Timestamps are in GMT/BST.
[1:37] * kg (helstrom@ip68-7-209-131.sd.sd.cox.net) has joined #freenet
[2:36] * kg (helstrom@ip68-7-209-131.sd.sd.cox.net) Quit (Read error: 104 (Connection reset by peer))
[3:25] * Hory (~Miranda@81.196.25.110) has joined #FreeNet
[3:35] * tyesinclair (~tyesincla@170.252.64.1) has joined #freenet
[3:35] * tyesinclair (~tyesincla@170.252.64.1) has left #freenet
[4:38] * stazz (~ville@a80-186-82-63.elisa-laajakaista.fi) Quit ("leaving")
[5:04] <i2p_iip> <Newsbyte> toad, you there?
[5:05] * Ash-Fox (Hal-9000@aas176.neoplus.adsl.tpnet.pl) Quit (Client Quit)
[5:08] * nextgens (~nextgens@d80-170-121-177.cust.tele2.fr) has joined #freenet
[5:09] <nextgens> hi
[5:11] <ShaunMacPherson> hi
[5:27] * pittaman (~mistery_b@d51A410CA.kabel.telenet.be) has joined #freenet
[5:27] * nextgens (~nextgens@d80-170-121-177.cust.tele2.fr) has left #freenet
[5:36] * Gunnar (~gunnar@bkkb-gw.bitcon.no) has joined #freenet
[6:11] * Gunnar (~gunnar@bkkb-gw.bitcon.no) Quit (Remote closed the connection)
[6:31] * 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!")
[6:37] * Hory (~Miranda@81.196.25.110) has joined #FreeNet
[6:49] * Zorix (Brandon@fl-69-69-7-231.dyn.sprint-hsd.net) Quit (Read error: 110 (Connection timed out))
[7:18] * MrNaughty (MrNaughty@d199-126-253-66.abhsia.telus.net) Quit (niven.freenode.net irc.freenode.net)
[7:18] * bluephile (bluephile@69-160-205-193.clvdoh.adelphia.net) Quit (niven.freenode.net irc.freenode.net)
[7:19] * MrNaughty (MrNaughty@d199-126-253-66.abhsia.telus.net) has joined #freenet
[7:19] * bluephile (bluephile@69-160-205-193.clvdoh.adelphia.net) has joined #freenet
[7:21] * mazzanet (~mazzanet@mazzanet.user) Quit (niven.freenode.net irc.freenode.net)
[7:21] * mazzanet (~mazzanet@mazzanet.user) has joined #freenet
[7:27] * guido^pe (~unknown@dsl-213-023-246-084.arcor-ip.net) has joined #freenet
[7:39] * toad_ (toad@82-32-16-91.cable.ubr03.azte.blueyonder.co.uk) Quit ("using sirc version 2.211+KSIRC/1.3.10")
[7:41] * toad_ (toad@82-32-16-91.cable.ubr03.azte.blueyonder.co.uk) has joined #freenet
[7:54] * Hory (~Miranda@81.196.25.110) Quit (Read error: 104 (Connection reset by peer))
[7:58] * Hory (~Miranda@81.196.25.110) has joined #FreeNet
[8:12] * Zorix (Brandon@fl-69-69-7-231.dyn.sprint-hsd.net) has joined #freenet
[8:22] <i2p_iip> <Newsbyte> toad?
[8:22] <i2p_iip> <Newsbyte> toad, you there?
[8:28] <toad_> yes
[8:28] <toad_> brb
[8:32] <toad_> hi Newsbyte
[8:32] * FallingBuzzard (~srademach@207.152.112.129) has joined #freenet
[8:44] <toad_> bbiab
[8:44] <toad_> ~ 30 minutes
[8:44] <toad_> might be 60
[8:55] * nextime (~nextime@danex.i-m-c.it) Quit ("nextime has no reason")
[8:57] * nextime (~nextime@danex.i-m-c.it) has joined #freenet
[8:58] * pupok_ (~r00t@81-178-112-130.dsl.pipex.com) has joined #freenet
[9:03] <i2p_iip> <Newsbyte> well, toad, you said unstable snapshot...will that interfere with my stable build? And how much juice does my puter need to run the simulation?
[9:05] * sanity (~ian@81-178-112-130.dsl.pipex.com) has joined #freenet
[9:08] <i2p_iip> <Newsbyte> ah, sanity.Saw my latest post on tech?
[9:09] <i2p_iip> <Newsbyte> you always seem to take things so personally...
[9:18] <sanity> if you don't want me to take things personally, then don't make them about me.
[9:25] <i2p_iip> <Newsbyte> well, there you go, you always think they are about you. If I say current NGR sucks, then, whether you agree with it or not, it's not about you, it's about NGR
[9:26] <hobx_> sanity: I have replicated your results with the link distribution
[9:27] <hobx_> It does not give the exactly right distribution by my count, but the performance actually is better than for networks that do have the right distribution...
[9:28] <i2p_iip> <Newsbyte> the same thing for the 'reluctance'; who was saying it's all your fault? Even if it were reluctance, it could well be, that toad had some valid reservations in making it open in details, for instance. But no, you asume I refer to a sinister plot on your part to refuse any openess.
[9:33] <sanity> hobx: cool
[9:35] <sanity> newsbyte: you have a selective memory, but i really can't be bothered to discuss this
[9:36] <sanity> hobx: how could the performance be better? i thought the "right" distribution was supposed to be optimal?
[9:37] <hobx_> yeah, that is not what I would have predicted
[9:37] <hobx_> it could be a problem with my simulator.
[9:38] <hobx_> But remember that the results are asymptotical. For any given n something that is O(n) and be smaller than something that is O(log n) depending on the constants.
[9:39] <sanity> hobx: anyway, this looks like it could be applied to creating an even simpler type of DHT than exists now
[9:40] <sanity> i suppose the next stage would be to compare Freenet's approach where data is de-coupled from the node it is on with a caching DHT approach where it isn't
[9:40] <hobx_> it has kind of already been proposed: http://www-db.stanford.edu/~manku/papers/03usits-symphony.pdf
[9:41] <hobx_> Only his method to to create the links was simply to generate positions with the right distribution, and then contact the node at that position.
[9:41] <toad_> hi
[9:41] <sanity> hobx: i think this way would be preferable
[9:41] <sanity> hobx: in that it is self-healing and adaptive
[9:41] <sanity> toad: howdy
[9:42] <toad_> so has there been a paper we somehow missed on how LRU is a near optimal connection strategy for DHTs?
[9:42] <hobx_> sanity: I agree. I actually proposed to the author a different method, where you use a weighted random walk to find nodes with the correct distribution, but he didn't seem interested.
[9:43] <hobx_> Here are what I see http://www.math.chalmers.se/~ossa/temp/datainv.png
[9:43] <sanity> hobx: NIH I suppose
[9:43] <sanity> hobx: perhaps we can get a paper out of it
[9:43] <sanity> hobx: i suspect several papers may emerge from our efforts to piece together some theoretical results around freenet
[9:44] <hobx_> sanity: If I can prove that it works, than defenitely.
[9:44] <sanity> hobx: any progress on that front?
[9:44] <hobx_> Maybe.
[9:44] <sanity> what is explicit and MH sampling?
[9:44] <hobx_> I'm begining to see what won't work anyways :-)
[9:44] * pupok_ (~r00t@81-178-112-130.dsl.pipex.com) Quit (Read error: 54 (Connection reset by peer))
[9:45] <hobx_> explicit sampling is what he proposed. Simply draw a value from the correct distribution, and route to the closest existing node.
[9:45] * pupok_ (~r00t@81-178-112-130.dsl.pipex.com) has joined #freenet
[9:45] <sanity> hobx: hmz, so LRU actually does better than his approach according to your sim
[9:46] <hobx_> MH is my idea, using a random walk with an algorithm called Metropolis-Hastings weighing the choices so that the distribution of nodes is passes is correct.
[9:47] <sanity> well, MH is closest to a straight line
[9:47] <hobx_> yeah. The deviation of the explicit pulling is surprising though
[9:47] * sanity (~ian@81-178-112-130.dsl.pipex.com) Quit (Read error: 54 (Connection reset by peer))
[9:47] * pupok_ (~r00t@81-178-112-130.dsl.pipex.com) Quit (Read error: 104 (Connection reset by peer))
[9:48] * sanity (~ian@81-178-112-130.dsl.pipex.com) has joined #freenet
[9:48] <sanity> FUCK!!!
[9:48] <sanity> my crappy fucking connection keeps crapping out, and i don't know why
[9:48] <sanity> anyway
[9:48] <hobx_> we noticed
[9:49] * pupok_ (~r00t@81-178-112-130.dsl.pipex.com) has joined #freenet
[9:49] <sanity> hobx: can you calculate the probability that a given node won't have a link closer to the destination than another given node relative to the latter node's distance from the destination?
[9:50] <hobx_> ouch
[9:50] <hobx_> what a sentence :-)
[9:50] <sanity> ok
[9:50] <sanity> node A, has a link to us and a request
[9:50] <sanity> we are node B
[9:51] <hobx_> ok
[9:51] <sanity> the message is intended for node C
[9:51] <sanity> probability that we get the message is the probability that A doesn't have a link that is closer to C than we are
[9:52] <sanity> [possible caveat: the fact that A has the message is not necessarily independent from the location of C because someone may have routed it to C]
[9:52] <hobx_> sorry phone. back now.
