Timestamps are in GMT/BST.
[0:04] * Treyvan (n=fred@) has left #freenet
[0:11] * Apophis2 (n=Apophis@) Quit (Read error: 54 (Connection reset by peer))
[0:38] * Caco_Patane (n=caco@) has joined #freenet
[0:43] * lattt (n=gerhard@) has joined #freenet
[1:06] * FCN (n=chatzill@) Quit ("Chatzilla 0.9.77 [Firefox 2.0/2006101023]")
[1:34] * droptothetop (n=droptoth@) has joined #freenet
[1:35] * droptothetop (n=droptoth@) Quit (Read error: 104 (Connection reset by peer))
[1:37] * heph (n=heph@) Quit (Remote closed the connection)
[1:45] * PraiseChaos (n=kcecil@) has joined #freenet
[1:59] <Andross> anyone know why my FCP would be messed up?
[2:00] <Andross> do I need to restart the node once I enable FCP?
[2:20] * lordwilliam (n=jesusisl@) has joined #freenet
[2:20] * lordwilliam (n=jesusisl@) has left #freenet
[2:21] * ATGeek (n=ATG@) Quit ("This computer has gone to sleep")
[2:24] * ATGeek (n=ATG@) has joined #freenet
[2:26] * attic (i=attic@) has joined #freenet
[2:28] <attic> I was thinking of reinstalling linux on my other box and run Freenet in a more dedicated environment. My question is, is there any way to transfer my nodes from this box or is my ref dependant on my current box?
[2:31] <Bombe> Nope, it isn't. Copying the complete freenet directory should preserve everything.
[2:32] <attic> Bombe, thanks, it was the answer I was hoping for :)
[2:32] <Bombe> Glad to be of service. :)
[2:33] <attic> You might need to be of service again, since I'm not too good at *ix yet, hehe. But I think I'll get it working in the morning.
[2:33] <Bombe> Good, I'll be gone by then. ;)
[2:34] <attic> :))
[2:59] * ATGeek (n=ATG@) Quit ("This computer has gone to sleep")
[3:07] * ATGeek (n=ATG@) has joined #freenet
[3:26] * needle (n=nevle@) has joined #freenet
[3:37] * needle (n=nevle@) Quit ("Leaving")
[4:30] * sham (n=sham@) has joined #freenet
[4:32] <sham> I need peers! Anyone peer with me?
[4:32] <attic> sham, please join #freenet-refs for those requests
[4:32] <sham> Oh Sorry
[4:33] <attic> np
[4:36] * kinzite (n=kinzite@) has joined #freenet
[4:39] * sham is now known as harlequin516
[4:42] * Moonboy (n=Joey@) has joined #freenet
[4:46] * dickie (n=pirch@) has joined #freenet
[4:46] * Moonboy (n=Joey@) has left #freenet
[4:59] * Andross (n=drew@) Quit (Remote closed the connection)
[5:17] * peerless (n=Peerless@) has joined #freenet
[5:28] * ATGeek (n=ATG@) Quit ("Leaving")
[5:38] * ATravelingGeek (n=atg@) Quit (Read error: 104 (Connection reset by peer))
[5:39] * dickie (n=pirch@) has left #freenet
[5:58] * ATravelingGeek (n=atg@) has joined #freenet
[6:28] * nharlequi516 (n=sham@) has joined #freenet
[6:33] * harlequin516 (n=sham@) Quit (Read error: 145 (Connection timed out))
[7:04] * needle (n=nevle@) has joined #freenet
[7:17] * needle (n=nevle@) Quit ("Leaving")
[7:22] * mk (n=mk@) has joined #freenet
[7:33] * peerless (n=Peerless@) Quit (Read error: 104 (Connection reset by peer))
[7:47] * nharlequi516 (n=sham@) Quit ("BitchX: nine out of ten gynecologists recommend it")
[7:51] * railk (n=railk@) has joined #freenet
[7:56] * railk (n=railk@) Quit (Remote closed the connection)
[8:24] * mk (n=mk@) has left #freenet
[8:27] * railk (n=railk@) has joined #freenet
[8:34] * IMCensored_ (n=KMIntern@) has joined #freenet
[8:35] * IMCensored1 (n=KMIntern@) Quit (Nick collision from services.)
[8:35] * IMCensored_ is now known as IMCensored1
[8:37] * sopues_ (n=sandos@) Quit (Read error: 131 (Connection reset by peer))
[8:40] * cyberdo (i=cyberdo@) has joined #freenet
[8:48] <IMCensored1> dunno if any devs are up... i have latest testing... around 11 PM EST my node shut off and wont restart ... the error message is in the wrapper log - http://dark-code.bulix.org/jvev4h-26502?raw - if its on my end, please pm me so i dont cause it to occur again, if not... heres a bug report =)
[8:49] <IMCensored1> im going to try update testing ... if that doesnt fix it... i'll update.cmd to get the latest stable
[8:51] <IMCensored1> oh and sectoor.de is always the slowest of all the mirrors
[8:53] <IMCensored1> well neither stable or testing is working now
[8:58] * Bardamu (i=Bardamu@) has joined #freenet
[8:58] <IMCensored1> i'll try a system restart
[8:59] * IMCensored1 (n=KMIntern@) Quit ("system restart")
[9:08] * railk (n=railk@) Quit (Remote closed the connection)
[9:09] <theseeker> just a note: awesome idea on the notification of bookmarked site updates!
