Timestamps are in GMT/BST.
[0:46] * mikeDOTd (~mikedotd@12.30.47.94) Quit (Read error: 54 (Connection reset by peer))
[1:02] * goatee (~goatee@ip216-239-84-102.vif.net) Quit (Read error: 110 (Connection timed out))
[1:46] * mikeDOTd (~mikedotd@plutonium.nexthop.net) has joined #freenet
[3:37] * yonkeltron (~yonkeltro@pcp04664407pcs.wilog501.pa.comcast.net) has joined #freenet
[3:37] * yonkel-cluster (~yonkeltro@pcp04664407pcs.wilog501.pa.comcast.net) has joined #freenet
[3:38] * cStyle (~cstyle@host-23-236-220-24.midco.net) Quit ()
[4:23] * briantd (~brian@dsl027-183-057.sfo1.dsl.speakeasy.net) has joined #freenet
[4:25] * briantd (~brian@dsl027-183-057.sfo1.dsl.speakeasy.net) has left #freenet
[4:25] * Kyon (~Kyon@134.53.120.239) has joined #freenet
[4:31] * Hory (~Miranda@81.196.25.110) has joined #FreeNet
[5:39] * d-ArkAngel (~Rob@213-131-104-86.onyx.net) has joined #freenet
[5:39] <d-ArkAngel> Morning all
[5:41] <ShaunMacPherson> hi
[6:41] * tyesinclair (~tyesincla@170.252.64.1) has joined #freenet
[6:41] * tyesinclair (~tyesincla@170.252.64.1) has left #freenet
[6:49] * plixed (~plixed@pD9551513.dip.t-dialin.net) has joined #freenet
[7:21] * kers (~kers@231.ppp132.rsd.worldonline.se) has joined #freenet
[9:46] * FallingBuzzard (~srademach@207.152.112.129) has joined #freenet
[10:01] * 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!")
[10:02] * Digger3 (~digger3@kf-dn-tg01-174.dial.kabelfoon.nl) Quit (Read error: 104 (Connection reset by peer))
[10:04] * Digger3 (~digger3@kf-dn-tg01-174.dial.kabelfoon.nl) has joined #freenet
[10:20] * toad_ (toad@82-32-16-91.cable.ubr03.azte.blueyonder.co.uk) has joined #freenet
[10:21] * Digger3_ (~digger3@kf-dn-tg01-174.dial.kabelfoon.nl) has joined #freenet
[10:22] * Digger3 (~digger3@kf-dn-tg01-174.dial.kabelfoon.nl) Quit (Read error: 104 (Connection reset by peer))
[10:22] * kers (~kers@231.ppp132.rsd.worldonline.se) Quit (Read error: 232 (Connection reset by peer))
[10:23] <toad_> hi ppl
[10:23] <d-ArkAngel> hi toad
[10:23] * moskawa23 (~Miranda@dsl-213-023-252-083.arcor-ip.net) has joined #freenet
[10:23] * moskau23 (~Miranda@dsl-213-023-252-083.arcor-ip.net) Quit (Read error: 232 (Connection reset by peer))
[10:24] <d-ArkAngel> how things with you this fine.. well afternoon here :-)
[10:26] <toad_> afternoon here too :)
[10:26] <toad_> fair
[10:26] <toad_> feeling inexplicably tired, possibly i have Yet Another Fscking Chest Infection
[10:26] <toad_> dad's toe is now much better, but both parents seem to have some chest problems lately
[10:26] <d-ArkAngel> I've got to say I've still not completed the threaded simulator, and I doubt I'll get much time to work on it this evening, I'm going out for a meal for my mothers brithday.
[10:27] <toad_> mum's site has got 2,855 page hits since the first of october
[10:27] <toad_> and an average of 60MB/day over the last week
[10:27] <d-ArkAngel> mum's site?
[10:28] <toad_> http://www.simechoes.org/
[10:28] <toad_> hosted on our cable modem
[10:28] <toad_> after they changed the T&C to allow hosting web sites
[10:28] * plixed is now known as plixed_away
[10:28] <toad_> sims fan-site
[10:29] <d-ArkAngel> :-)
[10:29] <toad_> she was using various mostly subscriber-pays hosting options
[10:29] <d-ArkAngel> nice.
[10:29] <toad_> maybe one day she can use freenet
[10:29] <toad_> a LOT of sims sites go down because of bandwidth costing too much
[10:30] * moskawa23 (~Miranda@dsl-213-023-252-083.arcor-ip.net) Quit (Read error: 104 (Connection reset by peer))
[10:30] <d-ArkAngel> my main web presence has had 4,199,858 hits since 1st oct ;-)
[10:30] <toad_> nice
[10:30] <toad_> that's profit making?
[10:31] <d-ArkAngel> it breaks even
[10:31] * moskau23 (~Miranda@dsl-213-023-252-083.arcor-ip.net) has joined #freenet
[10:31] <toad_> advertising funded or actual ebiz?
[10:31] <d-ArkAngel> adverts
[10:31] <toad_> that's the blizzard thingy?
[10:31] <d-ArkAngel> yeah
[10:31] * toad_ thinks that if we ever get freenet REALLY working there's a big potential in free unlimited bandwidth hosting...
[10:32] <toad_> which brings me back to... IT DOESN'T WORK, even in simulation!
[10:33] <toad_> d-ArkAngel: that's page hits, not total hits, btw?
[10:33] <d-ArkAngel> oh yeah, just page hits
[10:36] <toad_> nice
[10:36] <toad_> what's your connection?
[10:36] <d-ArkAngel> it's a dedicated server (well two dedicated servers) hosted in the US
[10:36] <d-ArkAngel> 100Mbit burstable, up to 1.5Tb per month
[10:37] <toad_> what do you actually transfer in a month?
[10:38] <d-ArkAngel> 350-500Gb
[10:41] <d-ArkAngel> lol, I had 1289 hits from the first untill last night on my cable modem hosted site, and then 7615 hits since midnight last night :-)
[10:41] <d-ArkAngel> I think someone on Blizzhackers has linked to my webcam by the look of the number of hits there has been :-)
[10:41] <toad_> what time is it now?
[10:42] <d-ArkAngel> it's 2pm
[10:42] <d-ArkAngel> well 2:40pm
[10:42] <toad_> 3 hours after midnight?
[10:42] <toad_> or 2:40 in the afternoon
[10:42] <d-ArkAngel> no, as in 2pm in the afternoon ;-) I'm in the UK too ;-)
[10:43] <d-ArkAngel> 14:40 if that's the way you like it.
[10:45] * toad_ downloads current simulation results...
[10:45] <toad_> is there anything on /. in the last two days to justify my checking the headline list emails they sent me?
[10:46] <d-ArkAngel> I've given up reading /. to be honest
[10:46] <d-ArkAngel> There's probably something about the whole Kodak vs Java thing, that I mentioned on the freenet-dev list
[10:46] <toad_> i don't read the threads usually
[10:46] <ShaunMacPherson> Let me see
[10:46] <ShaunMacPherson> Yes
[10:46] <ShaunMacPherson> Java is underattack by Kodak :)
[10:46] <ShaunMacPherson> Java/SUn
[10:47] <toad_> ShaunMacPherson: i heard that from the devl list
[10:47] <toad_> anything else?
[10:47] <ShaunMacPherson> Hmmmmm... nothign else then heh
[10:47] <toad_> okay, why has my download-simlog script not said anything yet? it's been running for a while...
[10:48] <toad_> the first line that does anything rsync's something from the dp opteron server...
[10:48] <toad_> ssh: connect to host 217.16.18.124 port 22: Connection timed out
[10:48] <toad_> rsync: connection unexpectedly closed (0 bytes read so far)
[10:48] <toad_> rsync error: error in rsync protocol data stream (code 12) at io.c(343)
[10:48] <toad_> arrgh
[10:48] <toad_> THAT is REALLY annoying
[10:48] <toad_> almost as annoying as the hoops we jumped through yesterday to avoid buying a licensed dvd player ;|
[10:49] <toad_> (exacerbated greatly by driver issues with ati cards)
[10:50] <d-ArkAngel> I've been thinking about swapping my ATI out for a matrox (a very old one, 2D only)
[10:50] <toad_> heh
[10:50] * Digger3_ is now known as Digger3
[10:50] <toad_> i need 3d
[10:50] <toad_> well i want 3d ;)
[10:50] <toad_> but i'll probably get an nvidia next time
[10:50] <d-ArkAngel> yeah, but software OGL would be faster than what my ATI does at the moment :-)
[10:50] <toad_> having open drivers that more or less work is nice, but getting tv-out is a PITA
[10:50] <toad_> d-ArkAngel: what is it, mach64?
[10:51] <toad_> well
[10:51] <toad_> the open drivers work a lot better than the closed ones for my R9K - except for tv-out issues
[10:51] <d-ArkAngel> nope, it's a Millenuim 2 ;-)
[10:51] <toad_> d-ArkAngel: that's a matrox card
[10:51] <d-ArkAngel> or do you mean the ATI? that's a damn R300
[10:51] <toad_> what's the ati?
[10:51] <toad_> ah ok
[10:52] <toad_> what's the problem with the r300?
[10:52] <toad_> it should be pretty fast with the closed drivers
[10:52] <d-ArkAngel> no open source drivers.
[10:52] <d-ArkAngel> No closed drivers that work with the x86_64 kernel
[10:52] <toad_> ahhh :)
[10:52] * lima (~unknown@195.226.109.66) has joined #freenet
[10:52] <toad_> hmmm
[10:52] <toad_> we could swap :)
[10:52] <d-ArkAngel> :-P
[10:52] <lima> hi
[10:53] <toad_> I could give the R300 to mum, you could have my R250, and then she could buy me an nvidia 5xxx
[10:53] <toad_> :)
[10:53] <toad_> lima: hi
[10:53] <d-ArkAngel> lol
[10:53] <toad_> R250 works very well even for games, with the open drivers, as long as you don't need tv-out
[10:53] <d-ArkAngel> I think my better half has a 9200 I might see if I can get the DRI drivers going on x86_64 and swap the cards over.
