read error: 104 (Connection reset by peer))
[20:41] * louisb (n=lou@CPE0013f70131cb-CM0012c91006bc.cpe.net.cable.rogers.com) has joined #freenet
[21:07] * Ribs (n=ribs@riblet.plus.com) has joined #freenet
[21:22] * Rom|Bed (n=romster@tor/session/x-cd09f48e17d1e838) has joined #freenet
[21:22] * Romster (n=romster@tor/session/x-a8166f6ba71af843) Quit (Nick collision from services.)
[21:22] * Rom|Bed is now known as Romster
[21:30] <Elly> holy crap
[21:30] <Elly> freenet is taking my router down hard
[21:33] <toad_> hi Elly
[21:35] <Elly> hi toad_
[21:58] * Hory (n=Miranda@82.78.27.85) has joined #FreeNet
[22:08] * smains (n=smains@tor/session/x-dd6c522c678d2530) Quit (Remote closed the connection)
[22:35] * linyos (n=linyos@109-17.73-24.tampabay.res.rr.com) has joined #freenet
[22:37] * meriad (n=parsec@snoid.gfed.net) has joined #freenet
[22:37] <meriad> when i run tspc Error is 1: TSP_ERROR
[22:37] * Ribs (n=ribs@riblet.plus.com) Quit ("Leaving")
[22:38] * meriad (n=parsec@snoid.gfed.net) has left #freenet
[22:39] * FallingBuzzard (n=FallingB@c-24-12-230-255.hsd1.il.comcast.net) has left #freenet
[22:46] * greycat (i=rfc1413@wooledge.org) Quit ("This time the bullet cold rocked ya / A yellow ribbon instead of a swastika")
[22:54] * hobx (n=hobx@c83-248-105-188.bredband.comhem.se) has joined #freenet
[23:03] * nextgens_ (n=nextgens@ADijon-151-1-28-147.w83-194.abo.wanadoo.fr) has joined #freenet
[23:10] * nextgens (n=nextgens@ADijon-151-1-28-52.w83-194.abo.wanadoo.fr) Quit (Read error: 110 (Connection timed out))
[23:23] * sleon|tuX (n=sleon|tu@e180018060.adsl.alicedsl.de) Quit (Read error: 110 (Connection timed out))
Irc logs of #freenet : 2008 2007 2006 2005
These logs were automatically created by FreenetLogBot on chat.freenode.net using the Java IRC LogBot.