[9:53] <sanity> oops, fucked that up, again : [possible caveat: the fact that A has the message is not necessarily independent from the location of C because someone may have routed it to A]
[9:53] <hobx_> exactly
[9:53] <hobx_> the last part is critical
[9:53] <sanity> so we need to account for the length of the path so-far
[9:54] <hobx_> at the first step, the query distribution is uniform
[9:54] <toad_> hmmm
[9:54] <sanity> hobx: of course
[9:54] <sanity> hobx: at the second step... its not :-)
[9:54] <hobx_> as the number of steps increases, it comes closer and closer to being concentrated around the nodes own value
[9:54] <sanity> hobx: QED!
[9:55] * toad_ has to reboot because of suspicions wrt cooling... fortunately the logs are up, so i can read backlog
[9:55] * toad_ (toad@82-32-16-91.cable.ubr03.azte.blueyonder.co.uk) Quit ("using sirc version 2.211+KSIRC/1.3.10")
[9:55] <sanity> hobx: yes, but not ever request received by a node will have the same HTL
[9:56] <hobx_> of course
[9:56] <sanity> so we want to prove that the destinations of the incoming requests a node will receive is inversely proportional to the eventual destinations of those requests
[9:57] <hobx_> clearly the prob that it is on its first hop is > 0, but the prob that it is on hop greater than size of network is 0.
[9:57] <hobx_> So that probability varies instead.
[9:58] <sanity> ok, so how are the HTLs of incoming requests distributed?
[9:58] <sanity> evenly across HTLs?
[9:58] <sanity> probably should be
[9:58] <hobx_> I think the path to a nice solution is to turn the problem around, and consider it as a chain on nodes coming to the query, rather than a query coming to different nodes. Then I think it can be expressed as a markov chain, after which things are easier.
[9:59] <sanity> ok
[9:59] <hobx_> sanity: No, obviously the prob that it is on it's first hop is greater than that it is on it's 1000th.
[9:59] <sanity> ah, correct
[10:00] <sanity> i was assuming all request lengths are the same, but obviously they aren't
[10:00] <hobx_> I'll figure it out, but I'm not quite clever enough that I don't need some time. If I fail, then I have access to people who are better at this.
[10:00] * pupok_ (~r00t@81-178-112-130.dsl.pipex.com) Quit (Read error: 54 (Connection reset by peer))
[10:00] <sanity> hobx: fair enough
[10:00] * pupok_ (~r00t@81-178-112-130.dsl.pipex.com) has joined #freenet
[10:01] <i2p_iip> <Newsbyte> *I* have a selective memory? heh...well, at least you didn't lose your sense of irony...
[10:02] <sanity> newsbyte: go away, we are trying to discuss something important
[10:02] <hobx_> Anyways, it is strange that the distribution with explicit sampling is so wrong. I checked the formula I use to generate it, and that is correct, so it can't be that.
[10:03] <sanity> hobx: what exactly is explicit sampling?
[10:03] <sanity> hobx: oh, another nice thing about LRU is that nodes can always get an introduction to each-other through other nodes
[10:04] <sanity> hobx: that is very useful if you are trying to do firewall circumvention without relying on a centralized Trent
[10:04] <hobx_> One explanation is that I followed the procedure of the paper I linked, creating the links as each node is connected. That means that the nodes that join early may end up with links that are far away (because there are few nodes so the closest may be far).
[10:04] <sanity> hobx: yes, perhaps you could create the nodes first and then link them up
[10:05] <hobx_> Explicit sampling: In this case, you want to draw links at distances with distribution p(d) = 1 / ln(n) d for d in [1/ln(n), 1]
[10:06] <sanity> hobx: i wonder if LRU will compensate for imperfect sampling that can't be fixed
[10:06] <hobx_> that is the Kleinberg distribution for a one dimensional ring.
[10:06] <sanity> hobx: ie. if some of the nodes have fixed random links, and the others use LRU, will they compensate somehow
[10:06] * pupok_ (~r00t@81-178-112-130.dsl.pipex.com) Quit (Read error: 104 (Connection reset by peer))
[10:07] <hobx_> The formula for making a random number with that distribution is: Math.exp(Math.log(n) * (re.nextDouble() - 1.0)) / 2.0;
[10:09] <hobx_> So the explicit sampling is: create such a number with that formula. Add that value to your position, and route to the node that is closest to the value. Choose him as a neighbor.
[10:11] <sanity> i see
[10:12] <sanity> i suppose another advantage of LRU is that it is constantly "rebalancing"
[10:12] <hobx_> It is essentially "cheating" to get the right link distribution.
[10:12] <hobx_> Yes.
[10:14] <sanity> ok, i will brb in 15 mins
[10:15] * toad_ (toad@82-32-16-91.cable.ubr03.azte.blueyonder.co.uk) has joined #freenet
[10:15] * pupok_ (~r00t@81-178-112-130.dsl.pipex.com) has joined #freenet
[10:15] * toad_ decides to do nothing about CPU temp, which brat advises me is rather high... ~ 70degC
[10:16] <i2p_iip> <Newsbyte> well, then, I should take advantage of the fact it isn't 'discussing anything important' for 15 min, then ;-)
[10:17] <i2p_iip> <Newsbyte> I agree codestuff is very important, but in my view, things like a finance-page are pretty important in their own area too.
[10:17] <toad_> [10:03] <sanity> hobx: oh, another nice thing about LRU is that nodes can always get an introduction to each-other through other nodes
[10:17] <toad_> eh?
[10:18] * C-M|morphium (~CMmorphiu@p213.54.211.53.tisdip.tiscali.de) has joined #freenet
[10:18] <C-M|morphium> hi, is it possible to run freenet server on a linux box, without using it for surfing (just as node)
[10:19] <i2p_iip> <Newsbyte> well, toad, you said unstable snapshot...will that interfere with my stable build? And how much juice does my puter need to run the simulation?
[10:19] <i2p_iip> <Newsbyte> yes, morph
[10:19] <toad_> Newsbyte: no, you don't need to shut down your stable node or start an unstable one
[10:19] <toad_> you just use the jar
[10:19] <toad_> C-M|morphium: yes
[10:20] <toad_> Newsbyte: re finance page, here are my thoughts: I don't think that keeping it all together using a script that parses emails from paypal is the way to go
[10:20] <toad_> I think that that would suck and get inconsistent
[10:20] <toad_> I think we need to use paypal's and egold's automated interfaces, on a cron job, or every time we get a donation
[10:20] * pupok_ (~r00t@81-178-112-130.dsl.pipex.com) Quit (Read error: 104 (Connection reset by peer))
[10:20] <toad_> and make a summary that way
[10:20] <i2p_iip> <Newsbyte> heh...well, that was sanity's idea, really
[10:21] <toad_> I am quite willing to do this - IF ian agrees that it's important
[10:21] * pupok_ (~r00t@81-178-112-130.dsl.pipex.com) has joined #freenet
[10:21] * sanity (~ian@81-178-112-130.dsl.pipex.com) Quit (Read error: 104 (Connection reset by peer))
[10:21] <toad_> ALSO, we have a substantial amount of money in the bank, however it appears the only person with the ability to get a balance is Steven Starr, who is on permanent vacation
[10:21] <toad_> I believe ian can write cheques from the account but can't query the bank balance (which is surreal)
[10:22] <i2p_iip> <Newsbyte> he already did, months ago
[10:22] * interrupt (~chatzilla@64.122.23.214) has joined #freenet
[10:22] <toad_> anyway, isn't ian forwarding the paypal mails to you now?
[10:22] * sanity (~ian@81-178-112-130.dsl.pipex.com) has joined #freenet
[10:22] <toad_> wb sanity
[10:24] <i2p_iip> <Newsbyte> you mean, moved from the paypal account to a seperate bankaccount?
[10:24] <i2p_iip> <Newsbyte> I didn't receive squat thusfar
[10:25] <C-M|morphium> the freenet download server is slow :/
[10:25] <toad_> hmm?
[10:25] <toad_> what freenet download server?
[10:25] <C-M|morphium> http://freenetproject.org/snapshots/freenet-latest.tgz << this server
[10:26] <i2p_iip> <Newsbyte> he means the snapshotthingy
[10:26] <sanity> toad: if you want to devote a few hours to it - be my guest
[10:26] <sanity> toad: feel free to add a recipe to the .procmailrc file in my home dir on dodo
[10:26] <toad_> okay
[10:27] <toad_> sanity: running the script from dodo won't cause problems?
[10:28] <sanity> toad: why would it?
[10:29] <C-M|morphium> i'm running some debian machines... i installed java on one now, freenet says, it's too old...is there a newer debian java package (testing/stable) ?
[10:29] * sanity (~ian@81-178-112-130.dsl.pipex.com) Quit (Read error: 104 (Connection reset by peer))
[10:30] <i2p_iip> <Newsbyte> toad, could you give a short summary of what the deal is with people that have routers/firewalls, etc.? (for the wiki)?
[10:30] * sanity (~ian@81-178-112-130.dsl.pipex.com) has joined #freenet
[10:31] <i2p_iip> <Newsbyte> morph, I'm curious; do you know about www.freenethelp.org?