[9:09] * IMCensored1 (n=KMIntern@) has joined #freenet
[9:10] <nextgens> IMCensored1> please fill in a ticket on the BTS ... and include long lines of your freenet.ini file
[9:10] * sanity (n=ian@) Quit ()
[9:10] <nextgens> IMCensored1> backup your freenet.ini and strip long lines if you need a quick workaround
[9:11] * nextgens sets mode +v Bardamu
[9:11] * lattt (n=gerhard@) Quit ("baba")
[9:11] <Bardamu> merci nextgens
[9:11] <Bardamu> sorry thanks nextgens :p
[9:13] * Dan541 (n=Trojan@) has joined #freenet
[9:17] <IMCensored1> alright
[9:18] <nextgens> it's likely it's a side effect of https://bugs.freenetproject.org/view.php?id=1022
[9:18] <nextgens> and you have been using a non-fixed build
[9:19] * Mrtwister (n=chatzill@) has joined #freenet
[9:20] * Dan541 (n=Trojan@) has left #freenet
[9:20] * Bardamu (i=Bardamu@) Quit (Remote closed the connection)
[9:21] * Mrtwister (n=chatzill@) Quit (Client Quit)
[9:21] <IMCensored1> yep i see that
[9:21] <IMCensored1> it was the bookmarks
[9:21] * whiterabbit (n=Whiterab@) has joined #freenet
[10:03] * railk (n=railk@) has joined #freenet
[10:18] * railk (n=railk@) Quit (Remote closed the connection)
[11:02] * IMCensored_ (n=KMIntern@) has joined #freenet
[11:15] * needle (n=nevle@) has joined #freenet
[11:20] * IMCensored1 (n=KMIntern@) Quit (Read error: 110 (Connection timed out))
[11:20] * IMCensored_ is now known as IMCensored1
[11:29] * sopues (n=sandos@) has joined #freenet
[11:47] <_ph00> on the new 'peer location distribution' thuingy, I am the blue X, right? but how are location rappresented? top left quadrant should be europe and the bottom right one is probably america or north america...
[11:47] <_ph00> or doest it work another way?
[11:48] <_ph00> I mean how do I read it? I only figured out that my node is 'probably' the blue X
[11:50] * railk (n=railk@) has joined #freenet
[11:51] * needle (n=nevle@) Quit ("Leaving")
[11:54] * whiterabbit (n=Whiterab@) Quit ("KVIrc 3.2.5 Anomalies http://www.kvirc.net/")
[11:55] * whiterabbit (n=Whiterab@) has joined #freenet
[11:55] * Caco_Patane (n=caco@) Quit ("Leaving")
[11:58] * Zothar is now known as Zothar_Work
[12:00] <Zothar_Work> _ph00: the locations on the circle are your peer locations between 0.0 and 1.0
[12:00] <_ph00> Zothar_Work: and how do I interpret those values as 'locations'?
[12:02] <Zothar_Work> _ph00: they're just numbers; the circular distance between your node and the peer determines swap behavior and the circular distance between a request's location and a peer determines routing
[12:02] <_ph00> how are they calculated? (roughly)
[12:03] <_ph00> and why call them 'location' anyway?
[12:03] <_ph00> 'circular distance'?
[12:03] <Zothar_Work> probably location because it's what's used for routing and it's what is supposed to be "small world"
[12:04] <_ph00> small world like 'internet within the internet'?
[12:04] <Zothar_Work> circular distance is basically the distance between two points on a circle; .6 and .7 is .1, but .1 and .9 is .2
[12:04] <_ph00> hm
[12:04] <_ph00> I thought those were calleds 'arch' or something like that
[12:05] <_ph00> (don't know the exact word used in English)
[12:05] <Zothar_Work> _ph00: "small world" as in the graph theory Freenet is designed around; read the white paper(s), watch the Freenet video and check Wikipedia
[12:05] <_ph00> k
[12:06] <Zothar_Work> arc maybe? nah, we're not doing geometry; we're just doing a number space between 0.0 and 1.0 that wraps, thus the circle
[12:07] <_ph00> hm
[12:09] <_ph00> uh. ok, I think I get it... somehow. (I should read more about it, but I get the idea of circular distance)
[12:15] * Cage (n=someone@) has joined #freenet
[12:19] * vfervers (n=vfervers@) Quit (Read error: 145 (Connection timed out))
[12:23] * Cage (n=someone@) Quit ()
[12:31] * rasco (i=rasco@) has joined #freenet
[12:32] * vfervers (n=vfervers@) has joined #freenet
[13:25] <sleon> FreenetLogBot: this is wonderfull :) i am so pleased to hear it
[13:26] * greycat (i=rfc1413@) has joined #freenet
[13:35] * toad_ (n=toad@) has joined #freenet
[13:35] * ChanServ sets mode +o toad_
[13:46] * railk (n=railk@) Quit ("Cya, wouldn't want ta be ya!")
[13:50] <nextgens> hi toad_
[13:52] <nextgens> toad_> may you have a look to https://bugs.freenetproject.org/view.php?id=1026 please ? I don't understand how it can possibly occur... As the database ought to be started before fproxy
[13:52] <nextgens> and the node constructor "done"
[14:06] <toad_> ok
[14:06] <toad_> did you make some graphs of downloads?