[10:53] <toad_> good idea
[10:54] <toad_> lima: trying to run freenet? failing miserably?
[10:54] <d-ArkAngel> or... I could buy an nvidia card... :-)
[10:54] <toad_> do i need to respond to the email apparently from ebay talking about my account being used fraudulently? is it a scam to get my account details somehow?
[10:55] <d-ArkAngel> probably.
[10:55] <d-ArkAngel> what's it say? click this link and login to confirm your address or something?
[10:55] <d-ArkAngel> if so then it's almost certainly a scam.
[10:56] <d-ArkAngel> pluss you could always login to ebay and see if your bidding history is full of bids you didn't make :-) (which it won't be) and then ignore the mail.
[10:57] * tyesinclair (~tyesincla@170.252.64.1) has joined #freenet
[10:57] * tyesinclair (~tyesincla@170.252.64.1) has left #freenet
[10:57] <toad_> d-ArkAngel: it asks me to log in on some url on ebay
[10:57] <lima> lima: nope, ran freenet for nearly 2 years till last year, now i want to look at it again
[10:57] <d-ArkAngel> I think there's a mail address you can forward such mail to, and they'll say "thanks" for letting them know
[10:57] <toad_> or something like that
[10:57] <d-ArkAngel> yeah, that URL will be "faulty" in some way
[10:58] <d-ArkAngel> right click it, and copy link/URL and paste it in here.
[10:58] <toad_> 5096
[10:58] <toad_> 27th September
[10:58] <toad_> Nothing too magical in this latest build. It doesn't look like the insert issues have been addressed, but so far no fatalities have been reported.
[10:58] <toad_> That is all.
[10:58] <toad_> hmmm
[10:58] <toad_> i suppose we probably should reset stable
[10:59] <toad_> to try to fix the insert/timeout issues
[10:59] <d-ArkAngel> someone was in saying that there are a number of old files laying arround in the store/tmp dir on his node.
[11:00] * lima (~unknown@195.226.109.66) Quit ("leaving")
[11:00] <d-ArkAngel> kept saying it was wasting his bandwidth or something. I didn't really understand if it was really a problem to him, or if it was just a random question
[11:01] <toad_> hmmm
[11:01] <toad_> there have been such bugs on various occasions...
[11:01] <toad_> heh, i hope we didn't put lima off...
[11:03] * sanity (~ian@81-178-94-146.dsl.pipex.com) has joined #freenet
[11:05] <toad_> hi sanity
[11:05] <sanity> hi toad
[11:05] <toad_> are you actually here or are you a ghost of ian's dsl connection?
[11:06] <sanity> actually here this time
[11:06] <toad_> ah cool, i'm probably not talking to a network artifact ;)
[11:06] <toad_> sanity: ok
[11:06] <toad_> what are we doing?
[11:06] <toad_> we may have to reset stable to fix the insert issues
[11:07] <toad_> and the simulations aren't going especially well
[11:07] <sanity> in what way are the simulations not going well?
[11:08] <toad_> that is made clear on the list
[11:09] <sanity> hoptime?
[11:10] <sanity> well, i stand by my original statement on that - curve fitting to anything less than 50 points is statistically meaningless
[11:10] <d-ArkAngel> Oh dear, it seems that MS may be somewhat involved in the Kodak vs Java issue. It seems that MS has a licence to the technology and is part of the reason that "Kodak Imageing" comes as part of windows these days... So .net isn't threatend by this, just java...
[11:10] <toad_> sanity: I stand by my original response: 8,12,17,35 is clearly not logarithmic
[11:11] <sanity> toad: it isn't clearly anything
[11:11] <toad_> and even if it becomes logarithmic after 35 (these are for doublings of network size btw), it will still need an absurd HTL limit to work
[11:11] <sanity> toad: with more points we can determine what is *is*, not just what it isn't
[11:11] <toad_> even if it then goes 35,36,37,38,39, the HTL is VERY high
[11:11] <sanity> and from there we can experiment with different approaches to see why routing sucks
[11:11] <sanity> we can also compare it to old-style routing
[11:12] <toad_> sanity: okay, so how do more points help us exactly?
[11:12] <toad_> say we can fit some formula on 50 points
[11:12] <sanity> i would say at least 50
[11:12] <toad_> how does that help us?
[11:12] <sanity> because knowing what shape the curve is allows us to deduce what might be happening
[11:12] <toad_> unless you want them from very small networks, getting 50 points will take a very long time in terms of processing power alone
[11:13] <sanity> why? any sim I ever created this was the first experiment I ran, and it took no time. why is this simulation so painfully slow?
[11:13] <toad_> because it's a sim of NGR
[11:13] <sanity> well, either way
[11:13] <sanity> either we give up on the sim, or we do a proper sim to help us learn something about scalability
[11:13] <d-ArkAngel> and it's takeing millions of requests to settle to a running value.
[11:14] <toad_> and if we have to revert NGR, combined with the fact that you and oskar have practically proved that harvesting will always be a serious vulnerability, well, I'd probably have to leave
[11:14] <sanity> proceeding on the basis of 5 sample points to do curve fitting is meaningless and would, quite rightly, make us a laughing stock
[11:14] <toad_> sanity: doing a bunch of sims that show that NGR sucks and then giving up would be the worst thing we could do
[11:14] <toad_> i mean giving up on the sims
[11:15] <toad_> and then getting back to the regular random-walk hacking
[11:15] <sanity> toad: nobody is suggesting reverting NGR, but deliberately avoiding experiments that can help us determine what the problem is is definitely not the answer
[11:15] <toad_> it would be a disaster if anyone outside the core group ever heard, and probably even if not
[11:15] <sanity> huh?
[11:15] <toad_> what sort of experiments?
[11:15] <sanity> why are you being so defeatist?
[11:15] <sanity> this is useful
[11:15] <toad_> you mean simulating classical routing?
[11:16] <sanity> if you have found that NGR has poor scability, then it is a *good thing* that we know that - and we can start to think about *why*
[11:16] <toad_> so think about it
[11:16] <sanity> in principal NGR should have small world scalability, but there are a veriety of reasons it might not - such as it taking too long to learn about other node's routes
[11:16] <toad_> why does NGR suck?
[11:17] <toad_> well if it just takes time it should get better over time right?
[11:17] <sanity> to think about it, i need to know what the current scalability characteristic is - for that we need a proper sim, not 5 samples
[11:17] <sanity> NGR either a) doesn't suck, there is a bug in your sim
[11:17] <sanity> b) there is a bug in your NGR implementation
[11:17] <toad_> there are two balancing problems here: 1 is that we keep inserting data, the other is connection flux
[11:17] <toad_> my NGR impl is the same as is being used on the network
[11:18] <sanity> which makes it extra important to determine whether it is buggy
[11:18] <toad_> c) there is a fundamental problem in this particular NGR estimator
[11:18] <sanity> are you seeing specialisation in the NGR estimators?
[11:18] <toad_> d) there is a fundamental problem in NGR
[11:18] <sanity> indeed
[11:18] <toad_> e) there is a fundamental problem with freenet routing full stop
[11:18] <sanity> c) is probably the most likely
[11:18] <sanity> e) is contradicted by both theory and experiment
[11:18] <toad_> no, the old simulations match O(N^0.28)
[11:19] <toad_> quite clearly not log in the IEEE paper
[11:19] <sanity> forget them, the stuff oskar and i have been doing *proves mathematically* that LRU (which is the least-likely to work part of Freenet) results in a network with small world characteristics
[11:20] <toad_> with a routing algorithm that looks a lot like classical routing
[11:20] <toad_> also that paper doesn't prove ANYTHING
[11:20] <sanity> what paper?
[11:20] <toad_> see oskar's response to it
[11:20] <sanity> who is talking about that? i am taking about the theoretical work we have been doing
[11:20] <toad_> and i never saw any proof from your sims either, just some nice graphs and some hand-waving
[11:21] <toad_> your simulations assume fixed specialization, don't they?
[11:21] <sanity> toad: actually, it wasn't hand waving, you obviously just didn't understand it
[11:21] <d-ArkAngel> the thing I'm having trouble getting my head arround is that the network never seems to have a stable point in terms of the connections. Everything is always moving, I can't see that the network can learn routing as quickly as it can change....
[11:21] <toad_> sanity: oskar didn't claim to have anything approaching a proof
[11:21] <sanity> actually, that is exactly what he claimed - that he was approaching a proof
[11:22] <toad_> and he said what you had said wasn't anything close to a proof
[11:22] <toad_> sanity: what are the assumptions on routing algorithm here?
[11:22] <d-ArkAngel> the sims are always creating huge volumes of new connections, in a network that is a fixed set of nodes, with a fixed limit to connections.
[11:23] <toad_> d-ArkAngel: the problem is that if you don't have connection flux, you'll NEVER get the 1/d distribution required for logarithmic scalability
[11:23] <toad_> if you do, the nodes won't learn fast enough
[11:23] <toad_> we can ameliorate this a bit with the RT-bigger-than-connections thing
[11:23] <toad_> but probably not enough
[11:23] <sanity> toad: we assume fixed specialisation and a perfect routing algorithm - and clearly NGR (and old routing) are only a loose approximation to this. the question is whether they are close enough, and if not, how we can make them closer
[11:24] <toad_> sanity: in which case you have something that is very close to a classical DHT
[11:24] <toad_> sanity: i have a proof of this theorem, which is too large to fit in this margin...