[10:31] <i2p_iip> <Newsbyte> sanity, your line doesn't seem to be much more stable then a few weeks ago. Is it just when connecting to IRC, or..?
[10:32] <C-M|morphium> no, thanks newsbyte
[10:32] <sanity> newsbyte: i don't know what the problem is, i have phoned my ISP and they blame my dsl modem (which, unfortunately, they didn't sell to me)
[10:34] <i2p_iip> <Newsbyte> I'm actually surprised how many dudes, especially on IRC, don't know about the wiki, even though sometimes they ask questions that are exactly explained there. It's mentionned more then once on the main site, though, so...how comes?
[10:35] <i2p_iip> <Newsbyte> if it's usb, it wouldn't surprise me. But I think you already said it's not usb, so...dunno...drivers? (but I suppose you use linux?)
[10:40] <i2p_iip> <Newsbyte> maybe you have a winmodem? ;-) j/k
[10:43] <sanity> my girlfriend did some debugging, but couldn't find anything
[10:43] <toad_> hmmm
[10:43] <sanity> i think i will wait and let her fix it :-)
[10:43] <toad_> they don't provide anything that does exactly what I want
[10:43] <toad_> so we basically have to write a screen scraper
[10:43] <sanity> toad: you are scraping from the web?!
[10:43] <toad_> paypal provides various tools but nothing to directly access the balance
[10:43] <sanity> toad: why not catch the incoming emails and use them
[10:44] <toad_> sanity: it'll get out of sync
[10:44] <sanity> a screen scraper will be a bitch
[10:44] <toad_> using the emails will get out of sync and be even more of a bitch
[10:44] * sanity (~ian@81-178-112-130.dsl.pipex.com) Quit (Read error: 104 (Connection reset by peer))
[10:45] * sanity (~ian@81-178-112-130.dsl.pipex.com) has joined #freenet
[10:45] <i2p_iip> <Newsbyte> there he goes again
[10:45] <i2p_iip> <Newsbyte> and there he is back again
[10:46] <hobx_> Why can't you just update by hand every few days?
[10:47] <toad_> because we don't want to give Newsbyte access to the account ;)
[10:47] <i2p_iip> <Newsbyte> well, can't you make an auto-reset every month or so? then it won't be much out of sync, and it's sufficient for a monthly update
[10:47] <sanity> hobx: some further thoughts:
[10:47] <i2p_iip> <Newsbyte> ermm..yeah, well, not every few days, hobx
[10:48] <i2p_iip> <Newsbyte> unless you take the manual part upon you :-)
[10:48] <sanity> hobx: the LRU scheme will tend towards each link being traversed by roughly the same number of messages over time, right?
[10:48] <i2p_iip> <Newsbyte> toad; lol
[10:48] <sanity> hobx: ie. if longer links were not being used as frequently as shorter links, they would be dropped, and visa-versa
[10:49] <i2p_iip> <Newsbyte> seems logical, indeed
[10:50] <sanity> [hobx: i did an experiment and it appears that requests *were* being evenly spread across different links - although you should verify this]
[10:51] <hobx_> sanity: I'm not sure about the load balancing thing
[10:51] * sanity (~ian@81-178-112-130.dsl.pipex.com) Quit (Read error: 104 (Connection reset by peer))
[10:51] * sanity (~ian@81-178-112-130.dsl.pipex.com) has joined #freenet
[10:52] <sanity> FUUUUUUUCK!!!!
[10:53] <sanity> hobx: if we could prove that the optimal distribution of links results in each link being used roughly the same number of times, then given that LRU will ensure that each link is used roughly the same number of times, then perhaps we prove that LRU results in an optimal distribution
[10:54] <hobx_> sanity: I had traffic data from my sims, and while no aweful (~70% of all data is handled by nodes doing less than twice the mean work), it was far from a completely even load.
[10:54] <sanity> ah, but that is requests per NODE, i am talking about requests per CONNECTION
[10:55] <sanity> hobx: 1) do you agree that LRU results in an even distribution of requests across >>connections<<
[10:56] <sanity> 2) can we prove that an even distribution of requests across connections implies an optimal network topology?
[10:58] <hobx_> It isn't a matter of agreeing. I don't see why that wouldn't be true, but I don't see why it would either.
[10:58] <sanity> well, first lets see if the simulation agrees
[10:58] <sanity> ...hang on
[10:59] <i2p_iip> <Newsbyte> so, I d/l unstable on the snapshot...and then what?
[10:59] <i2p_iip> <Newsbyte> speaking of which
[10:59] <hobx_> I think you should not emphasise the LRU part so much. The significant thing is that you are choosing neighbors distributed as the destination of incoming queries.
[11:00] <sanity> well, yes, but LRU is how to achieve that
[11:00] <sanity> or one way
[11:00] <i2p_iip> <Newsbyte> cough...i run it, how, again?
[11:01] <toad_> Newsbyte: it's in the email isn't it?
[11:01] <toad_> java -Xmx256M freenet.node.simulator.Main
[11:01] <toad_> and capture stderr
[11:01] <i2p_iip> <Newsbyte> no
[11:01] <sanity> hmz, actually that theory isn't quite as easy to test as i thought...
[11:02] <i2p_iip> <Newsbyte> you DO realise I'm a total noob on java? ;-)
[11:02] <hobx_> I can't believe you would want to write a simulator in teh framework of the existing Fred.
[11:02] <toad_> hobx: that way we can directly test NGRouting code
[11:02] <toad_> why is that a bad thing?
[11:02] <hobx_> I still have bad dreams about the heap of horror.
[11:02] <sanity> hobx: anyway, do you see where i am going with this?
[11:02] <toad_> we don't use Heap much now
[11:02] <hobx_> sanity: No, I don't.
[11:02] <i2p_iip> <Newsbyte> maybe it would make a nice addition for the wiki, for ppl that want to try out the simulator
[11:03] <sanity> hobx: if lru converges on a situation where each connection has an equal number of requests going across it
[11:03] <sanity> hobx: *and* we can prove that only an optimal network topology will have an equal number of requests going over each connection
[11:03] <sanity> hobx: *then* we have proved that LRU converges on an optimal network topology
[11:04] <hobx_> So you are saying i A implies B and C implies C then A implies C?
[11:04] <hobx_> So you are saying i A implies B and C implies B then A implies C?
[11:05] <sanity> um, no, if A implies B, and B implies C, then A implies C
[11:05] <hobx_> A = lru topology, B = equal number of connection, C = optimal topology
[11:05] <hobx_> ok, "only"
[11:05] <hobx_> But I don't see why that would be true at all.
[11:06] <sanity> me either - yet, but something tells me that it is
[11:07] <sanity> call it a gut feeling
[11:07] <sanity> AH
[11:08] <i2p_iip> <Newsbyte> indeed..gut feeling...
[11:08] <sanity> hobx: CONNECTIONS ARE A SCARCE RESOURCE!
[11:08] <i2p_iip> <Newsbyte> toad, ermm, so I d/l freenet-unstable-20040914.tgz ?
[11:08] <toad_> yes
[11:08] <sanity> hobx: if you aren't fully utilising the scarce resource then the topology must be sub optimal
[11:09] <sanity> hobx: only if each connection is used equally are you fully utilising the scarce resource, and thus an optimal network topology must use each connection equally
[11:09] <ShaunMacPherson> Sounds like economics, which proves economics is sometimes useful :)
[11:10] * MrNaughty (MrNaughty@d199-126-253-66.abhsia.telus.net) Quit ("\(^_^)/' No Soliciting!!! Unless you have legs way, way up and really, really big tits....")
[11:10] <sanity> hobx: sound plausable?
[11:11] <i2p_iip> <Newsbyte> yes, econmoics...that's why MS sells it's products so expensively; they ae scare resources! ;-)
[11:11] <i2p_iip> <Newsbyte> yes, econmoics...that's why MS sells it's products so expensively; they ae scare resources! ;-)
[11:15] <C-M|morphium> i'm unable to set up java properly on my debian box...someone here who could help me...?
[11:19] <i2p_iip> <Sonax> C-M|morphium: http://www.debian.org/doc/manuals/debian-java-faq/ch11.html
[11:20] <hobx_> sanity: Not particularly. And even if it did, that does not prove "only"
[11:21] <i2p_iip> <Sonax> Anyone know why gmane is down? Or is it just me having problems here?
[11:21] * sanity (~ian@81-178-112-130.dsl.pipex.com) Quit (Read error: 104 (Connection reset by peer))
[11:22] * sanity (~ian@81-178-112-130.dsl.pipex.com) has joined #freenet
[11:22] * sanity shoots isp
[11:23] <sanity> hobx: i think it is
[11:24] <sanity> hobx: can you think of a network topology that isn't optimal that will spread requests evenly over connections?
[11:24] <sanity> (not that your inability to do-so would prove anything)
[11:38] * MrNaughty (MrNaughty@d198-166-55-147.abhsia.telus.net) has joined #freenet
[11:39] <hobx_> I would suspect that any regular one would.
[11:39] <sanity> hobx: such as?
[11:39] <sanity> hobx: random interconnection would have a bias towards routing over shorter connections - so it wouldn't
[11:39] <hobx_> why?