[14:10] * nextgens doesn't remember
[14:10] <nextgens> no, I collected data but haven't managed to make gnuplot output anything usefull iirc
[14:11] <nextgens> toad_> http://emu.freenetproject.org/~nextgens/dl_stats.txt
[14:12] <nextgens> of course the sum is cumulative and not per month
[14:13] * nextgens updates the stats
[14:15] <nextgens> doh
[14:15] <nextgens> the sum is broken
[14:15] * nextgens fixes
[14:17] <Zothar_Work> of possible interest: http://theochem.ki.ku.dk/on_line_docs/gnuplot/gnuplot_16.html
[14:17] <nextgens> toad_> ok, the stats are available from august to now ... and the number of downloads decrease slowly
[14:17] <nextgens> no need to graph it :)
[14:18] <nextgens> 9143 for 08
[14:18] <nextgens> 9109 for 09
[14:18] <nextgens> 9032 for 10
[14:18] <nextgens> 7950 for 11
[14:18] <nextgens> 4844 for 12
[14:18] <nextgens> still, we have to keep in mind that those stats are for win32 nodes only
[14:19] <toad_> okay, so it declines
[14:19] <toad_> that's more or less what I expect :(
[14:20] <nextgens> stats for december are obviously lower as the month isn't finished
[14:20] <toad_> well yeah but you can guesstimate
[14:20] <toad_> 4844 * 31/18 = 8342
[14:20] <toad_> hmm that would be respectable if it was the case
[14:21] <nextgens> ;)
[14:21] <toad_> and it may well be
[14:21] <nextgens> doing stats over a longer period is really much more work
[14:21] <toad_> lots of ppl have time off over xmas
[14:21] <nextgens> in fact I'm doing stats since I've deployed the new installer
[14:22] <nextgens> toad_> I don't :)
[14:22] <nextgens> lucky them
[14:22] <toad_> well if you could do an earlier period that would be great
[14:22] <toad_> but ok
[14:22] <nextgens> I could but that wouldn't be very helpfull imo
[14:22] <toad_> ok the question before us is whether to do opennet and what the timing would be
[14:23] <toad_> the argument against / for delaying i can roughly sum up as "we should give fast darknet time to see if people will actually use it and virally grow it"
[14:23] <toad_> right?
[14:23] <nextgens> yes, that's part of it
[14:23] <toad_> this does mean limiting non-viral growth by not providing an opennet
[14:24] <toad_> and hoping that viral growth outstrips non-viral growth
[14:24] <toad_> nextgens: what's the other part?
[14:24] <nextgens> we can't reasonably deploy opennet until we have load-balancing sorted
[14:25] <toad_> sure
[14:25] <nextgens> so what's the aim of doing it before it's sorted out ?
[14:25] <toad_> that doesn't mean we can't get it ready beforehand
[14:25] <toad_> so that we can deploy both at the same time
[14:25] <toad_> which is what ian wants to do
[14:25] <nextgens> and break everything at once :)
[14:25] <toad_> obviously that means we have no cooldown time; we are relying 100% on simulations
[14:25] <nextgens> so that we are 100% sure not to be able to compare the old and new load-balancing algorithm
[14:25] <toad_> i would prefer to deploy opennet after we KNOW that load limiting works
[14:26] <nextgens> ;)
[14:26] <toad_> but to oppose ian i need something stronger than "I would prefer"
[14:26] <nextgens> I would prefer not to deploy opennet ... but I guess you already know that :p
[14:26] <toad_> well, we may as well examine that option
[14:27] <toad_> if we deploy opennet, it'll come with warnings saying get yourself some real darknet peers for security's sake
[14:27] <toad_> and many people will
[14:27] <toad_> it will almost certainly result in a larger network in the short run
[14:27] <nextgens> suppose the new load-balancing algorithm works ... and freenet grows virally without as we have achieved the "speed" goal; would openent still be needed ?
[14:27] <toad_> otoh in the long run viral growth may be much less than in the non-opennet case
[14:27] <nextgens> maybe because it will lack content
[14:28] <toad_> hmm?
[14:28] <nextgens> there is something we never consider : how will the churn be handled ?
[14:28] <toad_> yeah, we need to prototype it all in simulation, we have agreed this
[14:29] <toad_> i'm not sure i can defend the never-implement-opennet case, that's the problem
[14:29] <nextgens> ian said he has already simulated both in high level simulations
[14:29] <toad_> as ian says, people will find ways to do opennet
[14:29] <toad_> the most obvious one being #freenet-refs
[14:29] <toad_> which greatly reduces our adoption
[14:30] <toad_> nextgens: ian hasn't simulated load limiting
[14:30] <nextgens> toad_> I mean "hybrid net" : darknet and opennet cohabitation
[14:31] <toad_> nextgens: he has not simulated load limiting/balancing's interactions with opennet
[14:31] <toad_> which are a concern
[14:31] <nextgens> well, anyway we shouldn't spend too much time debating/argueing about it
[14:31] <toad_> what he has shown is that swapping can coexist with path folding
[14:31] <toad_> nextgens: i think we need to
[14:31] <nextgens> how long would it take to hack something up ?
[14:31] <toad_> nextgens: otherwise i will have to do what ian says since he's my boss
[14:31] <nextgens> then do it :)
[14:32] <toad_> well what he says is to implement opennet
[14:32] <nextgens> I do think the debate ought to take place on "when do we deploy opennet; hybridnet or separated networks" ?