[11:24] <toad_> :)
[11:25] <toad_> (fermat on fermat's last theorem)
[11:25] <toad_> it took hundreds of years, a huge proof, and some seriously innovative maths, to prove it
[11:25] <sanity> well, freenet is similar to a classical DHT in many ways. the differences are 1) LRU, 2) that nodes have an evolving rather than fixed specialisation, and 3) that routing decisions are approximate, not exact.
[11:25] * lolo-laptop (~lostlogic@68.251.84.226) has joined #freenet
[11:26] <sanity> oskar and I have experimentally demonstrated, and nearly proved that (1) isn't an issue
[11:26] * hobx_ (~chatzilla@ankh.math.chalmers.se) has joined #freenet
[11:26] <toad_> sanity: well what do you think is the problem with the current simulations? and what would you like to do with them?
[11:26] <sanity> next we will investigate (2), and then (3) - to see whether the things that make small world work are robust against (2) and (3)
[11:27] <sanity> toad: first we should investigate whether specialisations are emerging
[11:27] <sanity> toad: if not, that is a serious problem which will prevent everything else from working
[11:27] <toad_> sanity: why are specializations emerging on the real network?
[11:27] <toad_> apparently
[11:27] <sanity> i don't know
[11:28] <toad_> the simulation has the same algorithm
[11:28] <sanity> well, i mean, they should emerge
[11:28] <sanity> if they aren't emerging in the sim then we have a problem - are they?
[11:28] <toad_> i don't know
[11:29] <sanity> so first step is to find out
[11:29] <toad_> by specialization, what exactly are we looking for/at? the distribution of requests across the keyspace for popular nodes?
[11:29] <sanity> yes, and/or the shape of the estimator curves for other nodes, and how closely these match
[11:30] <sanity> related to this is how accurate the estimates are which the estimators are producing
[11:30] <toad_> so say every 100,000 requests, we dump all the keys in the datastore, the last 100 request keys, and the estimators for each node?
[11:30] * greycat (~wooledg@192.35.79.70) has joined #freenet
[11:31] <toad_> and we do this for 100x25 firstly, since that's the easiest to deal with
[11:37] <toad_> sanity: two things I'm not simulating that are on the real network and might be really important. one: estimator passing. two: offline RT (RT > open conns)
[11:38] <sanity> yeah
[11:38] <toad_> sanity: so the plan is to do all of the above, then get back to you
[11:38] <sanity> i think the most likely problem is simply that estimators are not learning fast enough, perhaps not learning at all
[11:38] <toad_> yes
[11:38] <toad_> it looks that way
[11:38] <sanity> and that will, obviously, totally fubar the whole network
[11:38] <toad_> thanks for being here
[11:39] <toad_> I'll be back in a couple hours, going shopping
[11:39] <sanity> ok
[11:39] <toad_> will be here this evening
[11:39] <toad_> will email you when/if i get some results
[11:39] <sanity> should be around too - txt me if you don't find me here and want to talk
[11:39] <toad_> ok
[11:43] <toad_> sanity: i think those two might be really important... anyway bbl
[11:46] * guido^pe (~unknown@dsl-213-023-247-108.arcor-ip.net) has joined #freenet
[12:00] * v3g`away (~sdfsd@hn-45-159.brookes.ac.uk) has joined #freenet
[12:01] <v3g`away> er hi, just curious about whether i am likely to get into trouble using freenet at my uni - they say no to p2p software. and if for example one is found with kazza they loose their internet
[12:01] <d-ArkAngel> ask them, it's thier rules.
[12:01] <v3g`away> thanks
[12:02] <v3g`away> ok
[12:02] * dli_ (~tst@anna-bhat.uchicago.edu) has joined #freenet
[12:02] <dli_> hello
[12:02] <v3g`away> does freenet connect through ports?
[12:03] <dli_> v3g`away, yes
[12:03] <dli_> v3g`away, but not a universal one
[12:03] <v3g`away> which ports? can u change them/
[12:03] <dli_> v3g`away, in your freenet.conf file
[12:03] <v3g`away> is it the same ports as other p2p software uses?
[12:04] <v3g`away> ok ok
[12:04] <dli_> v3g`away, I think you can change the one in your freenet.conf, different nodes have different ports, not a universal one
[12:04] <v3g`away> what is a node - i am very new to this
[12:05] <dli_> v3g`away, your box is one :)
[12:05] <v3g`away> :o ok
[12:06] <d-ArkAngel> it's the name for what the software client runs... it's not the same as other P2P tools
[12:06] <v3g`away> ok
[12:07] <dli_> d-ArkAngel, is toad_ around?
[12:07] <d-ArkAngel> essentialy the more you contribute the better your connection will be. A node is what you run, and you ask your node for files.
[12:07] <d-ArkAngel> He's gone shopping about 15 min ago
[12:07] <dli_> d-ArkAngel, oh, I usually go shopping after sun set :)
[12:08] <v3g`away> :o ok
[12:09] <dli_> d-ArkAngel, my friend told me, freenet is available there, not blocked
[12:10] <d-ArkAngel> where is it not blocked vs where is it blocked?
[12:11] <dli_> d-ArkAngel, toad_ thought freenet be blocked in China, and wanted to know how it's blocked, but it's available in China indeed
[12:12] <hobx_> SS1 has landed!
[12:13] <dli_> d-ArkAngel, as far as there's no clear trace of downloading sensitive things, it's safe to use freenet in China
[12:13] <d-ArkAngel> well with freenet there's no clear tace of anything is there? :-)
[12:14] <dli_> d-ArkAngel, not so obiviously, if they take your computer as evidence
[12:14] <d-ArkAngel> :-)
[12:15] <dli_> d-ArkAngel, I don't know whether toad can improve on this point
[12:16] <d-ArkAngel> well when people extract things from freenet, then that's outside of toad's control really...
[12:16] <dli_> d-ArkAngel, after you downloaded some pages, those pages are in your local cache
[12:16] <dli_> d-ArkAngel, no, I mean even people save nothing on purpose
[12:17] <d-ArkAngel> but they can get in your cache anyway.
[12:17] * Ash-Fox (Hal-9000@aba127.neoplus.adsl.tpnet.pl) Quit (Nick collision from services.)
[12:17] <d-ArkAngel> or do you mean in the IE cache?
[12:17] * Ash-Fox (Hal-9000@aaf197.neoplus.adsl.tpnet.pl) has joined #FreeNET
[12:17] <dli_> d-ArkAngel, there are news in China, when police seize all laptops, and scan internet cache
[12:17] <dli_> d-ArkAngel, both IE and mozilla keep certain page cache
[12:18] <d-ArkAngel> well that again is out of toads control...
[12:18] <dli_> d-ArkAngel, maybe, we can reconfigure mozilla to save nothing in cache
[12:19] <d-ArkAngel> almost certainly
[12:20] <dli_> d-ArkAngel, after all, the chance of being caught would be essentially none, if there's nothing in cache, when police knocking your door, you can always reboot
[12:21] <d-ArkAngel> maybe you could make a ram drive, and get it to store the temp internet files on the ram drive.
[12:21] <dli_> d-ArkAngel, so, after reboot, it's gone
[12:22] <d-ArkAngel> yeah
[12:22] <dli_> d-ArkAngel, would a ram disk be possibly swap to hard drive?
[12:23] <dli_> d-ArkAngel, that's too much for dummy users :(
[12:23] <d-ArkAngel> dunno. might do. In linux I'm sure you can make secure partitions so that they're in ram, and protected from other stuff.
[12:24] <dli_> d-ArkAngel, it's still a great news, if freenet continues to be available in China, it's a market of dozens of millions of computers
[12:25] <d-ArkAngel> you might be able to script something so that it makes a ram drive, and then installs Firebird to it automaticaly during start up.
[12:25] <dli_> d-ArkAngel, when the user number is over certain critical value, the speed would improve dramatically
[12:25] <d-ArkAngel> assuming that routing really works that way.
[12:25] <dli_> d-ArkAngel, to turn off disk cache is a simple solution
[12:26] <v3g`away> i am loading the web interface for the first time and it looks like its stuck on the build load
[12:26] <v3g`away> on 45% - has been for the last like 10mins - any recommendations?
[12:26] <dli_> v3g`away, your node needs some time to study the net
[12:26] <v3g`away> how much time?
[12:26] <d-ArkAngel> or maybe a boot CD that runs it. only use the HDD for the freenet node's store.
[12:27] <d-ArkAngel> load is the amount of Load on your node, not as in % complete
[12:29] <dli_> v3g`away, depends, on my nodes, it usually takes several hours to get about 60 connections
[12:30] <dli_> v3g`away, freenet couldn't be fast in the beginning, but it's practically fast after you get 70 like connections
[12:34] <dli_> v3g`away, at the web interface, click "switch to advanced mode" at the upper right corner
[12:35] <dli_> v3g`away, then you have a link of "open connections" at the left column, it links to your connections
[12:42] * Hory (~Miranda@81.196.25.110) has joined #FreeNet
[12:44] * v3g`away (~sdfsd@hn-45-159.brookes.ac.uk) Quit (Read error: 110 (Connection timed out))
[12:59] * d-ArkAngel (~Rob@213-131-104-86.onyx.net) has left #freenet
[13:08] <i2p_iip> <Newsbyte> toad, you there?