[11:40] <sanity> because it routes towards peers that are closer to the destination, and if they have a connection to the destination, then it will be shorter than average
[11:42] <hobx_> If the destination is far away, it will use the longest one possible in that direction.
[11:42] <C-M|morphium> i know, i'm asking stupid questions...but how to configure my freenet client/server on the linux shell? i started it, it didn't ask me something...
[11:43] <sanity> hobx: yes, but that is only one hop over a long connection, followed by multiple hops over shorter connections
[11:45] <hobx_> I don't think it works that way.
[11:45] * sanity (~ian@81-178-112-130.dsl.pipex.com) Quit (Read error: 104 (Connection reset by peer))
[11:46] * sanity (~ian@81-178-112-130.dsl.pipex.com) has joined #freenet
[11:46] <toad_> C-M|morphium: hmmm?
[11:46] <toad_> what did you do exactly?
[11:46] <C-M|morphium> i started it
[11:46] <C-M|morphium> it says some things about java detected etc, and then
[11:46] <C-M|morphium> Starting Freenet now: Command line: java -Xmx128m -XX:MaxDirectMemorySize=128m freenet.node.Main
[11:47] <C-M|morphium> Done
[11:47] <C-M|morphium> and there aren't really much options in the config file
[11:48] <toad_> hmmm
[11:48] <toad_> didn't it ask you various questions?
[11:48] <C-M|morphium> no
[11:48] <toad_> rm freenet.conf
[11:48] <toad_> and then run start-freenet.sh again
[11:48] <toad_> after killing java obviously
[11:51] <toad_> okay, I got the screen scraper working
[11:51] * sanity (~ian@81-178-112-130.dsl.pipex.com) Quit (Read error: 54 (Connection reset by peer))
[11:51] <toad_> what to do with the value?
[11:51] <C-M|morphium> ok it started, but i can't access it remote anyway
[11:51] <toad_> huh?#
[11:51] * sanity (~ian@81-178-112-130.dsl.pipex.com) has joined #freenet
[11:51] <toad_> sanity: what to do with the $ value?
[11:52] <sanity> toad: what do you mean?
[11:52] <toad_> I wrote a screenscraper in perl
[11:52] <toad_> it will download from paypal the current total in USD
[11:52] <toad_> what do I do with that once I have it?
[11:53] <sanity> put it on the website
[11:53] <toad_> so include a file in donate.php that only has the current balance in?
[11:54] <sanity> ssh toad@freenet.sf.net echo $AMOUNT > .../htdocs/amount.txt
[11:54] <sanity> yeah, something like that
[11:54] <toad_> ahh, send it up directly... yeah
[11:54] <sanity> ssh toad@freenet.sf.net "echo $AMOUNT > .../htdocs/amount.txt"
[11:54] <toad_> and then include that from the php
[11:54] <toad_> how do i do that btw?
[11:54] <sanity> yeah
[11:54] <sanity> um
[11:54] <sanity> imclude("amount.txt");
[11:54] <sanity> er
[11:54] <sanity> include
[11:55] <toad_> @?
[11:55] <sanity> it is used elsewhere in the website
[11:55] <toad_> what's the formatting
[11:55] <toad_> ?
[11:55] * toad_ checks...
[11:55] <sanity> look in index.php
[11:55] <sanity> i don't think you need a @
[11:55] <toad_> <? include("menu.php"); ?>
[11:56] <sanity> yes, well - the <? indicates the start of some PHP code, and ?> indicates the end of it
[11:56] <sanity> so you might do:
[11:56] <toad_> do i need to go through cvs?
[11:56] <toad_> do non-cvs files get deleted?
[11:56] <toad_> <p>Current PayPal total:<? include("paypaltotal.txt"); ?></p>
[11:56] <sanity> no, i don't think they get deleted
[11:56] <sanity> yes
[11:57] <sanity> although you could include it from your home directory
[11:57] <sanity> be sure to give it liberal read permissions
[11:58] <sanity> a+r
[11:59] <toad_> how long does it normally take for it to update?
[11:59] <toad_> yay, it worked
[11:59] <toad_> okay..
[12:00] <C-M|morphium> is there a graphical FCP for windows?
[12:02] <toad_> C-M|morphium: there are several utilities that use FCP
[12:02] * sanity (~ian@81-178-112-130.dsl.pipex.com) Quit (Read error: 104 (Connection reset by peer))
[12:02] <toad_> that run on windows
[12:03] <toad_> e.g. FUQID, Frost
[12:04] <C-M|morphium> freenet is unable to find it's own readme...
[12:05] <C-M|morphium> nice...
[12:08] * sanity (~ian@81-178-112-130.dsl.pipex.com) has joined #freenet
[12:09] <C-M|morphium> hmm contact me if freenet works better...the idea is nice, and i would run it on some 100 mbit machines...but at the moment, it's not user-friendly
[12:11] <toad_> C-M|morphium: you expect it to be user friendly when you want to use the command line?
[12:11] <C-M|morphium> no
[12:12] <toad_> so what's the problem? it doesn't give you instant filez? we never claimed it did!
[12:12] <toad_> we explicitly say it won't on the download page
[12:12] <C-M|morphium> i expect something, that i can use to surf with freenet running on another machine
[12:12] <toad_> sure, so ssh tunnel it
[12:13] <toad_> what's the problem there?
[12:13] <C-M|morphium> why can't i simply use http://othermachine:8888/ ?
[12:13] <toad_> because that is a security risk
[12:13] <toad_> due to the nature of content that may be posted on freenet, it's not a good idea to run a gateway that is public for all the world to see
[12:14] <C-M|morphium> and then, why isn't the readme available...?
[12:14] <toad_> isn't it in the tarball?
[12:14] <C-M|morphium> it is CHK@ix0Z0wFt5jfWFGmbIZaLud~VOIsPAwI%2c2MBdgGPT57XMzl0AxpSKyA ...
[12:14] <C-M|morphium> and it isn't available
[12:15] <toad_> it should be in the unix tarball
[12:15] <C-M|morphium> ok, imagine i'm a windows user...
[12:15] <toad_> then you'd have to wait for it to fetch it
[12:15] <toad_> because the wininstaller doesn't include it
[12:15] <toad_> but it should eventually become available
[12:15] <toad_> HOWEVER windows users NEVER
[12:15] <toad_> EVER read the readme anyway
[12:16] <toad_> unix users very rarely read the readme
[12:16] <C-M|morphium> i'm a windows user, too
[12:16] <i2p_iip> <Newsbyte> geez.My mother just told me a brother of my uncle died by falling from a ladder. He was going to fix some stuff in my redoing of my house. :-/
[12:17] <i2p_iip> <Newsbyte> how weird, I spoke to him only a week ago.
[12:17] <toad_> any perl experts here?
[12:17] <toad_> I'm calling system "/usr/bin/rsync", "-e", "ssh", "~/paypaltotal.txt
[12:17] <toad_> ", "amphibian@shell1.sf.net:/home/groups/f/fr/freenet/htdocs/paypaltotal.txt"
[12:17] <toad_> when i run that from a shell, it works
[12:17] <toad_> when i run it from... oh, it's the @ isn't it..
[12:18] * C-M|morphium (~CMmorphiu@p213.54.211.53.tisdip.tiscali.de) Quit (Remote closed the connection)
[12:18] <i2p_iip> <Newsbyte> it was not close family, and I didn't know him all that well, but still. It's strange to realise the guy I just spoke a week ago is now dead
[12:19] <i2p_iip> <Newsbyte> ah, well, that's of no interest to anybody here, I guess. Dunno why I brought it up. Just got me by surprise.
[12:19] * sanity (~ian@81-178-112-130.dsl.pipex.com) Quit (Read error: 104 (Connection reset by peer))
[12:19] * sanity (~ian@81-178-112-130.dsl.pipex.com) has joined #freenet
[12:23] <toad_> Newsbyte: hrrm :(
[12:23] <toad_> Newsbyte: http://www.freenetproject.org/index.php?page=donate
[12:24] <toad_> it's updated hourly at present
[12:24] * toad_ will put it on dodo and make it be updated every time a donation happens
[12:25] <i2p_iip> <Newsbyte> yeah, k. Sorry about that emotional bit :-)
[12:26] <i2p_iip> <Newsbyte> as I was trying to ask, before I got a call from my mom; about the simulation, I d/l the tgz, and now I klick on..what?
[12:26] <toad_> sanity: how do I connect it up so that it runs whenever you get an email that would go into IN.freenet-donations?
[12:29] <i2p_iip> <Newsbyte> well, mentionning the total is a good start... but I had it a wee bit more detailed in mind, actually
[12:29] * sanity (~ian@81-178-112-130.dsl.pipex.com) Quit (Read error: 104 (Connection reset by peer))
[12:29] <toad_> Newsbyte: what sort of detail?
[12:30] * sanity (~ian@81-178-112-130.dsl.pipex.com) has joined #freenet
[12:31] <toad_> wb sanity, what ISP are you on? ;)
[12:34] <toad_> okay, updating it every 15 minutes
[12:34] <i2p_iip> <Newsbyte> well, like, what comes in and out on a daily or weekly basis (and sum that up to the total amount each month), anf to what/whome.