[14:32] <toad_> you mean when we do?
[14:32] <toad_> that debate has been had, and is over
[14:32] <toad_> it has to be a hybrid
[14:33] <toad_> otherwise there won't be any content on one or other network
[14:33] <toad_> probably being the darknet
[14:33] <nextgens> what I mean is that both questions are related
[14:33] <nextgens> I'm not sure about that
[14:33] <toad_> also i don't think we can keep them separate if we wanted to
[14:33] <nextgens> how so ?
[14:34] <nextgens> just make them incompatible enough :p
[14:34] <toad_> well we can keep them MOSTLY separate
[14:34] <toad_> but we would want to have key compatibility
[14:34] <toad_> so people can migrate content
[14:34] <toad_> using binary blobs
[14:34] <nextgens> ok, but that wouldn't hurt routing/the network/load balancing
[14:34] <nextgens> so it's not an issue
[14:35] <toad_> no but hacked nodes on both networks would
[14:35] <toad_> better that we make explicit provision for a hybridnet, then we can have content on both
[14:35] <nextgens> well, if the api is incompatible enough, noone is gonna do it :)
[14:35] <toad_> also more importantly opennet can link all the darknets together
[14:35] <nextgens> the problem with hybridnet is "when shall we deploy it"
[14:35] <toad_> and it can form a pool to increase the chance that you know somebody on freenet and therefore can have a darknet connection
[14:35] <nextgens> not before load balancing is sorted out : we agreed on that
[14:36] <toad_> well yes but how do you determine when load balancing is sorted out?
[14:36] <nextgens> but shouldn't we give some time to the darknet to settle once done ?
[14:36] <nextgens> so that we can compare before/after
[14:36] <toad_> ian would argue that we should simulate it, implement it, and deploy both at once
[14:36] <toad_> otherwise i really don't see any urgency in implementing opennet
[14:37] <toad_> from the other end ...
[14:37] <nextgens> well, if we do, we won't be able to evaluate the new load-balancing policy on the same grounds
[14:37] <nextgens> the only thing we will have is simulations
[14:37] <toad_> if i don't implement opennet, is there anything i can do to speed up implementation of the new load balancing algorithm?
[14:38] <toad_> i can ask jflesch again to write up the middle layer (congestion control)
[14:38] <toad_> and then implement the bottom layer, middle layer and STS all at once
[14:38] <Zothar_Work> from my reading of logs and yesterday, the urgency seemed to be because of refbot's inter-bot ref trading, but that doesn't seem to be used much at the moment; it seems that post-inter-bot ref trading doesn't get used any more than pre-inter-bot ref trading; usually there's a few extra bots running right after it's promoted/hyped and then it drops to zero/near zero; the only refbot.py on...
[14:38] <Zothar_Work> ...#freenet-refs ATM was around before inter-bot ref trading
[14:39] <nextgens> neither opennet nor load balancing have been simulated enough, right ? so why is implementing opennet now more sensible that implementing an unfinished draft of the new low-balancing ?
[14:40] <nextgens> *than
[14:40] <toad_> nextgens: the low-level stuff has no direct effect but is convenient to implement with the middle level; the middle level has been extensively simulated
[14:40] <nextgens> ok, then I suggest you implement the middle level *now*
[14:40] <toad_> nextgens: okay, so the argument to put to ian is that we don't WANT to deploy opennet and load balancing at the same time
[14:40] <nextgens> on the basis that : 1) it has been simulated 2) it will be required for opennet anyway
[14:41] <nextgens> toad_> yes, that's how I see it
[14:41] <Zothar_Work> ian was lobbying me yesterday to implement opennet, but suggested I wait after finding that I'll be on vacation for most of the rest of the year; I've looked at the wiki page, but there are a lot of terms on there I don't understand yet; I dunno if I want to undertake opennet anyway as I don't plan to use it; I can see how things are in January I suppose
[14:41] <toad_> ugh
[14:42] <Zothar_Work> yeah, sounds like implementing the middle level now is good since it's the only thing that is completely simulated and why wait; we should get 1008 released first though I think
[14:43] <Zothar_Work> (I'll be off work 6 hours or so)
[14:43] <nextgens> Zothar_Work> they are bugs that need to be addressed before we do
[14:43] <toad_> yeah we need to think about how to exit from the plugins changes if I'm going to abandon them
[14:43] <toad_> i'll have to revert all my API changes to make the old plugins work again :(
[14:43] <Zothar_Work> nextgens: extactly; I'm saying we should sort those out and release 1008 before middle level, which should possibly be next
[14:43] <toad_> or just plough ahead with it
[14:43] <nextgens> I suggest you don't abandon them
[14:44] <toad_> ian wants me to
[14:44] <nextgens> :(
[14:44] <Zothar_Work> can they just be
[14:44] <Zothar_Work> uh
[14:44] <nextgens> a new plugin (freenetwiki) will have to be ported then :/
[14:44] <Zothar_Work> can they just be swapped out with the old API and then swapped back in later?
[14:44] <toad_> nextgens: there are good reasons for changing plugins
[14:45] <toad_> nextgens: however none of them are more important than opennet
[14:45] <nextgens> toad_> how many hours are we talking about now ?
[14:45] <nextgens> how long would it take to finish "plugins" ?