[13:09] * cbreak (~cbreak@dhcp-32-094.via-eth.ch) has joined #freenet
[13:18] * Digger3 (~digger3@kf-dn-tg01-174.dial.kabelfoon.nl) Quit (Read error: 104 (Connection reset by peer))
[13:21] * Digger3 (~digger3@kf-dn-tg01-174.dial.kabelfoon.nl) has joined #freenet
[13:21] * sanity (~ian@81-178-94-146.dsl.pipex.com) Quit (Read error: 232 (Connection reset by peer))
[13:22] * sanity (~ian@81-178-94-146.dsl.pipex.com) has joined #freenet
[13:41] * kers (~kers@hd5e25c8f.gavlegardarna.gavle.to) has joined #freenet
[13:42] * Hory (~Miranda@81.196.25.110) Quit (Read error: 54 (Connection reset by peer))
[13:43] <toad_> <dli_> d-ArkAngel, toad_ thought freenet be blocked in China, and wanted to know how it's blocked, but it's available in China indeed
[13:43] <toad_> woah cool
[13:43] <toad_> they got some of the sites?
[13:43] <dli_> toad_: yeah, they got freedom engine
[13:43] <dli_> toad_: and one site from there
[13:43] <dli_> toad_: even more amazing, using the central seednodes
[13:44] <toad_> dli_: cool!
[13:45] <dli_> toad_: there's a chance, freenet become popular in China
[13:45] <toad_> dli_: you know they'll come for us next, right? what are the viable options in china at the moment?
[13:45] <toad_> dli_: if they want to publish, there are some issues with charsets - the usual META HTTP-EQUIV trick doesn't work, you have to have the insert tool specify it in metadata
[13:45] <toad_> and I don't think the mainstream ones do that
[13:46] <dli_> toad_: since there was freenet-china, I think people figured out that trick
[13:47] <toad_> dli_: well, the people who did freenet-china did
[13:48] <toad_> they probably just enabled meta http-equiv though
[13:48] <toad_> but freenet-china is dead...
[13:48] <dli_> toad_: there was triangleBoy, powered up with big money, but freenet still available, the fact itself is impressive
[13:48] <dli_> toad_: yes, because nobody uses it :(
[13:49] <dli_> toad_: because people use things simpler, and more traditional, like ultrasurf and freegate
[13:50] <dli_> toad_: but both ultrasurf and freegate are essentially the same as triangleBoy, they would fail easily
[13:51] <dli_> toad_: freenet is older than triangleBoy, but still not blocked, while they blocked the word freenet and *.sf.net
[13:53] <i2p_iip> <Newsbyte> toad, how's things with the 10% of free HD as datastore thingy?
[13:54] <dli_> toad_: there are several cases this year in China, when police seized some porno (either VCD or downloaded in computers), but later, cases were all dismissed, because 1. for private use is not illegal by the law; 2. police didn't collect the evidence properly. I mean using freenet is not illegal, even in China
[13:57] <dli_> toad_: there's a huge market, but unfortunately, people are investing in close source tools like freegate and ultrasurf, both are not viable in long term
[14:08] * unknown_ (~unknown@dsl-213-023-245-024.arcor-ip.net) has joined #freenet
[14:09] * guido^pe (~unknown@dsl-213-023-247-108.arcor-ip.net) Quit (Read error: 113 (No route to host))
[14:17] * moskau23 (~Miranda@dsl-213-023-252-083.arcor-ip.net) Quit (Read error: 60 (Operation timed out))
[14:20] * tyesinclair (~tyesincla@170.252.64.1) has joined #freenet
[14:20] * tyesinclair (~tyesincla@170.252.64.1) has left #freenet
[14:26] * joltrush (~josh@paradroid.com) has joined #freenet
[14:30] * plixed_away (~plixed@pD9551513.dip.t-dialin.net) Quit (Client Quit)
[14:31] * plixed (~plixed@pD9551513.dip.t-dialin.net) has joined #freenet
[14:35] * hirvox (~hirvox@cs181027153.pp.htv.fi) has joined #freenet
[14:40] * joltrush (~josh@paradroid.com) has left #freenet
[14:43] * moskau23 (~Miranda@dsl-082-082-237-069.arcor-ip.net) has joined #freenet
[14:44] * kenny (~daniel@80.178.101.55.forward.012.net.il) has joined #freenet
[14:44] * kenny (~daniel@80.178.101.55.forward.012.net.il) has left #freenet
[14:51] * tav is now known as tav|offline
[14:51] * sanity (~ian@81-178-94-146.dsl.pipex.com) Quit (Read error: 232 (Connection reset by peer))
[14:51] * pittaman (~mistery_b@d51A410CA.kabel.telenet.be) has joined #freenet
[14:52] * sanity (~ian@81-178-94-146.dsl.pipex.com) has joined #freenet
[14:53] * tav|offline is now known as tav
[14:53] * toad_ back
[14:55] <toad_> <hobx_> SS1 has landed! - ah, that SS1. Yeah, very cool, see them in 2030 when they make orbit ;)
[14:56] <toad_> <dli_> d-ArkAngel, as far as there's no clear trace of downloading sensitive things, it's safe to use freenet in China - umm, well.. don't they invent arbitrary retroactive new laws? i know they torture people to extract confessions as part of the regular judicial process...
[14:56] <toad_> <dli_> d-ArkAngel, maybe, we can reconfigure mozilla to save nothing in cache - we do, i think it's by default, but would have to check
[14:57] <toad_> <d-ArkAngel> maybe you could make a ram drive, and get it to store the temp internet files on the ram drive.
[14:57] <toad_> <dli_> d-ArkAngel, so, after reboot, it's gone
[14:57] <toad_> well.. then they'll assume you have something to hide and torture you until you admit to something :|
[14:58] * Kyon (~Kyon@134.53.120.239) Quit (Read error: 110 (Connection timed out))
[14:58] <cbreak> at least with freenet you have something to admit...
[14:58] <dli_> toad_: but things are changing fast in China, last year, police can still arrest anyone on street, for not holding a valid city pass (zan zhu zheng, or temporary residence), this year, it's illegal for police to do so
[14:58] <toad_> <i2p_iip> <Newsbyte> toad, you there? - i am now...
[14:59] <toad_> <dli_> toad_: but both ultrasurf and freegate are essentially the same as triangleBoy, they would fail easily - all of the above are blocked now?
[14:59] <dli_> toad_: theoretically, yes, but people start to ask openly, whether it's constitutional.
[14:59] <toad_> <i2p_iip> <Newsbyte> toad, how's things with the 10% of free HD as datastore thingy? - i have no idea...
[15:00] <toad_> <dli_> toad_: there are several cases this year in China, when police seized some porno (either VCD or downloaded in computers), but later, cases were all dismissed, because 1. for private use is not illegal by the law; 2. police didn't collect the evidence properly. I mean using freenet is not illegal, even in China - freenet was on these machines??
[15:00] <hobx_> toad_: Carrier ship the size a 747, a first stage booster rocket that takes it to 100 km and is manned and glides back home. At 100 km you have seperation of the orbital craft, which fires the rest of the distance.
[15:00] <hobx_> I could see it happen...
[15:00] <dli_> toad_: yeah, ultrasurf stopped working long ago, they are still making new versions of freegate, but most people say, it's not accessible
[15:01] <toad_> <dli_> toad_: but things are changing fast in China, last year, police can still arrest anyone on street, for not holding a valid city pass (zan zhu zheng, or temporary residence), this year, it's illegal for police to do so - cool
[15:01] <hobx_> I should get a rendering program and start doing mockups.
[15:01] <toad_> hobx_: two stages? cool
[15:02] <toad_> ok, where was i?
[15:02] <toad_> dli_: is triangleboy blocked?
[15:02] <toad_> and ultrasurf?
[15:02] <hobx_> I think single stage to orbit is stupid. Why carry all the weight to space? But, OTOH, the first stage need not be a NASA type rocket that falls in the ocean. Build it as a reusable Booster Ship.
[15:03] <dli_> toad_: triangleBoy was totally breached, and the project terminated
[15:03] <dli_> toad_: all my friends couldn't use ultrasurf
[15:04] <toad_> triangleBoy was an internal project?!
[15:04] <toad_> hobx_: yes, if you can make all components reusable then multiple stages to orbit is fine
[15:05] <dli_> toad_: I remember triangleBoy got a contract from voa.gov totals $8 million
[15:05] <toad_> assuming recovery, refueling, retesting isn't expensive
[15:05] <toad_> dli_: it was internally developed?
[15:05] <dli_> toad_: "internal"?
[15:05] <toad_> or they terminated the project after it was blocked and they ran out of ideas?
[15:06] <toad_> within china
[15:06] <dli_> toad_: I think developped in canada
[15:06] <toad_> dli_: so why was the project killed?
[15:06] * pupok (~r00t@81-178-94-146.dsl.pipex.com) Quit (Read error: 104 (Connection reset by peer))
[15:06] <hobx_> The only problem is how one makes the rocket aerodynamic both before and after seperation...
[15:06] * toad_ thinks $8M is absurd..
[15:07] <dli_> toad_: because Chinese net police figured out how to block it completely
[15:07] <toad_> dli_: for $8M, surely they could develop a countermeasure? how did it work?
[15:07] * sanity (~ian@81-178-94-146.dsl.pipex.com) Quit (Read error: 104 (Connection reset by peer))
[15:07] <hobx_> I think all the get around censorship systems for china have the wrong idea.
[15:07] <dli_> toad_: triangleBoy is basically an https proxy network, my impression
[15:07] <hobx_> And so does Freenet of course
[15:07] <hobx_> Nothing like that will work.
[15:07] * pupok (~r00t@81-178-94-146.dsl.pipex.com) has joined #freenet
[15:07] <toad_> hobx_: yeah, because everyone in china has a western friend they can proxy over
[15:08] <hobx_> toad_: Exactly
[15:08] <hobx_> that can't work
[15:08] <toad_> hobx_: small group networks do not automatically work better, and have serious discovery problems
[15:08] <hobx_> I agree
[15:08] <hobx_> Not what I was thinking at all.