[12:36] <toad_> uh
[12:36] <toad_> some smallish amount for hosting dodo
[12:36] <toad_> that's $50-$100
[12:36] <toad_> not sure
[12:36] <toad_> and my pay
[12:36] <i2p_iip> <Newsbyte> wait, I'll give you an example, so you'll have a better idea
[12:36] <toad_> which is ?1200 per 140 hours
[12:36] <toad_> that's about it iirc
[12:37] <i2p_iip> <Newsbyte> there you go: http://www.emule-project.net/home/perl/general.cgi?l=29&rm=donations
[12:38] <toad_> hmmm possibly
[12:38] <toad_> but it would be considerably more work
[12:38] <toad_> suggest it on tech or somewhere
[12:38] <i2p_iip> <Newsbyte> now, THAT is a coolio donationpage :-)
[12:38] <i2p_iip> <Newsbyte> now, THAT is a coolio donationpage :-)
[12:39] <i2p_iip> <Newsbyte> yes, true, but it doesn't have to be like that in one shot. It would be nice already if you had the seperate donations mentionned, and I guess that would be more easy to do, since there is an email when a donation is made?
[12:40] <toad_> relatively easy
[12:40] <toad_> but not a priority ATM
[12:40] <toad_> if you want it to be one, start a thread on tech
[12:40] <i2p_iip> <Newsbyte> note, btw, that they work with paypal too :-)
[12:41] <i2p_iip> <Newsbyte> ah, yes, tech...ermm...you DID read the post where I said I already contacted a devl that claimed he wanted to do it...TWICE?
[12:43] <toad_> http://amphibian.dyndns.org/simulations2-nonewbie.png
[12:43] <toad_> hmmm
[12:43] <toad_> Newsbyte: that's my fault?
[12:43] <toad_> ian laid the blame squarely on that developer
[12:43] <toad_> the graph suggests psuccess might actually be falling... oh well, keep it running for another few days, and we'll know...
[12:45] <i2p_iip> <Newsbyte> I thought Ian said it wasn't anyones' fault ;-)
[12:46] <i2p_iip> <Newsbyte> anyway, where was I? yes, the tgz, openend it with winzip, got the jar and all the other files; what now?
[12:46] <toad_> you do what i said in the email
[12:46] <toad_> and what i've already said once today
[12:46] <toad_> java -cp freenet.jar:freenet-ext.jar -Xmx256M freenet.node.simulator.Main <options> > logfile
[12:47] <toad_> where <options> is options from the list i supplied in the email
[12:49] <i2p_iip> <Newsbyte> it says; 'fail to load Main-Class mainfest attribute'
[12:52] * jay (jcl@ool-18bf6dac.dyn.optonline.net) has joined #freenet
[12:53] <toad_> hi jay
[12:53] <toad_> Newsbyte: you have to be in the dir with freenet.jar
[12:55] <jay> whaddup toad!
[12:55] * sanity (~ian@81-178-112-130.dsl.pipex.com) Quit (Read error: 104 (Connection reset by peer))
[12:56] <jay> or "what's up"?
[12:56] <i2p_iip> <Newsbyte> indeed, jay, indeed
[12:57] <i2p_iip> <Newsbyte> toad, you are aware I'm only a poor misguided windowsuser, right?
[12:57] <jay> Newsbyte: what version of windows do you use?
[12:58] <i2p_iip> <Newsbyte> 3.11
[12:58] <toad_> Newsbyte: cd into the right directory
[12:58] <toad_> in a command prompt
[12:58] <toad_> and then run the above command
[12:58] <i2p_iip> <Newsbyte> j/k ;-) win2000
[12:58] <toad_> with your choice of options
[12:59] <jay> Newsbyte: i figured u were kidding
[12:59] <jay> toad_: good work on the paypal counter thingie on the website
[12:59] <i2p_iip> <Newsbyte> well, I've been asked the same, and sometimes, ppl actually think I'm using 3.11
[13:00] <i2p_iip> <Newsbyte> cd? you mean, like in DOs...geez...ok
[13:00] <jay> some people still use 3.11, 95, 98 etc.
[13:00] <jay> i wasn't that surprised to be honest
[13:01] <toad_> jay: hint: win3.11 doesn't have a 1.4 JVM
[13:01] * sanity (~ian@81-178-112-130.dsl.pipex.com) has joined #freenet
[13:01] <toad_> I doubt it has a 1.1 JVM ;)
[13:01] <toad_> a lot of people use 98
[13:01] <i2p_iip> <Newsbyte> well, 3.11 would really be an oldie. Though, it happens to be so, that I'm searching for it, for an old 386 I wanna donate
[13:01] <jay> toad_: heh true
[13:01] <jay> yeah there's lots of 98 users still out ther
[13:01] <jay> e
[13:02] <toad_> because it runs all the games and it was shipped with their generic off the shelf PC
[13:02] <jay> windows 3.1 was my intro to windows actually
[13:02] <jay> toad_: yup.. and it's real good at it too.. provided you do nothing else with it ;)
[13:03] <jay> like install office or anything
[13:03] <toad_> ;)
[13:03] <toad_> Longhorn will of course be required for DX10
[13:03] <toad_> DX10 games will not run on anything else - not linux either because they'll use TCPA+DMCA to prevent us emulating it
[13:03] * goatee__ (~goatee@ip216-239-82-182.vif.net) Quit (Read error: 104 (Connection reset by peer))
[13:04] <toad_> but hopefully we'll have more ports by then
[13:04] * goatee__ (~goatee@ip216-239-82-182.vif.net) has joined #freenet
[13:04] <jay> DMCA is rough.. to more i learn about it the more i understand the ramifications
[13:04] <jay> especially the ISP requirements that haven't yet gone through
[13:04] <toad_> DMCA on its own is irritating but not a big problem
[13:04] <toad_> DMCA+widespread adoption of Longhorn is scary
[13:05] <toad_> anyway...
[13:06] <toad_> should I try different numbers of buckets? IMHO it's a reasonable approximation for movementFactor, but buckets I'm not so sure about...
[13:06] * sanity (~ian@81-178-112-130.dsl.pipex.com) Quit (Read error: 54 (Connection reset by peer))
[13:06] <jay> we're OT again?
[13:06] <jay> heh
[13:06] * sanity (~ian@81-178-112-130.dsl.pipex.com) has joined #freenet
[13:06] <toad_> or should I try to get something with connections working?
[13:07] <toad_> I think I should let the current sims run before getting into non-fully-connected
[13:07] * sanity (~ian@81-178-112-130.dsl.pipex.com) Quit (Read error: 54 (Connection reset by peer))
[13:07] <toad_> hmmmm
[13:08] * sanity (~ian@81-178-112-130.dsl.pipex.com) has joined #freenet
[13:08] <toad_> if I want to do non-simulation things... then I need to be able to run the node...
[13:08] <i2p_iip> <Newsbyte> toooaaaadddieee! It doesn't work?
[13:08] <toad_> okay, if I renice them, and tune the node's priority up slightly... then maybe I can run one at the same time?
[13:08] <toad_> Newsbyte: hmm?
[13:08] <i2p_iip> <Newsbyte> you sure your command is correct, as in, cut&pastable?
[13:09] <toad_> does just doing java work? i.e. is java in your path?
[13:09] <toad_> hmmm
[13:10] <toad_> okay, I start my stable node...
[13:10] <toad_> I think the immediate usability issue now is transfer failures...
[13:10] <i2p_iip> <Newsbyte> it should, I already have a JVM that the stable node uses
[13:10] <toad_> which are not handled in a very friendly way
[13:10] <toad_> Newsbyte: type java -version
[13:10] <toad_> what happens?
[13:11] <i2p_iip> <Newsbyte> it says 1.4.2_04, just as I installed it
[13:11] <jay> on win 3.11 it causes an Abort, Retry, Fail, Ingore message..
[13:12] <jay> you know that 1.4.2_04 isn't the latest?
[13:12] <jay> the latest 1.4.2 build i mean
[13:12] <i2p_iip> <Newsbyte> it was when I downloaded it ;-)
[13:13] <jay> im speaking about the present tense, today
[13:13] <i2p_iip> <Newsbyte> and it also says 'mixed mode'
[13:14] <sanity> toad: what runs the screenscraper?
[13:14] <i2p_iip> <Newsbyte> ok, next, toad?
[13:14] <sanity> toad: i can't see it in my crontab
[13:14] <toad_> sanity: it's in mine atm
[13:15] <sanity> toad: can you make it ever 12 hours, ever 15 minutes seems *way* excessive, and we could get in trouble with Paypal (again)
[13:15] <jay> Newsbyte: does 'pwd' work on windows? i don't recall..
[13:15] <toad_> I couldn't find the filter in your .procmailrc - I was going to make it run on a mail coming to IN.freenet-donations
[13:15] <jay> sanity: yeah good point
[13:15] <i2p_iip> <Newsbyte> pwd? yeah, right ;-)
[13:15] <jay> i think CD will print the current dir
[13:16] <jay> resolving java classpath issues is tough for the newbie
[13:16] <jay> the students i tutor in it have much trouble figuring it out
[13:17] <sanity> toad: i made a few changes to the site
[13:17] <toad_> sanity: does it have to be 2x / day? what's reasonable? is 4x/day reasonable?