[14:45] <toad_> i suspect i can get the new plugins basically sorted out in a few days
[14:45] <toad_> ideally Bombe would do it
[14:45] <nextgens> Bombe> around ?
[14:46] <toad_> sadly Bombe hasn't been around since we last talked about plugins
[14:46] <toad_> which was early last week :<
[14:46] <nextgens> ok, we are the 18. of december today
[14:46] <Zothar_Work> to me, if opennet's implementation depends on low level/middle level stuff, then it seems we'd want those two implemented before opennet, so opennet can't trump yet it seems to me
[14:46] <Bombe> Yeah, my life keeps interfering. :)
[14:46] * toad_ would like to take some time off for christmas of course
[14:46] <Bombe> Sometimes I prefer lurking here, though. ;)
[14:46] <nextgens> if you start working on opennet, what/when will be done before next year ?
[14:46] <toad_> Zothar_Work: opennet depends ultimately on the new load balancing system
[14:46] <nextgens> toad_> hence I ask :)
[14:47] <toad_> Zothar_Work: the problem is I am not personally involved very much in that
[14:47] <toad_> Zothar_Work: and it's not progressing as fast as we might like
[14:47] <Zothar_Work> implementation or deployment? If implementation, then it seems opennet can't be implemented even if we don't deploy yet
[14:47] <toad_> Zothar_Work: so i work on stuff like plugins
[14:47] <nextgens> now you know what to do with plugins ... opennet is yetAnotherArea of the code
[14:47] <toad_> Zothar_Work: but that pisses off ian
[14:47] <toad_> Zothar_Work: for the reason that I should be working on opennet, so that I can deploy it as soon as load balancing is deployed
[14:48] <nextgens> would anything usefull be done before christmas if you were working on opennet ?
[14:48] <toad_> if I implement opennet before xmas then ian will deploy it while i am away, and several devs will leave the project
[14:48] <toad_> :)
[14:48] <toad_> okay i do trust ian a bit more than that
[14:48] <Zothar_Work> perhaps you need to be able to demonstrate the opennet dependency on load balancing; I'm not convinced opennet should be implemented now either
[14:49] <toad_> certainly there are useful things I could do
[14:49] <nextgens> toad_> but to you think you can implement it before hollidays ? :)
[14:49] <toad_> not fully no
[14:49] <nextgens> if so, by any means, do it
[14:49] <nextgens> ;)
[14:49] <Zothar_Work> (inter-bot ref trading hasn't grown as big as I think/thought it might yet, so it's not a "threat" yet)
[14:49] <nextgens> ok, I'm getting back to *real work*
[14:50] <nextgens> lurk me if needed
[14:50] <nextgens> toad_> have a look to mantis if you want to do a new release :)
[14:51] <toad_> :)
[14:52] <toad_> i'm writing an email to ian
[14:52] <nextgens> Cc me please :p
[14:52] <Zothar_Work> heh, yeah, that could be an interesting email :)
[14:56] <toad_> bbiab
[14:57] <toad_> Bombe: will you be able to spend some time on plugins over xmas or not?
[14:57] <toad_> Bombe: if it's with me it needs to be soon; tomorrow would be good for example
[14:58] <Bombe> Probably. I'll be sitting at home with not much to do.
[14:58] * Apophis2 (n=Apophis@) has joined #freenet
[14:58] <toad_> tomorrow, this evening
[14:58] <toad_> maybe wed
[14:58] <sopues> "fast darknet" ?
[14:58] <sopues> is that already here or....
[14:59] <sopues> should I upgrade :)
[14:59] <toad_> somebody explain to sopues :)
[14:59] <toad_> bbiab
[14:59] <toad_> back ~ 3 hours
[15:00] <toad_> bye
[15:01] <Zothar_Work> sopues: what are you running right now? Is it 0.5 or 0.7 or nothing?
[15:04] <sopues> 0.7
[15:04] <sopues> Freenet 0.7 Build #1007 r11372
[15:05] <Zothar_Work> where did you hear the term "fast darknet"?
[15:06] <sopues> at 15:22 CET: http://pastebin.ca/283688
[15:06] <sopues> I assume toad means with working load limiting
[15:06] <Zothar_Work> oh, that's talking about yet to be implemented stuff; 1007 is still the latest
[15:06] <sopues> ok
[15:07] <sopues> I heard there are newer build that are very broken. Is this still true for latest svn
[15:07] <Zothar_Work> that's testing builds
[15:07] <Zothar_Work> there are bugs preventing 1008 from being released, yes
[15:07] <sopues> ok
[15:16] * Apophis2 (n=Apophis@) Quit (Read error: 110 (Connection timed out))
[15:20] <sopues> it would be insanely cool with store/cache histograms in the webinterface ;)
[15:22] <Zothar_Work> sopues: well, code it and they'll be there :)
[16:12] * wkatcar9 (i=wkatcar9@) has joined #freenet
[16:12] * wkatcar9 (i=wkatcar9@) has left #freenet
[16:15] * railk (n=railk@) has joined #freenet
[16:30] * sanity (n=ian@) has joined #freenet
[16:30] * ChanServ sets mode +o sanity
[16:31] * sanity (n=ian@) Quit (Client Quit)
[16:48] * MikeW (n=em@) has joined #freenet
[17:01] * kinzite_ (n=kinzite@) has joined #freenet
[17:08] * lattt (n=haha@) has joined #freenet
[17:14] * kinzite (n=kinzite@) Quit (Read error: 110 (Connection timed out))
[17:26] * lattt (n=haha@) Quit ("Verlassend")
[17:34] * Urs_ShPo (n=gaim@) Quit ("Leaving.")