[15:08] <toad_> ok, we agree :) - then why is freenet the wrong idea? apart from "it can't possibly work" issues?
[15:08] <hobx_> I think one needs to step further back.
[15:09] <hobx_> Trying to give the chinese free net access is hopeless
[15:09] <hobx_> It is a question of trying to smuggle information.
[15:10] <toad_> maybe in the long term
[15:10] <dli_> hobx_: it's not that hopeless, now in China you can watch cnn, fox news, hbo, if you can afford it, it's partly opening up
[15:10] <toad_> dli_: satellite dishes aren't illegal?
[15:10] <hobx_> I think they should allow Foxnews.
[15:10] <toad_> they were in iraq and probably still are in some middle eastern nations...
[15:10] <hobx_> Makes chinese media look credible! :-)
[15:11] <dli_> toad_: yes, it's still illegal, but they provide you cable, in highend apartments
[15:11] <dli_> toad_: interesting, none provides bbc, but cnn is everywhere
[15:12] <toad_> they don't have some sort of special deal with cnn?
[15:12] <toad_> hobx_: LOL
[15:12] <toad_> ;)
[15:12] <dli_> hobx_: foxnews sounds just like communists, but the other end
[15:13] <hobx_> Watch O'Reilly for a while and you'll start thinking "Maybe Mao had a point"
[15:13] * sanity (~ian@81-178-94-146.dsl.pipex.com) has joined #freenet
[15:13] <hobx_> Actually I don't mind it so much. At least you know where Foxnews stands, so you can apply a filter to what you read.
[15:13] <toad_> ah
[15:14] <toad_> but fox is one of the independants, isn't it? i.e. it's not owned by a defence contractor? ;)
[15:14] <dli_> hobx_: independent doesn't mean unbiased
[15:14] <toad_> indeed
[15:14] <toad_> dli_: toad_: not hobx_:
[15:15] <toad_> most of the really-independant media is hard left and definitely not unbiased :)
[15:15] <toad_> in the west
[15:16] <KenMan> "the world is not enough" - that 007 flick ?
[15:16] <Ash-Fox> Show me some unbiased media
[15:16] <dli_> toad_: interesting, all my friends, who watch cnn in China, are playing around with the net blocking
[15:16] <toad_> dli_: hmm?
[15:16] <toad_> hmmm
[15:16] <toad_> yeah, it's curious that they allow CNN
[15:16] <dli_> Ash-Fox, how about online? you only read what you like
[15:16] <Ash-Fox> dli_, it's still biased to what you like
[15:17] <dli_> toad_: not for most people, only for those who can afford
[15:17] <KenMan> how many chinese people afford computers, percentage wise ?
[15:17] <dli_> Ash-Fox, my point is that, when info flow is free, it could be more biased
[15:17] <toad_> KenMan: a RAPIDLY increasing number
[15:18] * Ash-Fox is hosting 6312 chinese websites
[15:18] <toad_> KenMan: which generally includes the potentially political-activist classes
[15:18] <dli_> KenMan, I think the number is around 200 million
[15:18] <toad_> Ash-Fox: woah
[15:18] <Ash-Fox> and I'm hosting 10642 websites currently
[15:18] <toad_> Ash-Fox: you work for a hosting company? :)
[15:18] <dli_> KenMan: about 100 million surfing online regularily
[15:18] <toad_> woah
[15:19] <toad_> isn't that the entire urban population of china?
[15:19] <Ash-Fox> toad_, I provide free hosting (Currently suspended new registrations untill I get more bandwith)
[15:19] <toad_> Ash-Fox: total size? bandwidth use per month?
[15:19] <KenMan> that's like 1 in 10 chinamen. I thought it was more like 1 in 1000...
[15:19] <Ash-Fox> toad_ http://users.quickfox.org/ - bandwith usage.. too much, it's increasing constantly
[15:19] <Ash-Fox> (seeing that you can't register, this isn't advertising)
[15:20] <dli_> toad_: basically, poor country side
[15:20] <dli_> toad_: so poor, they are forbidden by the law to live in cities
[15:21] <dli_> KenMan: you never being to China :)
[15:21] * Ash-Fox has a friend who went to China
[15:21] <Ash-Fox> He said he could get internet access in the tiny villages
[15:21] <Ash-Fox> infact I talked to him though VOIP with it
[15:22] <plixed> well, we live in 21st century. this means technology exists, but only few can afford it
[15:22] <dli_> KenMan: if you go to Guangzhou, you would simply impressed by the luxury cars on street, no more rusty second cars, half are bmw, benz, luxus
[15:22] <toad_> dli_: the peasants are legally forbidden to drive into the cities? how do they buy supplies then?
[15:23] <dli_> toad_: no drive, police can stop you and arrest you. in many case, put you into a labor camp for months, and ask your family to buy your release
[15:24] <dli_> toad_: peasants don't have cars :)
[15:24] <KenMan> whatever generates the revenues :p
[15:24] <toad_> plixed: i suspect the ratio of bandwidth costs to food costs is very different in places like china to here
[15:24] <toad_> dli_: don't they have tractors?
[15:24] <toad_> farming is all fairly large scale post mao, right?
[15:25] <dli_> KenMan: exactly, it was a business so profitable, many government depts stepped in
[15:25] <KenMan> somebody over there is ordering a lot of large farming equipment. Perhaps it is just to make WMD.
[15:25] <dli_> toad_: for most, no, for those who have, there's no need to become cheap labor in cities
[15:25] <plixed> they have so many people that they don't need tractors ;-) but seriously, do tractors help with rice farming?
[15:26] <greycat> KenMan: Fear the Weapons of Mass Dairy! Mooooo!
[15:26] <KenMan> heehee
[15:26] <dli_> KenMan: good thing is that, since last year, the kidnapping business was ordered to stop, now, police couldn't just stop anyone and check for id
[15:26] <toad_> <dli_> KenMan: exactly, it was a business so profitable, many government depts stepped in - uh, what was?
[15:27] <dli_> plixed: we don't use people, we use cattles
[15:27] <toad_> <KenMan> somebody over there is ordering a lot of large farming equipment. Perhaps it is just to make WMD. - LOL
[15:28] <dli_> greycat: fortunately, there's no radical religion in mainstream of chinese
[15:30] <dli_> toad_: I belive bbc still got blocked, because it hosts a Chinese version, while cnn provides only Japanese version for asia
[15:31] <toad_> greycat: xtianity doesn't count because it's not quite mainstream :)
[15:31] <toad_> err i mean dli_:
[15:32] <greycat> not sure I'd count that one as radical...
[15:33] <toad_> :)
[15:34] <dli_> greycat: most religions are radical, buddhism is a good alternative for those
[15:34] * cbreak belives in physics...
[15:35] <dli_> cbreak: don't just believe in physics books
[15:36] <cbreak> not books, the concept. The World works after fixed rules, we just have do find them...
[15:37] <toad_> [VERY OFF TOPIC] well anyway, the point i was making is that if you don't think xtianity is radical you probably don't know many real xtians well enough to form an opinion
[15:37] * sanity (~ian@81-178-94-146.dsl.pipex.com) Quit (Read error: 232 (Connection reset by peer))
[15:37] <greycat> There are also many, many different flavors of christianity.
[15:38] * sanity (~ian@81-178-94-146.dsl.pipex.com) has joined #freenet
[15:38] <plixed> there are also many, many different flavors of stupidity ;-)
[15:38] <toad_> ;)
[15:38] <dli_> cbreak: what concepts? physics (or physical sciences) is more based on scientific methods, not the conclusions
[15:38] <toad_> uh oh..
[15:38] <cbreak> exactly.
[15:40] <dli_> cbreak: then you have to accept we can not ask why on this question. like the time before the big bang, like why this partible is detected here, with it's diffuse wavefunction
[15:41] <toad_> dli_: which question? the validity of empiricism?
[15:41] <i2p_iip> <Newsbyte> woohoo! flamewar on devl!
[15:41] <i2p_iip> <Newsbyte> ;-)
[15:41] <dli_> toad_: physics doesn't tell you everything, because you simply shouldn't be asking, like the time before the big bang
[15:41] <i2p_iip> <Newsbyte> Ian is tiresome. You hardly can debate anything rationally. He's even worse then a xian, toad! ;-)
[15:42] <cbreak> Nothing tells you everything.
[15:42] <toad_> LOL Newsbyte
[15:42] <i2p_iip> <Newsbyte> why is he always so damn irritating
[15:42] <toad_> dli_: why not? many physicists have speculated about it.. but there is no evidence, and there probably never will be any
[15:43] <dli_> cbreak: but one thing for sure, our physical structure is quit suitable to creat and stick to religions
[15:43] <i2p_iip> <Newsbyte> I know he says the same of me, but at least I have a notion of what he finds irritating. He doesn't seem to have the capacity to see why ppl think HE is irritating, however
[15:43] <cbreak> dli_: Not quite. No "In six days he creaded the world" mumbo jumbo.
[15:43] <plixed> dli_: yeah, what you can't explain, you explain with a singular enitity of you imagination
[15:44] <dli_> toad_: speculations, it can easily start as science and soon turns to scientific fictions
[15:44] <i2p_iip> <Newsbyte> and his 'you never contributed anything technical' is SO damn adjusted to be irritating. I mean, why do you think he added the last part, you think?
[15:45] <toad_> dli_: yeah, untestable theories are not science
[15:45] <dli_> plixed: religions are part of the brain structure, a byproduct of evolution
[15:45] <dli_> toad_: now, the standard theory is simply, don't ask. obiviously, a puzzle for those who believe in physics
[15:45] <plixed> dli_: nothing wrong with speculations, as long as you don't claim to explain more that you can prove
[15:46] <i2p_iip> <Newsbyte> which isn't even true; I have given technical suggestions more then once, be it of a general nature. It's not my fault he choses to ignore it because he thinks it's bull
[15:46] <plixed> dli_: imagination is a part of the brain structure, not religious fanatism
[15:46] <i2p_iip> <Newsbyte> one prime example is the increase of htl, for instance. That was considered bull too, but yet, it got implemented after a while.