[13:17] <sanity> toad: i don't think 2x / day is *un*reasonable
[13:17] <sanity> i don't want to get banned from paypal again
[13:17] <toad_> true
[13:17] <i2p_iip> <Newsbyte> yeah, and those students aren't half the stubborn know-it-alls that I am, jay, lest not forget that! ;-)
[13:17] <toad_> well i've set it to 4x/day
[13:17] <sanity> toad: and i suspect they have bot detection
[13:17] <toad_> err 2x/day
[13:18] <toad_> Current PayPal balance (updated every 12 hours): $188.90
[13:18] <toad_> good
[13:18] <jay> Newsbyte: they all debate my advice to them, since i look more like a rock star than a coder
[13:18] <jay> so im used to it ;)
[13:18] <toad_> sanity: what should I be doing now?
[13:18] <toad_> I have several sims running:
[13:18] <sanity> toad: what are the current issues in stable?
[13:18] <toad_> locally I have fastestimators pcaching smoothing nonewbie noannounce movementfactor={0.2|0.1}
[13:18] <toad_> to find out whether we need a different movementFactor
[13:18] <sanity> good plan
[13:19] <toad_> on the big server, I have two more...
[13:19] <sanity> if there is nothing obvious for you to do with the sims, perhaps time to look at more practical stuff again
[13:19] <sanity> don't forget that we are still under pressure to make 0.6 viable
[13:20] <toad_> on the big server I have fastestimators pcaching nonewbie {smoothing|}
[13:20] <toad_> yeah, that's what I was thinking
[13:20] <toad_> there IS obvious stuff for me to do on the sims
[13:20] <toad_> however I'm inclined to do it after I've got some more results
[13:20] * pupok_ (~r00t@81-178-112-130.dsl.pipex.com) Quit (Read error: 104 (Connection reset by peer))
[13:21] <sanity> ideally i wish others would start screwing around with the sims leaving you with more practical (albeit less fun) stuff....
[13:21] <toad_> if we can speed things up a bit more, that would make it a lot easier to simulate 500 nodes with 50 node RTs
[13:21] * pupok_ (~r00t@81-178-112-130.dsl.pipex.com) has joined #freenet
[13:21] <i2p_iip> <Newsbyte> sigh. I *knew* this wasn't gonna work, getting advice from linuxusers and all that...
[13:21] <toad_> well, somebody on devl suggested he might work on his own sims a while back
[13:21] <toad_> maybe you should ask him?
[13:21] <toad_> a few days ago
[13:21] <sanity> who was that?
[13:22] <sanity> people offer to do stuff all the time, the number that actually end up doing it tends to be a much smaller subset :-(
[13:22] <jay> heh
[13:22] <toad_> hmmm
[13:22] <toad_> no he didn't
[13:22] <toad_> :(
[13:22] * moskau23 (~Miranda@dsl-082-082-224-074.arcor-ip.net) has joined #freenet
[13:22] <i2p_iip> <Newsbyte> well, for just running the standard stuff, I'm willing to do it (if I ever get it to run)
[13:23] <toad_> Jusa Saari
[13:23] <sanity> toad: i think if you put together a simple HOWTO, it will make things easier for people to muck around with it
[13:23] <jay> thelema talked about it for some time
[13:23] <i2p_iip> <Newsbyte> I'll note that most of the things I said I was going to do, I did so accordingly
[13:23] <toad_> sanity: yeah but there's still major stuff to be done to get simulations of non-fully-connected networks working
[13:23] <sanity> i know
[13:23] <toad_> but I just don't feel like doing that until it works considerably faster
[13:23] <sanity> ok
[13:24] <toad_> there are tweaks i could do e.g. trying out a different announce strategy (use perverse estimators instead of newbie routing; announcements using newbie seem to suck...)
[13:24] <sanity> toad: perhaps a good exercise would be for you to install a stable node from scratch and see what problems you encounter
[13:24] <i2p_iip> <Newsbyte> so, about the classpath; since I standard installed the java, shouldn't it be a normal path as well?
[13:25] * pupok_ (~r00t@81-178-112-130.dsl.pipex.com) Quit (Client Quit)
[13:25] <toad_> I'm getting incoming NaN MRI's occasionally...
[13:25] <toad_> I don't think we send them now but we may...
[13:25] <toad_> well, we should get error messages if we do
[13:25] <toad_> also, we need to handle transfer failed properly in fproxy
[13:26] <toad_> i suppose we should make a proper error message for it a la DNF/RNF?
[13:26] <sanity> yeah
[13:26] <toad_> the other option would be to retry
[13:26] <sanity> ok, i need to pop out soon
[13:26] <toad_> hmmm
[13:26] <toad_> I need to change port, wipe store, wipe RT, wipe node file...
[13:27] <sanity> toad: why not try installing it on dodo?
[13:28] <toad_> sanity: bandwidth/transfer limits?
[13:28] <toad_> sanity: what's our transfer limit on dodo?
[13:29] <toad_> seems a pretty critical piece of info...
[13:29] <i2p_iip> <Newsbyte> it says exeption in thread "main" java.lang.noclassdeffounderror yadayada
[13:30] <toad_> Newsbyte: are the files in the right dir?
[13:30] <toad_> in the same dir you're running it from?
[13:31] <i2p_iip> <Newsbyte> yes, I went with cd in the commandline to the dir, as you suggested
[13:31] <toad_> hmmm
[13:31] <toad_> odd
[13:31] <toad_> show me the command line, the pwd, and the output of "dir", via privmsg
[13:31] <i2p_iip> <Newsbyte> could be I've got thewrong tgz, ofcourse, but you said it was the right one
[13:32] <toad_> oh you can't privmsg
[13:32] <toad_> ugh
[13:32] <i2p_iip> <Newsbyte> of the 3 things you ask me, I don't know 1, and another is impossible
[13:32] <i2p_iip> <Newsbyte> indeed
[13:34] <i2p_iip> <Newsbyte> mind you, I have stable on the same puter, but you also said that wasn't a prob
[13:34] <toad_> okay, new stable node from scratch
[13:34] <toad_> uptime ~ 2 mins
[13:35] <toad_> open connections: nil
[13:35] <i2p_iip> <Newsbyte> who, me?
[13:35] <toad_> hmm?
[13:35] <toad_> several connection attempts have timed out...
[13:36] <i2p_iip> <Newsbyte> ermm...couldn't I jsut give you ftp access or sort, and you install/fix it right? :-)
[13:36] * sanity (~ian@81-178-112-130.dsl.pipex.com) Quit (Read error: 104 (Connection reset by peer))
[13:36] * sanity (~ian@81-178-112-130.dsl.pipex.com) has joined #freenet
[13:37] <i2p_iip> <Newsbyte> wb, sanity
[13:37] <toad_> you could install VNC maybe
[13:37] <toad_> you could email me the things I asked for above, at least
[13:37] * sanity (~ian@81-178-112-130.dsl.pipex.com) Quit (Read error: 104 (Connection reset by peer))
[13:39] <i2p_iip> <Newsbyte> yeah...how do I copy all that stuff in the commanddosthingy? Or must I use screencaptures? and what's that about pwd?
[13:41] <i2p_iip> <Newsbyte> VNC is, basically, a trojan-like-thingy, sub7like?
[13:42] <toad_> got TFE's alink...
[13:43] <toad_> Newsbyte: click the icon in the top left corner
[13:43] <toad_> there's a menu
[13:43] <toad_> including such useful stuff as "copy"
[13:43] * sanity (~ian@81-178-112-130.dsl.pipex.com) has joined #freenet
[13:43] <toad_> got TFE, CofE, TFHI alinks
[13:43] <toad_> ... at least...
[13:43] <toad_> with 3 open outbound conns
[13:44] <toad_> got TFHI site
[13:45] <toad_> the HTML, not all the images yet
[13:45] <toad_> got YoYo alink
[13:45] <toad_> FIND alink failed
[13:45] <i2p_iip> <Pseudonym> yay TFHI!
[13:45] <toad_> RNF on TFE and CofE
[13:46] <toad_> almost all the alinks on TFHI have RNFd...
[13:47] <i2p_iip> <Pseudonym> images on TFHI are active links to the other sites linked to
[13:47] <i2p_iip> <Pseudonym> images on TFHI are active links to the other sites linked to
[13:47] <i2p_iip> <Newsbyte> mail send away
[13:48] <toad_> the gameplan would be to get the simulations up to a convincing level with simulation of say 500 nodes with 50 node RTs and an HTL of 10
[13:48] <toad_> and fix a few bugs in the current codebase, which seems to be working reasonably well
[13:48] <toad_> and call that 0.6-rc1
[13:48] <toad_> 6 connections...
[13:48] <toad_> Pseudonym: indeed
[13:49] <i2p_iip> <Newsbyte> and sell the simulations as 0.6! Good idea! ;-)
[13:49] <toad_> Newsbyte: show me the output of dir as well please
[13:50] <i2p_iip> <Newsbyte> give me a sec so I can try to comprehend what you're asking :-)
[13:50] <toad_> Newsbyte: go to the command line, type "dir"
[13:50] <toad_> and press enter
[13:51] <toad_> got Dolphin activelink
[13:51] <i2p_iip> <Newsbyte> o, that. You coders have funny ways of asking simple things :-p
[13:52] <toad_> well, the other option: continue to fix bugs, do some simulations, present the simulations to the world along with a demand for more cash or we'll go bust, get slashdotted, get more cash, continue to solve the Harder Problems, plod towards 0.6
[13:53] <toad_> having said that, I believe there is enough in the bank to pay me for this month... IF ian can get hold of steven
[13:53] <toad_> yay got FIND HTML
[13:53] <i2p_iip> <Newsbyte> mail send
[13:54] <toad_> hmmm
[13:54] <toad_> i don't know
[13:54] <i2p_iip> <Newsbyte> now, no doubt, it will show I lack several packages...but hey, I asked if it was that which I should d/l :-)
[13:54] <toad_> is the syntax for -cp different on windows?