[17:43] <sandos> Zothar_Work, hehe true
[17:47] * egon2003 (n=fargod@) has joined #freenet
[17:50] * PraiseChaos (n=kcecil@) Quit ("Leaving")
[17:50] * PraiseChaos (n=kcecil@) has joined #freenet
[17:51] * PraiseChaos (n=kcecil@) Quit (Read error: 104 (Connection reset by peer))
[17:53] * PraiseChaos (n=kcecil@) has joined #freenet
[17:54] * railk_ (n=railk@) has joined #freenet
[17:58] * PraiseChaos (n=kcecil@) Quit (Remote closed the connection)
[17:59] * railk (n=railk@) Quit (Nick collision from services.)
[17:59] * railk_ is now known as railk
[18:01] * sanity (n=ian@) has joined #freenet
[18:01] * ChanServ sets mode +o sanity
[18:02] * railk_ (n=railk@) has joined #freenet
[18:05] * Yukishiro (n=zhaan@) has joined #freenet
[18:06] * maelztrom (i=nobody@) Quit (Read error: 110 (Connection timed out))
[18:10] * railk (n=railk@) Quit (Read error: 145 (Connection timed out))
[18:13] * railk_ is now known as railk
[18:34] * Apophis2 (n=Apophis@) has joined #freenet
[18:42] * sandos (n=sandos@) Quit (Connection timed out)
[18:42] * sandos (n=sandos@) has joined #freenet
[18:43] * DebolazY (n=DebolazX@) has joined #freenet
[18:56] * Fennesniehier is now known as Fennes
[19:00] * DebolazX (n=DebolazX@) Quit (Connection timed out)
[19:09] * MikeW (n=em@) Quit (Read error: 110 (Connection timed out))
[19:10] * attica (i=attic@) has joined #freenet
[19:13] * Fennes is now known as Fenneszz
[19:28] * attic (i=attic@) Quit (Read error: 110 (Connection timed out))
[19:28] * Yukishiro (n=zhaan@) Quit ("Chatzilla 0.9.75 [Firefox 1.5.0.8/2006102516]")
[19:33] * attic (i=attic@) has joined #freenet
[19:34] * attica (i=attic@) Quit (Read error: 54 (Connection reset by peer))
[19:36] * attica (i=attic@) has joined #freenet
[19:36] * attic (i=attic@) Quit (Read error: 104 (Connection reset by peer))
[19:37] * DebolazY is now known as DebolazX
[19:38] * attic (i=attic@) has joined #freenet
[19:50] * JCTV (i=JJK@) has joined #Freenet
[19:52] * attic (i=attic@) Quit (Read error: 60 (Operation timed out))
[19:52] * attic (i=attic@) has joined #freenet
[19:57] * attica (i=attic@) Quit (Read error: 110 (Connection timed out))
[20:07] * Randan (n=Randan@) has joined #freenet
[20:13] * JCTV (i=JJK@) Quit (Client Quit)
[20:15] * Andross (n=drew@) has joined #freenet
[20:29] * egon2003 (n=fargod@) Quit (Read error: 113 (No route to host))
[20:31] * MikeW (n=em@) has joined #freenet
[20:34] <toad_> nextgens: you around?
[20:43] * fasta (n=fasta@) has joined #freenet
[20:58] * NullAcht15 (n=NullAcht@) has joined #freenet
[20:58] * Andross (n=drew@) Quit (Client Quit)
[20:58] * whiterabbit (n=Whiterab@) Quit ("KVIrc 3.2.5 Anomalies http://www.kvirc.net/")
[21:14] <fasta> Where can I find the Peer Location Distribution map?
[21:15] <toad_> fasta: hmmm?
[21:15] <toad_> fasta: telnet port 2323 and do PROBEALL:
[21:17] <fasta> toad_: What does that do?
[21:17] <toad_> fasta: what are you trying to do?
[21:17] <toad_> "peer location distribution map" ?
[21:17] <fasta> toad_: Someone posted that on Frost.
[21:18] <toad_> well, PROBEALL:
[21:18] <toad_> will produce, to wrapper.log, a list of locations of nodes on the network
[21:18] <fasta> toad_: I don't have the PROBEALL documented
[21:18] <toad_> no, it's undocumented :)
[21:18] <fasta> heh
[21:26] <toad_> it takes a long time
[21:26] <toad_> but i'd be interested in the results
[21:27] <fasta> What exactly does it do?
[21:27] <toad_> nextgens: Bombe: please tell me if you appear
[21:27] <fasta> toad_: Is there a way to uniquely refer to a message on frost, btw?
[21:28] <toad_> Bombe: we need to talk plugins; I need you to finish them, basically
[21:28] <toad_> Bombe: tell me when you're on
[21:29] <toad_> nextgens: I'm going to have to support opennet; #freenet-refs appears to produce a very poor network topology
[21:30] <toad_> Bombe: I am able to put in a reasonable amount of help, but I can't do all the work ... I probably can't even do most of the work
[21:31] <nextgens> toad_> _o/
[21:31] <toad_> well
[21:32] <toad_> that's one explanation of the wierd clustering behaviour
[21:32] <nextgens> any proof it's the only one ?