[15:46] <dli_> plixed: but speculations are not science, because a the testing is not complete, the logic is still not self-consistent
[15:47] <cbreak> Newsbyte: At that time, a change was introduced which decreased the range of a message at a given htl.
[15:47] <dli_> plixed: I believe there was a reccent paper in nature, correlation between religious feelings and brain activities
[15:48] <i2p_iip> <Newsbyte> toad, do you find I have never contributed anything worthwhile to the freenet-project?
[15:48] <toad_> Newsbyte: no, you contributed freenethelp.org
[15:48] <i2p_iip> <Newsbyte> cbreak, that could be, but I was the one that first suggested the htl increase
[15:48] <cbreak> :)
[15:49] <toad_> heh
[15:49] <toad_> dli_: which proves what exactly?
[15:49] <cbreak> so, I propose that freenet increases routing efficiency, and decreses load!
[15:50] <cbreak> (And I am the FIRST :)
[15:51] <i2p_iip> <Newsbyte> and I helped secure some money at the time it was getting finacially tough, and I run a freenode (though not always), and I have made a lot of contributions to the website, and I offered to run a testnetworknode with a shell for you, etc. But according to Ian, that doesn't count, because I'm no javacoder.
[15:51] <i2p_iip> <Newsbyte> he can be such an ass, sometimes
[15:51] <i2p_iip> <Newsbyte> well, most of the time, actually.
[15:51] <i2p_iip> <Newsbyte> ;-)
[15:51] * toad_ has a word for an untestable theory... philosophy? ;)
[15:51] <cbreak> Newsbyte: You don't contribute to freenet the programm, but to freenet, the project.
[15:52] <cbreak> untestable theory is called axiom.
[15:52] <dli_> toad_: religious feeling is simply a brain function, and which can be further interpreted as an advantage in the evolution process
[15:53] <plixed> i bet you can prove a correlation between "philosophical feelings" and brain activity, you just need a very broad definition of "philosophical feelings" ;-)
[15:53] <toad_> dli_: as an openly religious person i don't have a problem with "religious feeling" being a brain function of some sort
[15:53] <i2p_iip> <Newsbyte> right
[15:53] <i2p_iip> <Newsbyte> I would like to see one other non-coder that did as much for the project as me, frankly
[15:54] <dli_> plixed: the fundamental question is still understanding brains by brains
[15:54] <toad_> dli_: depending on how broad your definition is (the experiments in question deal with a very narrow definition)
[15:55] <plixed> all hail newsbyte.
[15:55] <toad_> hail to thee, that art king hereafter...
[15:55] <i2p_iip> <Sonax> Newsbyte: I see all your hard (non coding) work, and raise you by a "I uploaded pr0n to the network". :)
[15:56] <toad_> lol
[15:56] <dli_> toad_: right, if you ask as question to broad, it's usually not science, because conditions are not controlled
[15:56] * kers (~kers@hd5e25c8f.gavlegardarna.gavle.to) Quit ("Leaving")
[15:56] <i2p_iip> <Newsbyte> hmm...religion, again. I know some religious guys that refute the hypothesis that it's a brainfunction just by being so. They are simply too braindead to have any function! ;-)
[15:57] <toad_> lol
[15:57] <i2p_iip> <Newsbyte> pron is good
[15:57] <plixed> no: porn is god ;-)
[15:57] <toad_> lol
[15:58] <plixed> it shurly activates your brain
[15:58] <i2p_iip> <Newsbyte> which reminds me; I'm in the middle of making a freesite for the testnetwork full of legal (copyright) mp3s and maybe some video as well, so you can use it for pull/push tests without danger. :-)
[15:58] <toad_> Newsbyte: cool
[15:58] <i2p_iip> <Newsbyte> I'll add it's a worthless non-technical contribution, of course.
[15:59] <i2p_iip> <Sonax> Newsbyte: key?
[15:59] <i2p_iip> <Newsbyte> I'm still making it, it's not finished yet. Besides, it's for the testnetwork
[16:00] <i2p_iip> <Newsbyte> (if that ever gets done)
[16:00] <i2p_iip> <Sonax> Any plans for a soon to be test net that i have missed?
[16:00] <toad_> only in Newsbyte's imagination
[16:01] <cbreak> there's unstable and the simulations
[16:02] <plixed> do the 20 nodes in unstable count as a network? ;-)
[16:03] <cbreak> there are at least 50
[16:09] * kers (~kers@hd5e25c8f.gavlegardarna.gavle.to) has joined #freenet
[16:18] <toad_> |Getting
[16:18] <toad_> into discussions with you is less productive than masturbation, and
[16:18] <toad_> much less enjoyable.
[16:18] <toad_> [from ian's post, to newsbyte]
[16:24] * unknown_ (~unknown@dsl-213-023-245-024.arcor-ip.net) Quit (Remote closed the connection)
[16:28] <i2p_iip> <Newsbyte> back...did I miss anything? :-)
[16:29] <greycat> something about Ian masturbating
[16:30] <i2p_iip> <Newsbyte> lol
[16:30] <i2p_iip> <Newsbyte> I think I've seen that somewhere on the list allready. Diplomatic talk at its best! ;-)
[16:31] * Hory (~Miranda@81.196.25.110) has joined #FreeNet
[16:33] <i2p_iip> <Newsbyte> toad gone?
[16:34] <toad_> hi
[16:34] <i2p_iip> <Newsbyte> I was going to ask him what he consideres a reasonable amount of mp3 and such, to have an adequate amount of data on the site to push/pull
[16:35] <i2p_iip> <Newsbyte> ah, you're still there. :-) I seem to have trouble with my IIP connection, these last weeks..always lagging and discon and such... :-/
[16:40] <plixed> if the nodes pass data-messages around without storing them, freenet could work as an irc proxy ;-)
[16:40] * sanity (~ian@81-178-94-146.dsl.pipex.com) Quit (Read error: 104 (Connection reset by peer))
[16:41] <i2p_iip> <Newsbyte> so, would a site with, say 10 mp3 suffice?
[16:42] <toad_> it'd be a start
[16:44] <plixed> it is way more attractive than an outdated linux-distribution
[16:45] * sanity (~ian@81-178-94-146.dsl.pipex.com) has joined #freenet
[16:45] <toad_> plixed: i don't think FC2 is outdated, is it?
[16:46] <toad_> sanity: hi, implementing estimator passing
[16:46] <sanity> toad: good
[16:47] <plixed> a distribution is generally unattractive because you find a mirror which exceeds your download speed in no time
[16:49] <i2p_iip> <Newsbyte> well, I could add more. I could simply repeat the stuff with other names, even; it's just for testing, after all. What would be a good enough for you, toad?
[16:50] <toad_> plixed: well, most legal files will be available on the regular internet faster
[16:51] <i2p_iip> <Newsbyte> right
[16:51] <toad_> for the time being anyway
[16:51] * kers (~kers@hd5e25c8f.gavlegardarna.gavle.to) Quit (Read error: 104 (Connection reset by peer))
[16:52] <toad_> although certainly there are files that have bandwidth issues...
[16:52] * kers (~kers@hd5e25c8f.gavlegardarna.gavle.to) has joined #freenet
[16:52] <i2p_iip> <Newsbyte> how's the /; simulatorthingy comming along, toad?
[16:52] <plixed> toad_: so you have to offer other advantages, a custom collection of legal mp3 is a good example
[16:53] <i2p_iip> <Newsbyte> plixed: I'm working on it :-) (sort off)
[16:53] <i2p_iip> <Newsbyte> though some will consider it to be worthless, because it's not a technical contribution....
[16:55] <plixed> without users and content a p2p network is just technical masturbation ;-)
[16:56] <toad_> plixed: indeed, but not all content is illegal
[16:56] <toad_> and not all illegal content is universally regarded as bad
[16:56] <toad_> e.g. the diebold files
[16:56] <i2p_iip> <Newsbyte> well, yes, compared to masturbation, some things suck even more
[16:57] <i2p_iip> <Newsbyte> whores suck too, for instance
[16:59] <i2p_iip> <Newsbyte> the focus on purely technical achievements is misplaced, me things. As plix says, without users and content, all the technical contribution amounts to little
[17:00] <i2p_iip> <Newsbyte> which doesn't mean we should strive for a better working network, of course :-) But, equally as important are the non-technical issues that makes people actually USE Freenet
[17:00] <i2p_iip> <Newsbyte> Sometimes, I think Ian fails to see that.
[17:01] <i2p_iip> <Newsbyte> shite. I think I'm lagging again
[17:01] <i2p_iip> <Newsbyte> 'shouldn't strive'
[17:01] <plixed> i do not think that. the network is still in beta
[17:01] <i2p_iip> <Newsbyte> you can say that again ;-)
[17:02] <i2p_iip> <Newsbyte> but even beta's need users/sponsors, if it has to pay a coder...
[17:06] <i2p_iip> <Newsbyte> I'm reminded of the VHS vs betamax thingy: even though VHS was technologically inferior and what not, it still won and beat the crap out of betamax. Why? Simple; because it had to offer more. ultimately, ppl are interested in things that give a benefit to them.
[17:07] <i2p_iip> <Newsbyte> the technology for the (normal) enduser really comes second (well, even lower, probably).