[13:54] <toad_> very probably...
[13:54] <i2p_iip> <Pseudonym> I vote for 0.6-rc1
[13:54] <toad_> got Content of Evil HTML
[13:55] <toad_> 509Whatever
[13:55] <toad_> 15th September
[13:55] <toad_> Thanks to all those who pointed out I'd missed the announcement of Freenet 5094 and 5095. Actually I didn't, but after finding that they both appeared pretty much as predicted in the last entry I realised I couldn't be bothered to say I told you so.
[13:55] <toad_> Well that and the small matter of being too afraid to alter this page because it'd mean waiting another three bloody hours for FIW to insert it, and then not have the page retrievable the next day anyway.
[13:55] <toad_> Yes I'm pissed.
[13:55] <toad_> Off that is.
[13:55] <toad_> okay, that's another argument for not shipping yet
[13:55] <toad_> got all the images on CofE except for one
[13:56] <toad_> if we had transfer coalescing we'd have got all of them...
[13:56] <toad_> got 10 connections
[13:56] <toad_> reloading TFHI, maybe we get some images...
[13:57] <toad_> browsing-out-of-the-box performance really isn't bad - at least by historical standards...
[13:57] <i2p_iip> <Newsbyte> ah, yes...well, my connection deteriorated, if I may recall
[13:58] <toad_> Newsbyte: you have a firewall you can't mess with?
[13:58] <toad_> eww broken images on YoYo... transfer coalescing would definitely help the first-time-user experience
[13:59] <i2p_iip> <Newsbyte> I have a whole bunch of firewalls I can extensively mess with (and have)
[13:59] <toad_> Newsbyte: so do you have the port forwarded or not?
[13:59] <i2p_iip> <Newsbyte> on which puter?
[14:00] <toad_> on the one where it collapsed to 1 connection
[14:01] <toad_> Unknown Network Error: freenet.client.WrongStateException: Wrong state: FAILED should be DONE: null: Transfer failed with 168578 moved.
[14:01] <toad_> hehe, the infamous...
[14:01] <i2p_iip> <Newsbyte> well, at least on my local firewall it has a whole, dunno about the other
[14:02] <i2p_iip> <Newsbyte> that said, why would my unstable work on my other puter, if the router itself didn't leave the port open?
[14:02] <i2p_iip> <Newsbyte> besides, I thought the new builds were supposed to work with firewalls and routers and all that?
[14:02] <toad_> Newsbyte: different port...
[14:03] <toad_> bidi gives you some gain, but if the FNP port isn't forwarded you'll still have problems
[14:03] <i2p_iip> <Newsbyte> I have difficulties understanding this: shareaza, emule and all the like are P2P systems too, on strange ports...THEY seem to work ok
[14:04] <soros> who has the sexier head , ben kingsley or jean luc picard ?
[14:04] <i2p_iip> <Newsbyte> Worf
[14:06] <toad_> hehe
[14:07] <i2p_iip> <Newsbyte> so, any news on the command java thingy?
[14:07] * sanity (~ian@81-178-112-130.dsl.pipex.com) Quit (Read error: 104 (Connection reset by peer))
[14:08] <toad_> bidi conns are nice... you shut down, restart, and THEY connect to YOU!
[14:08] * sanity (~ian@81-178-112-130.dsl.pipex.com) has joined #freenet
[14:08] <toad_> Newsbyte: hmmm
[14:08] <toad_> try ; instead of :
[14:08] <toad_> in the -cp argument
[14:08] <i2p_iip> <Newsbyte> you mean: in Russia?
[14:08] <toad_> hmm?#
[14:09] <i2p_iip> <Newsbyte> "In Russia, THEY connect to YOU"
[14:09] <toad_> heh
[14:09] <i2p_iip> <Newsbyte> /. thingy :-)
[14:10] <toad_> Newsbyte: did it make any difference?
[14:11] <i2p_iip> <Newsbyte> ermm, can you give me the command again? sorta lost it in the other brouha
[14:12] <toad_> java -cp freenet.jar;freenet-ext.jar -Xmx256M freenet.node.simulator pcaching smoothing nonewbie fastestimators > logfile
[14:12] <toad_> try that
[14:13] <i2p_iip> <Newsbyte> Exception in thread "main" java.lang.NoClassDefFoundError: freenet/node/simulato
[14:16] <i2p_iip> <Newsbyte> you know, if that VNC thingy is free and not to difficult to install and run, I might exactly do that. On the puter you provided some extra ram for, that is. I was going to do use that for the testnetwork anyway, but seems the testnetwork is not for this year (me avoids looking for sanity)
[14:16] <i2p_iip> <Newsbyte> as long as it's safe, I don't mind having you total control of that puter
[14:18] <i2p_iip> <Newsbyte> consider it an extra testbed, or something
[14:18] <toad_> cool
[14:18] <toad_> I imagine it supports some sort of passwording
[14:18] <i2p_iip> <Newsbyte> though, the wiki is on there too, so you will have to be a bit careful
[14:19] <toad_> on the same PC?
[14:19] <i2p_iip> <Newsbyte> passwording fine, but does it encrypt the whole tunneling thingy?
[14:19] <toad_> I think so
[14:19] <i2p_iip> <Newsbyte> currently, yes
[14:20] <i2p_iip> <Newsbyte> I only have three puters, for the moment, alas
[14:21] <i2p_iip> <Newsbyte> well, actually 7, but 2 are not mine, and 2 are to old
[14:21] <i2p_iip> <Newsbyte> it does mean, that you couldn't use the b/w full throtle on that puter...
[14:22] <i2p_iip> <Newsbyte> but it would be fine to run a simulation, me thinks?
[14:24] <toad_> simulation will use as much cpu as it can
[14:25] <toad_> but apart from that should be fine
[14:26] <i2p_iip> <Newsbyte> well..it's an AMD K6-2 500Mhz...oldy, of course, but not completely useless in that respect, me thinks.
[14:26] <toad_> hmmm
[14:26] <toad_> probably pretty useless at the moment
[14:26] <toad_> simulations are still pretty slow
[14:26] * Ash-Fox (Hal-9000@aav130.neoplus.adsl.tpnet.pl) has joined #FreeNET
[14:26] * jay (jcl@ool-18bf6dac.dyn.optonline.net) Quit (".")
[14:26] <toad_> well that's not entirely true; routing learns far too slowly
[14:26] <toad_> even in simulation
[14:27] <i2p_iip> <Newsbyte> it is? I thought you said it was good enough to run Freenet?
[14:27] <toad_> it's good enough for freenet
[14:27] <toad_> but simulations run slooowly
[14:27] <i2p_iip> <Newsbyte> or is the simulation that more 'heavy' on cpu then a freenetnode?
[14:27] <i2p_iip> <Newsbyte> right
[14:27] <toad_> yes, sim will use 100% cpu
[14:28] <i2p_iip> <Newsbyte> well...100% is a bit.. much ;-)
[14:28] <toad_> :)
[14:28] <toad_> all simulators do that
[14:29] <i2p_iip> <Newsbyte> well, anyway, I hope you're happy your ram is already in good use for the wiki :-)
[14:30] <toad_> very happy
[14:31] <i2p_iip> <Newsbyte> now, about the router/firewall thingy: DOES one, or not, have to make a whole in the firewall, adjust the NAT etc, or not?
[14:31] <i2p_iip> <Newsbyte> it would be good to know, for the wiki
[14:34] <toad_> it may be possible to run a node behind a firewall without forwarding the port
[14:35] <toad_> but it is much more likely to work well, especially during the bootstrap phase, if you forward the listenPort
[14:36] <toad_> bbiab
[14:36] * sanity (~ian@81-178-112-130.dsl.pipex.com) Quit (Read error: 54 (Connection reset by peer))
[14:36] <i2p_iip> <Newsbyte> hmmm, that's like...ambigue. I mean, current view on the lists and even the wiki, says that the new builds work without any adaptations. Do you think that might be exagerated, and that the probs of my node may be related to that?
[14:36] * sanity (~ian@81-178-112-130.dsl.pipex.com) has joined #freenet
[14:37] <i2p_iip> <Newsbyte> and, as a side-matter: who's the Freenet Project Board exactly?