[21:32] <toad_> well do you really think the 0.7 network is only 300 nodes large?
[21:32] <toad_> It could also be just that the probe requests is still just biased to
[21:32] <toad_> see nodes in your neighborhood.
[21:32] * lattt (n=gerhard@) has joined #freenet
[21:32] <toad_> => degenerate routing
[21:32] <nextgens> at the time I've done my probes, it's quite possible
[21:33] <toad_> i can't see any problem with routing; it implies degenerate network topology IMNSHO
[21:33] <nextgens> it was late sunday afternoon
[21:33] <Zothar_Work> fasta: perhaps you're asking about the "Peer Location Distribution" box on the Darknet page of testing builds in advanced mode
[21:33] <nextgens> and btw, how comes opennet would produce a better one ?
[21:33] <toad_> nextgens: do you think we could get away with exposing the ENTIRE network topology?
[21:34] <toad_> nextgens: opennet would produce a much better topology because opennet is known to produce a small world link distribution
[21:34] <toad_> oskar's done a load of work on that
[21:34] <nextgens> sure we could ... but the whole point of darknet is not to
[21:34] <fasta> toad_: I see >500 nodes now, btw.
[21:34] <toad_> nextgens: I mean for debugging purposes
[21:34] <nextgens> that would be handy ...
[21:34] <nextgens> but we have to warn users somehow
[21:34] <nextgens> no ?
[21:35] <toad_> we ask a node for H(identity+"darknet"), its own location, its last 10 locations, and its peers
[21:35] <toad_> i suppose
[21:35] <toad_> if most of them are connected via #freenet-refs then we're not really losing any security :)
[21:35] <toad_> even if they're not, i'm not sure it's a big deal
[21:35] <toad_> you can scan the network topology from swapping already
[21:35] <nextgens> may you explain to me what the "size estimate" is on probe requests ?
[21:35] <toad_> sure
[21:35] <nextgens> does it come from the target node ?
[21:35] <toad_> no
[21:36] <nextgens> ok, I'll RTFS then :)
[21:36] <toad_> it's simply the number of locations seen so far divided by the highest location so far
[21:36] <toad_> on the theory that locations are distributed evenly across the keyspace
[21:36] <nextgens> wich isn't the case ;)
[21:37] <toad_> the thing is, either they aren't, AND the network is tiny, or they are, but the network topology is rubbish and we can't see the nodes further away from our location
[21:37] * nextgens reads his mails
[21:37] <toad_> don't reply to what i sent to ian yet
[21:37] <toad_> tell me first what you're gonna say...
[21:38] <nextgens> I won't anyway
[21:38] <toad_> you know what we could do?
[21:39] <toad_> we could just change the location swap requests to include H(identity+<some constant>) for the swapper and for each node ...
[21:39] <toad_> brb
[21:40] <nextgens> as we don't have update over mandatory, it's gonna take a month before we can use it :/
[21:40] <nextgens> use results of the change I meant
[21:43] <toad_> actually i know what i should do to chase this
[21:44] <toad_> spy on other people's swap requests!
[21:44] <toad_> that can be done now
[21:45] <nextgens> ?
[21:45] <toad_> no mandatory changes needed, no more info given to attackers
[21:45] * greycat (i=rfc1413@) Quit ("This time the bullet cold rocked ya / A yellow ribbon instead of a swastika")
[21:45] <toad_> but we'd have a much more accurate estimate of network size
[21:45] * nextgens mandates toad_ to spy on people
[21:45] <nextgens> you're sill missing your NSA badge though :)
[21:46] <toad_> nextgens: most swap requests we are involved in aren't directed to us
[21:46] <toad_> they're forwarded to a node behind us
[21:46] <toad_> we can however decode them and see the locations involved
[21:46] * MikeW (n=em@) Quit ()
[21:47] <nextgens> we could add them to a hashmap, like we do for networkSizeEstimate, right
[21:47] <toad_> at the moment our size estimate (on the stats page) is based only on swaps involving us
[21:47] <toad_> right
[21:47] <toad_> ok, that's the next job
[21:47] <toad_> should yield pretty fast results too
[21:47] <nextgens> ok, it seems sensible to do it :)
[21:48] <Randan> node/wrapper crash: http://dark-code.bulix.org/n1gnqs-26555?raw
[21:48] * SQFreak (n=SQ@) has joined #freenet
[21:50] <Bombe> toad, I'm here now.
[21:51] <nextgens> Randan> you aren't using sun's jvm
[21:51] <nextgens> Randan> are you ?
[21:51] <Randan> I am
[21:51] <nextgens> please paste the result of "java -version"
[21:52] <toad_> Bombe: hi
[21:52] <Randan> just noticed that I still run the old java 1.6 beta...
[21:52] <toad_> Bombe: how much time will you have to work on plugins over xmas?
[21:52] * Randan will update and report if it happens again
[21:52] <toad_> Bombe: the situation is this:
[21:53] <Bombe> toad, pretty much... I'll start my 5-day party marathon on 27th.