[17:08] * kers (~kers@hd5e25c8f.gavlegardarna.gavle.to) Quit ("Leaving")
[17:09] <i2p_iip> <Newsbyte> btw, toad, have you had a look at the updatingthingy of the cvs? It seems not to work, again (as I've said in a mail)
[17:09] <i2p_iip> <Newsbyte> or maybe Ian can look at it, if he's not too much in the 'ignore-newsbyte'-mood ;-)
[17:10] <toad_> which is interesting considering the corps at the time seemed to think that the main use for both was to pirate films ;)
[17:11] <i2p_iip> <Newsbyte> well, it is/was. But it had other uses too, and at the end they gained by it, instead of losing.
[17:11] <toad_> no, it isn't
[17:12] <hobx_> betamax superiority is a myth
[17:12] <hobx_> basically it was only superior for having smaller tapes
[17:12] <i2p_iip> <Newsbyte> but that's how corps reason; there is a technological change, and they are afraid they'll lose their long-cherised&established method of making profit. In reaction, they ALWAYS try to halt the new development, instead of searching for ways to adapt and profit from it.
[17:12] <hobx_> which were also shorter incidentally (not long enough for many movies in the initial version).
[17:14] <i2p_iip> <Newsbyte> I'm incilined to disagree, hobx. While it was somwhat before my time ;-) I've read some old reports that mentionned betamax to be of superior quality (in image)
[17:15] <i2p_iip> <Newsbyte> usually though, corps don't succeed in halting a new technology
[17:15] <toad_> Newsbyte: they've realized that they can cheat lately..
[17:16] <toad_> well ok they've always cheated
[17:16] <toad_> what's different with p2p is that there's no big money behind it
[17:16] <i2p_iip> <Newsbyte> what, by lobbying lawmakers? They've done that before...
[17:16] <hobx_> no big money is also no big money to take
[17:17] <hobx_> You worry about the wrong things.
[17:17] <toad_> hobx_: what do you mean? buying out the devs?
[17:17] <hobx_> Worry about:
[17:17] <hobx_> The firewalled consumer.
[17:17] <hobx_> The user hostile computer.
[17:17] <hobx_> And only that.
[17:17] <hobx_> The rest is bullcrap.
[17:17] <toad_> the user hostile computer is forced upon the user by the corporations :)
[17:17] <cbreak> who would buy a TCPA Enabled computer anyway? (Besides IBM Customers...)
[17:18] <i2p_iip> <Newsbyte> and don't forget to look at the cvs-update-thingy, btw. I'm puzzled why it keeps reverting to non-working? I've emailed you several times before, for it. and then it works like a charm, but some weeks later, it's back to the old 'doesn't update' issue (?)
[17:18] <hobx_> Actually it is forced upon us by the government
[17:18] <toad_> cbreak: people who want to watch non-
[17:18] <toad_> non-pirated online movies?
[17:18] <toad_> or everyone who isn't an outlaw if they force it upon us
[17:18] <hobx_> Without anti-circumvention laws they could try all they wanted.
[17:18] <toad_> Newsbyte: I have no idea really..
[17:18] <toad_> hobx_: sure, anti-circumvention is a bitch
[17:19] <hobx_> Corportations, by their very nature, cannot force anything on us.
[17:19] <toad_> but mandatory TC will be MUCH WORSE
[17:19] <hobx_> corporations cannot make TC mandatory.
[17:19] <plixed> maybe they introduce it like the region-code in dvd-players
[17:19] <toad_> corporations can force anything on the minority
[17:19] <hobx_> no
[17:19] <toad_> it takes two parties to execute a bribe
[17:19] <hobx_> democracy can force anything on the minority
[17:20] <hobx_> Only the receiver is guilty.
[17:20] <toad_> under which countries' law?
[17:20] <hobx_> Under my moral law.
[17:20] <toad_> try to bribe a policeman and you may find the opposite..
[17:21] <hobx_> Under any sensible law. It is one who accepts a bribe who betrays a confidence.
[17:21] <toad_> ah, under your law constructed specifically to demonize the state and ignore the problems caused by corps? ;)
[17:22] <hobx_> Depends what you mean by "coporations"
[17:22] * sanity (~ian@81-178-94-146.dsl.pipex.com) Quit (Read error: 104 (Connection reset by peer))
[17:22] <hobx_> If you mean "the set of government laws that regulate commercial businesses" then I agree that lots of problems are caused by it.
[17:22] <toad_> anyhow there's not much we can do about TC
[17:23] <toad_> no, I would argue that a lot of problems are caused by government failure to regulate corporations
[17:23] <toad_> :)
[17:23] <hobx_> You mean government failure to regulate enterprise?
[17:23] <toad_> yes
[17:23] <toad_> e.g. Bhopal
[17:23] <greycat> no, more like "government giving blatant handouts to corporations"
[17:23] <hobx_> You mean government failure to tell people that they cannot interact with eachother voluntarily as they see fit?
[17:24] <plixed> they do that even if are not told so
[17:24] <greycat> "What's that Mr. Eisner? Your precious Mickey is in danger of being in the public domain? No fear! We'll extend copyright law again!"
[17:24] <hobx_> You mean government failure to tell one man that he cannot have the fruits of his own labor, when he has received them from nobody who did not choose to give him the money?
[17:24] <toad_> hobx_: I mean government failure to prevent corporations from doing things to seriously to the detriment of the surrounding, exploited or employed people
[17:24] <cbreak> the failure to tell people they can't just take other's money and fly to the bahamas...
[17:25] <hobx_> employed people make money because the company exist
[17:25] <hobx_> that they would not make otherwise
[17:25] <greycat> The failure to enforce laws against dumping toxic levels of mercury into Ohio's water supply.
[17:25] <hobx_> they have the job by choice
[17:25] <hobx_> How are they exploited?
[17:26] <toad_> usually by cartels, anti-union legislation, anti-union illegal sanctions, and so on
[17:26] <hobx_> anti-union legislation is enacted by free enterprise?
[17:26] <hobx_> How on earth does that work?
[17:26] <toad_> no
[17:26] <hobx_> didn't think so
[17:26] <greycat> that's a much trickier issue
[17:26] <toad_> well yes, via the government
[17:26] <hobx_> wow
[17:26] <hobx_> surprise
[17:26] <toad_> but see the second point
[17:27] <i2p_iip> <Newsbyte> well, there was an interesting docu on BBC (panorama) yesterday, which showed clearly that the greed of corporations are detrimental to the health and lives of ppl.
[17:27] <toad_> if there is no government, the corporations take matters into their own hands
[17:27] <toad_> if there's a strike, kill some of the strikers
[17:27] <toad_> no more strike
[17:27] <toad_> this has happened on occasion
[17:27] * sanity (~ian@81-178-94-146.dsl.pipex.com) has joined #freenet
[17:27] <hobx_> People aren't allowed to kill others
[17:27] <i2p_iip> <Newsbyte> btw, toad, should I re-send the cvsthingy? It's been 2-3 days, since I send it.
[17:27] <greycat> That comes back of failure of the government to enforce the few good laws that exist.
[17:27] <hobx_> That is the law, and has nothing to do with corporations or not corporations.
[17:28] <plixed> blaim the voters for putting incompetent people into the government
[17:28] <toad_> hobx_: indeed, and who's going to prevent them from killing others? who's going to prevent them from selling toxic byproducts as fertilizer?
[17:28] <toad_> the government!
[17:28] * cbreak (~cbreak@dhcp-32-094.via-eth.ch) Quit ("exam training")
[17:28] <greycat> if the CEO of a corporation hires a thug to kill the leader of the workers' union, then the CEO should be found guilty of murder and hanged (or incarcerated, deported, name your punishment of choice).
[17:28] <greycat> but since the CEO is rich and the union leader is poor, it doesn't work that way in practice. the government caters to the rich because the rich keep the government in power. vicious circle.
[17:28] <toad_> corps are not morally competent to make decisions, so need to be supervised
[17:29] <toad_> if corps can buy the government, then the result is terrible
[17:29] <greycat> as we have seen.
[17:29] <hobx_> plixed: I blaim people like toad who attack the foundations that which has built the foundation of all of our wealth. Who attack the simple moral premise that people have a right to interact by there own choice, that people who earn money only from people who give it to them by choice have a right to that money, and that it is government, and government only, that acts over us by force.
[17:30] <greycat> hobx_: the problem is that the government is corrupt, and does not apply the natural laws (no murder, etc.) equally
[17:30] <toad_> that's part of the problem
[17:30] <greycat> then it makes up new laws to keep the rich in power
[17:30] <toad_> but it's not just the government
[17:30] <hobx_> greycat: All power is corrupt. Liberal market economies are a hell of a lot less corrupt then anything else we have seen.
[17:31] <toad_> one function of the government is to prevent cartels
[17:31] <greycat> this causes a blatant distortion of the free market which would normally tend to equalize wealth
[17:31] <toad_> something it does reasonably well - sometimes ;)
[17:31] <hobx_> no
[17:31] <i2p_iip> <Newsbyte> should, should...when you see the reality, for instance with that farmaceutical company that knowingly sold medicaments to ppl, even knowing for years it caused suicidal tendencies, then faith in corps is diminuishing. Faith in the regulators too, btw, because they were shown to have been severely lacking in spotting the problem, and coincidentally, many of the members of their board were ex-employees from that company
[17:31] <toad_> Newsbyte: well, that's murder, isn't it?
[17:31] <toad_> manslaughter or negligence or something
[17:31] <hobx_> Cartels are people volutarily working together. There is nothing wrong with that.
[17:31] <toad_> if the corps do that, the govt has a duty to catch them and punish them
[17:31] <toad_> hobx_: woah
[17:31] <greycat> Newsbyte: good example. Donnie Rumsfeld, before he became the czar we know and love today, helped push Aspartame into the US market over the objections of the scientists who studied it.
[17:31] <i2p_iip> <Newsbyte> it should be
[17:32] <toad_> who agrees with hobx that cartels are good?