[14:41] * Sugadude (foobar@outpost.zedz.net) Quit (Remote closed the connection)
[14:44] * nextgens (~nextgens@d80-170-160-200.cust.tele2.fr) has joined #freenet
[14:44] <Ash-Fox> The borg queen, unimatrix 001
[14:59] <hobx_> T1000
[15:16] * nextgens (~nextgens@d80-170-160-200.cust.tele2.fr) Quit ("Download Gaim: http://gaim.sourceforge.net/")
[15:16] * Ash-Fox (Hal-9000@aav130.neoplus.adsl.tpnet.pl) Quit (Read error: 104 (Connection reset by peer))
[15:17] * nextgens (~nextgens@d213-103-71-33.cust.tele2.fr) has joined #freenet
[15:23] * Ash-Fox (Hal-9000@aav130.neoplus.adsl.tpnet.pl) has joined #FreeNET
[15:34] * Zorix (Brandon@fl-69-69-7-231.dyn.sprint-hsd.net) Quit (Read error: 110 (Connection timed out))
[15:38] * nextgens is now known as nextgens|away
[15:38] * sanity (~ian@81-178-112-130.dsl.pipex.com) Quit (Read error: 104 (Connection reset by peer))
[15:39] * sanity (~ian@81-178-112-130.dsl.pipex.com) has joined #freenet
[15:46] * Sugadude (debian-tor@209.142.37.21) has joined #freenet
[16:02] <i2p_iip> <Newsbyte> so, anyway, we've been waiting so long for the 0.6, maybe we can wait a bit longer and slashdot with the simulations, gain some money, and work further for the 0.6 (especially the small chunks thingy). What you guys think?
[16:08] <i2p_iip> <Pseudonym> I think we've been waiting so long for 0.6 that we ought to release anything that halfway works as soon as possible.
[16:08] * sanity (~ian@81-178-112-130.dsl.pipex.com) Quit (Read error: 104 (Connection reset by peer))
[16:08] * sanity (~ian@81-178-112-130.dsl.pipex.com) has joined #freenet
[16:09] <hobx_> ah
[16:09] <Ash-Fox> <Ash-Fox> Ah... I just found the first instant messenger application I wrote in 1997 ^^
[16:09] <Ash-Fox> <Ash-Fox> Might be full of spelling errors, but it worked better than ICQ ever did
[16:09] <hobx_> The release strategy of all
[16:09] <Ash-Fox> <Ash-Fox> Hmm.. wtf, I was 12 when I wrote that?
[16:09] <Ash-Fox> <Ash-Fox> @.@ I was really talented
[16:09] <hobx_> And look what you made of it...
[16:10] * nextgens|away (~nextgens@d213-103-71-33.cust.tele2.fr) Quit (Read error: 60 (Operation timed out))
[16:18] <i2p_iip> <Newsbyte> hobx; lol ;-)
[16:18] <i2p_iip> <Newsbyte> hobx; lol ;-)
[16:20] <i2p_iip> <Newsbyte> anyway, the big prob for working further on it ad infinitum was that you can't keep that up, finance wise; you need some fresh input. 0.6 would do, but let's face it, it doesn't perform much better then 0.5..and the simulations would be a cool in-between
[16:20] * sanity (~ian@81-178-112-130.dsl.pipex.com) Quit (Read error: 104 (Connection reset by peer))
[16:22] <i2p_iip> <Newsbyte> reaping some gain, easing the financial pressure for a while, and not costing us much in time and effort, in comparison
[16:24] * Ash-Fox adds i2p_iip, as he finds relays anoying
[16:24] <Ash-Fox> *to ignore
[16:26] * sanity (~ian@81-178-112-130.dsl.pipex.com) has joined #freenet
[16:27] <i2p_iip> <Newsbyte> /me sympathises with ash-fox' problems
[16:30] * pittaman (~mistery_b@d51A410CA.kabel.telenet.be) Quit ("Bethje roelt!!!! >(")
[16:51] * |ux (~|ux@82-37-17-24.cable.ubr03.wolv.blueyonder.co.uk) has joined #freenet
[16:59] * |ux (~|ux@82-37-17-24.cable.ubr03.wolv.blueyonder.co.uk) Quit ("Leaving")
[16:59] * Zorix (Brandon@fl-69-69-7-231.dyn.sprint-hsd.net) has joined #freenet
[17:02] * FallingBuzzard (~srademach@207.152.112.129) has left #freenet
[17:31] * sanity (~ian@81-178-112-130.dsl.pipex.com) Quit ()
[18:02] * interrupt (~chatzilla@64.122.23.214) Quit ("Chatzilla 0.9.64g [Mozilla rv:1.7/20040803]")
[18:09] * robilad (~topic@mpiat2313.ag2.mpi-sb.mpg.de) Quit (Remote closed the connection)
[18:10] * Iakin (Iakin@as25-4-6.kp.g.bonet.se) Quit (Read error: 60 (Operation timed out))
[18:28] * Canar (~canar@acacia.bcit.ca) Quit (Read error: 104 (Connection reset by peer))
[18:28] * Canar (~canar@acacia.bcit.ca) has joined #freenet
[18:41] * 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!")
[19:08] <toad_> <i2p_iip> <Newsbyte> hmmm, that's like...ambigue. I mean, current view on the lists and even the wiki, says that the new builds work without any adaptations. Do you think that might be exagerated, and that the probs of my node may be related to that? - yes, you need to forward the port
[19:08] <toad_> for best performance
[19:09] <toad_> <i2p_iip> <Newsbyte> and, as a side-matter: who's the Freenet Project Board exactly? - ian, steven, oskar, not sure who else
[19:09] <toad_> <Ash-Fox> The borg queen, unimatrix 001 - are you a bot that randomly spouts trek lore?
[19:10] <toad_> <i2p_iip> <Newsbyte> so, anyway, we've been waiting so long for the 0.6, maybe we can wait a bit longer and slashdot with the simulations, gain some money, and work further for the 0.6 (especially the small chunks thingy). What you guys think? - that's what I said :)
[19:10] <toad_> woah Ash-Fox
[19:27] <toad_> http://secunia.com/advisories/12526/
[19:27] <toad_> son of a ... we're supposed to be milking M$'s vulnerabilities for all they're worth, not getting clogged up with our own!!
[20:22] * lolo-laptop (~lostlogic@68.251.84.226) Quit ("Client exiting")
[20:23] * guido^pe (~unknown@dsl-213-023-246-084.arcor-ip.net) Quit (Remote closed the connection)
[20:31] * moskau23 (~Miranda@dsl-082-082-224-074.arcor-ip.net) Quit (Read error: 113 (No route to host))
[20:40] * Zorix (Brandon@fl-69-69-7-231.dyn.sprint-hsd.net) Quit (Read error: 110 (Connection timed out))
[21:00] * Sugardude (~Sugadude@216.32.201.125) has joined #freenet
[21:02] * Sugadude (debian-tor@209.142.37.21) Quit (Remote closed the connection)
[21:04] * Zorix (Brandon@fl-69-69-7-231.dyn.sprint-hsd.net) has joined #freenet
[21:05] <Zorix> worst irc network ever :/
[21:05] <Zorix> dunno why i get dropped so often and half the time it takes 2 hours to ping out and i just think ppl are being quiet.. i can remain connected to 2 other irc networks with no problems
[21:10] <toad_> odd
[21:10] <toad_> it's not that bad for me
[21:11] <Zorix> irc.freenode.net is what i use
[21:11] <Zorix> is there a better one?
[21:11] <Zorix> there must be multiple servers here
[21:13] <toad_> i use irc.debian.org
[21:14] <toad_> hmmm
[21:14] <toad_> movementFactor=0.2 is disastrous
[21:15] <toad_> movementFactor=0.1 is similar to existing code, we'll see where it goes...
[21:15] <toad_> it takes maybe 2 million reqs to reach 93% success
[21:15] <toad_> and then it hits an asymptote at around 95%
[21:15] <toad_> with announcements it actually starts to fall...
[21:29] <Zorix> toad_ theres a lot of weird stuff goin on with freenet lately.. most people call it broke now heh
[21:30] <Zorix> stuff is hard to upload and download
[21:48] <toad_> Zorix: talk to you tomorrow about it? it's 3AM here...
[21:48] <toad_> bbl zzz
[22:06] * Zorix (Brandon@fl-69-69-7-231.dyn.sprint-hsd.net) Quit (Read error: 110 (Connection timed out))
[22:11] * hoppalong (~hedvig@d207-216-10-15.bchsia.telus.net) has joined #freenet
[22:14] * Zorix (Brandon@fl-69-69-7-231.dyn.sprint-hsd.net) has joined #freenet
[22:14] <Zorix> see if debian.org is better
[22:18] * goatee__ (~goatee@ip216-239-82-182.vif.net) has left #freenet
[22:18] * goatee (~goatee@ip216-239-82-182.vif.net) has joined #freenet
[22:22] * goatee (~goatee@ip216-239-82-182.vif.net) Quit (Client Quit)
[22:22] * goatee (~goatee@ip216-239-82-182.vif.net) has joined #freenet
[22:39] * hoppalong (~hedvig@d207-216-10-15.bchsia.telus.net) has left #freenet
[22:57] * Zorix (Brandon@fl-69-69-7-231.dyn.sprint-hsd.net) Quit (Read error: 110 (Connection timed out))
[23:14] * dystopia (admin@gaia-wan0.ipv6.freeshell.bofx.net) Quit (Read error: 113 (No route to host))
[23:19] * dystopia (admin@gaia-wan0.ipv6.freeshell.bofx.net) has joined #freenet
These logs were automatically created by Jay Oliveri with his gimp hapi on irc.freenode.net.