[21:53] <toad_> Bombe: I will be taking off some time over xmas. Ian wants me to implement opennet immediately. I've done a lot of work on plugins, including some refactoring _which breaks existing plugins_
[21:54] <toad_> Bombe: we can auto-migrate if we delay 1008 until the plugin API has been sorted out and implemented, including a new deployment system
[21:54] <Randan> nextgens: still strange, it ran for days with that vm (1.6.0-beta2-b86)
[21:54] <toad_> Bombe: I think I can justify writing the deployment system myself, but I can tell you how to do it if you want
[21:54] <NullAcht15> what? I always thought you were long doing that anyway ("spying" on other people's swap requests for size estimates)
[21:55] <toad_> s/a lot/a bit :)
[21:55] <Bombe> toad, deployment for what? Plugins?
[21:56] <Bombe> brb
[21:57] <toad_> yes
[21:57] <toad_> deployment of plugins
[21:57] <toad_> I want to move them to auto-update-over-freenet
[22:01] <toad_> if that's too hard I suppose we'll have to ask the user whether they want to redownload
[22:01] <toad_> and if THAT'S too hard I suppose we'll have to just tell them to reinstall their plugins :|
[22:01] * SQFreak is now known as SQFreak_bot
[22:02] <toad_> NullAcht15: no, it's one of 800 odd unresolved bugs
[22:02] <nextgens> toad_> have you received my mail about PROBEALL ?
[22:02] <toad_> what did it say? just a probeall trace with most of the locs near 0.0 or 1.0?
[22:02] <nextgens> yes
[22:03] * SQFreak_bot is now known as SQFreak
[22:04] <CIA-14> jflesch * r11472 /trunk/apps/Thaw/src/thaw/core/ (Core.java MainWindow.java NodeConfigPanel.java): Woops forgot to save the state of the option 'thaw and the node are on the same computer' in the configuration
[22:06] <CIA-14> toad * r11473 /trunk/freenet/src/freenet/node/LocationManager.java: Spy on other people's swaps that go through us. Should result in much faster and more accurate size estimates.
[22:07] <nextgens> # networkSizeEstimateSession: 737 nodes
[22:07] <nextgens> # networkSizeEstimate24h: 549 nodes
[22:07] <nextgens> # networkSizeEstimate48h: 626 nodes
[22:07] <nextgens> it's weird that probe requests aren't showing more :/
[22:07] <toad_> nextgens: that's with the old code
[22:07] <toad_> right
[22:07] <toad_> it suggests a problem with routing
[22:07] <nextgens> or the request code
[22:08] <toad_> which means either a) a bug in routing, b) a bug in routing/processing probe requests, or c) a degenerate network topology
[22:08] <nextgens> raaah
[22:09] <nextgens> it has dropped my global queue :((
[22:09] <toad_> eh?
[22:09] <toad_> hmm my fault
[22:10] <nextgens> what have you done ?
[22:11] * Zothar_Work is now known as Zothar
[22:15] * nextgens wonders why @cvs is so laggy
[22:18] <CIA-14> zothar * r11474 /trunk/apps/pyFreenet/ (minibot.py refbot.py):
[22:18] <CIA-14> refbot: Hopefully better handle a DNS error looking up the IRC server hostname.
[22:18] <CIA-14> Fix usage of the pinger to keep the watchdog happy. Check the identity of a bot
[22:18] <CIA-14> traded ref after we know we have an identity field. Hopefully actually fix the
[22:18] <CIA-14> refsperrun count bug in inter-bot ref trading.
[22:22] <toad_> nextgens: ok, i know what the problem is, i will fix it
[22:22] <toad_> after dinner :)
[22:22] <toad_> bbiab
[22:22] * railk (n=railk@) Quit ("Leaving")
[22:29] * electrosys (n=steo@) has joined #freenet
[22:29] <electrosys> Hia.
[22:33] * fasta (n=fasta@) Quit ("Lost terminal")
[22:42] * SQFreak (n=SQ@) has left #freenet
[22:46] * rebo123 (n=rebo123@) Quit (Read error: 145 (Connection timed out))
[22:48] <CIA-14> zothar * r11475 /trunk/freenet/src/freenet/clients/http/ (DarknetConnectionsToadlet.java StatisticsToadlet.java): Move Peer Location Distribution circle to /stats/ page
[23:00] * NullAcht15 (n=NullAcht@) Quit (Remote closed the connection)
[23:04] * mozillaman (n=self@) has joined #freenet
[23:07] * TheBishop_ (n=bishop@) Quit (Read error: 131 (Connection reset by peer))
[23:08] * TheBishop_ (n=bishop@) has joined #freenet
[23:13] * blimeyboy (n=chatzill@) has joined #freenet
[23:17] <nextgens> toad_> did you get my mail about the php warning on the donation webpage ?
[23:18] <CIA-14> zothar * r11476 /trunk/freenet/src/freenet/ (3 files in 2 dirs): Reduce precision of positioning of peer marks on Peer Location Distribution circle (do we even need hundreths of a pixel?). Clean a few XHTML validation issues.
[23:20] * electrosys (n=steo@) Quit ()
[23:33] <toad_> <li>networkSizeEstimateSession: 400 nodes</li>
[23:33] <toad_> <li>nodeUptime: 1h30m</li>
[23:33] <toad_> hmmm
[23:34] <toad_> i'd expected it to be higher
[23:36] * dickie (n=pirch@) has joined #freenet
[23:44] <nextgens> toad_> did you get my mail about the php warning on the donation webpage ?
[23:49] <nextgens> bbl
Irc logs of #freenet : 2008 2007 2006 2005
These logs were automatically created by FreenetLogBot on chat.freenode.net using the Java IRC LogBot.