[17:32] <hobx_> greycat: The free market does not equalize wealth because people are not equal. The free market places wealth where it is deserved.
[17:32] <plixed> hobx_: cartels kill the market
[17:32] <hobx_> bullshit
[17:32] <toad_> no, the free market is "to those who have, more will be given; to those who don't, what they have will be taken away"
[17:32] <toad_> most of the time
[17:33] <hobx_> Cartels produce an arbitrage opportunity for somebody to enter the market and get rich.
[17:33] <i2p_iip> <Newsbyte> if *I* poisoned a person, I would go faster to jail then I could blink my eyes. When THEY do it with dozens of people, nothing happens, exept an 'investigation' by the board of regulators, who were supposed to have protected the public in the first place.
[17:33] <toad_> http://amphibian.dyndns.org/simulations2-fetchfrac0.025-100x25.png
[17:33] <hobx_> Cartels are only harmful when backed by government sanctioned monopolies (regulations, copyrights, patents)
[17:33] <greycat> Newsbyte: correct. If Justice were served, Rumsfeld would have been in jail for poisoning US children.
[17:33] <toad_> Newsbyte: to be fair, sometimes they get done... and slapped on the wrist ;|
[17:33] <toad_> interesting new graph
[17:34] <toad_> hobx_: no, for example: M$ gives cut price deals to vendors who only sell Windows PCs
[17:34] <toad_> now, you can get around this by not having copyright
[17:34] <toad_> but there are similar practices elsewhere
[17:34] <hobx_> MS monopoly is backed by a government monopoly
[17:34] <hobx_> next
[17:34] <toad_> unions are similar
[17:34] <i2p_iip> <Newsbyte> it's a disgrace
[17:34] <toad_> unless there is a legal basis for unions, unions are illegal, because they are a breach of contract
[17:35] <i2p_iip> <Newsbyte> corps get away with things any other *real* person would not
[17:35] <toad_> but it is pretty clear that the only way to get decent pay and conditions of work, and to get looked after when there are incidents at work (accidents etc) is to have unions
[17:35] <i2p_iip> <Newsbyte> same with polticians, really
[17:36] <toad_> now does anyone like my graph? see the red-orange line, right on top of the purple?
[17:36] <toad_> that's with estimator passing
[17:36] <hobx_> The only way to get decent conditions at work is not to take the job unless it gives you decent conditions.
[17:36] * pupok (~r00t@81-178-94-146.dsl.pipex.com) Quit (Read error: 104 (Connection reset by peer))
[17:36] * sanity (~ian@81-178-94-146.dsl.pipex.com) Quit (Read error: 54 (Connection reset by peer))
[17:36] <hobx_> If you would be worse off not taking the job, then maybe the conditions are so indecent after all.
[17:37] * sanity (~ian@81-178-94-146.dsl.pipex.com) has joined #freenet
[17:37] <hobx_> s/are/aren't/
[17:37] <toad_> hobx_: well, I've made my case
[17:37] <i2p_iip> <Newsbyte> I remember a politician that (claimed to have) burned 1 million euro (illegal party funds and all that)....any other person would fly in jail ,and have a huge fine. You know what happend to that politician (minister, I believe)? He was 'slapped on the wrist' by sending him to a university-course abroud for two years, after which he just got back in the government
[17:38] <hobx_> Not to me you haven't.
[17:38] <toad_> hobx_: that's not my problem
[17:38] <hobx_> true enough.
[17:38] <toad_> I don't have to convince everyone of my viewpoint
[17:38] <i2p_iip> <Newsbyte> and toad is right: cartels suck. For that matter, raw capitalism sucks too.
[17:38] <plixed> i am not so shure about that, we have never seen it, like we have never seen raw socialism
[17:39] <toad_> I would just point to the numerous occasions throughout history where we have had systems very close to what hobx proposes
[17:39] <i2p_iip> <Newsbyte> I'm a liberal, but that doesn't mean I can't see the disadvantage of a non-social corrigated form of raw capitalism
[17:39] * greycat (~wooledg@192.35.79.70) Quit ("Lick Bush in '04")
[17:39] <toad_> and also, that there is clearly a regulatory role in at least that governments have to prevent corps from KILLING people and doing other "non-consensual" interactions
[17:39] <toad_> so it is not possible to abolish the state in favour of the corps
[17:39] <hobx_> 1) No we haven't.
[17:40] <toad_> we just disagree on what its role is
[17:40] <i2p_iip> <Newsbyte> raw socialism was communism. We have seen that, and it failed miserably.
[17:40] <toad_> hobx_: workers not being allowed to unionize, and consequently having shite working conditions? we see that all around the world, and we saw it in the west in the 19th century
[17:40] <toad_> and earlier
[17:40] <hobx_> 2) What does that have to do with it? Laws against violance against people have nothing to do with regulating enterprise, and I have never said that it should be legal.
[17:40] * nextgens (~nextgens@d80-170-167-224.cust.tele2.fr) has joined #freenet
[17:41] <nextgens> hi
[17:41] <i2p_iip> <Newsbyte> I think a mixture betwen the two is best. Communism or extreme forms of socialism just don't work, in general. Raw capitalism does work, for a small elite. A mixture of the two works for most.
[17:41] <toad_> hobx_: well, corporate manslaughter and manslaughter are completely different; we have different offences, or we should, because it's really hard to prove individual guilt
[17:41] <hobx_> The 19th century was not in any way a liberal market economy. It wasn't even a free society.
[17:42] <hobx_> toad_: Eh? Somebody pulls the fucking trigger.
[17:42] <hobx_> Somebody tells him to do it.
[17:42] <toad_> hobx_: MANSLAUGHTER
[17:42] <toad_> not murder
[17:42] <hobx_> What is so different?
[17:42] <toad_> yes, with murder
[17:43] <toad_> but with manslaughter, usually it's negligence, and you have to prove individual guilt; withholding research sometimes might elevate into murder though IMHO
[17:43] * kers (~kers@hd5e25c8f.gavlegardarna.gavle.to) has joined #freenet
[17:43] <hobx_> An enterprise is just people working together.
[17:43] <hobx_> You can make the same case for any time people work together.
[17:43] <toad_> okay, now how can i get some work done? hmmm
[17:43] <hobx_> going home now.
[17:44] <hobx_> later
[17:44] <toad_> maybe i can /ignore hobx_ ;)
[17:44] <toad_> seeya :)
[17:44] <toad_> hmmm
[17:44] <toad_> it was rising
[17:44] <toad_> now it seems to be falling
[17:44] <hobx_> Read Atlas Shrugged a couple of times.
[17:44] <i2p_iip> <Newsbyte> nah...the idea that you can leave it up to decide to corps about ethical behaviour is silly. Corps focus on profit; it's the only thing that counts. If they have to use sweatshops full of kids to make more profit, they will do it, it's as simple as that. Let corps govern, and you end up with the worst exesses of 19th century capitalism
[17:45] <toad_> yes
[17:45] <toad_> corps are LEGALLY REQUIRED to do anything necessary to profit
[17:45] <toad_> therefore they need to be regulated
[17:45] * pupok (~r00t@81-178-94-146.dsl.pipex.com) has joined #freenet
[17:45] <toad_> however, you could argue that enforcing laws on murder etc is enough
[17:45] <toad_> i think that's what hobx_ is saying
[17:46] <i2p_iip> <Newsbyte> well, ermm...not quite, toad
[17:46] <toad_> Newsbyte: articles of association. Granted they can't be required to do that which is illegal
[17:46] <toad_> but they are required to do anything they legally can to improve the bottom line
[17:46] <i2p_iip> <Newsbyte> you can make a perfect case for corporate government, as long as you don't think ethical behaviour is of any importance
[17:47] <toad_> anyway..
[17:47] * sanity (~ian@81-178-94-146.dsl.pipex.com) Quit (Read error: 232 (Connection reset by peer))
[17:47] * sanity (~ian@81-178-94-146.dsl.pipex.com) has joined #freenet
[17:47] <i2p_iip> <Newsbyte> enforcing laws on murder alone is not enough :-)
[17:48] <toad_> there we agree :)
[17:48] <i2p_iip> <Newsbyte> what about stealing? negligence? using kids in sweatshops? All these things are forbidden by the governments (in western countries), yet, they are not about murder
[17:48] <toad_> hehe
[17:48] <toad_> they're even consensual in most cases!
[17:49] <toad_> the kids are better off, because if they didn't, they wouldn't get to eat :)
[17:49] <toad_> of course if the parents were paid more, they could pay to feed their kids
[17:49] <toad_> but the jobs market is carefully stacked against them
[17:49] <i2p_iip> <Newsbyte> when you choice is limited to survive, it's no real consent
[17:50] <toad_> now, what shall I do while this sim is running?
[17:50] <i2p_iip> <Newsbyte> otherwise, you might argue that kids that prostitute themselves to get food are doing so consensual too
[17:52] <toad_> well, my argument is that unions are needed to get good conditions, that the jobs market is bound to be fixed up much of the time so can't be relied on to deliver, and that capitalism moves wealth towards those who already have it, rather than those who deserve it
[17:52] * sanity (~ian@81-178-94-146.dsl.pipex.com) Quit (Read error: 104 (Connection reset by peer))
[17:52] <plixed> who deserves wealth?
[17:52] <toad_> plixed: i'm not going to even get into that one
[17:52] <i2p_iip> <Newsbyte> and stealing isn't concensual neither. And then their is racism; it's no exeption that corps follow a rather racist line when employing people.
[17:53] <toad_> plixed: hobx basically said the market moves wealth to those who deserve it
[17:53] <i2p_iip> <Newsbyte> well, I do, plix, but that's only MHO ;-)
[17:53] <toad_> Newsbyte: well most of the time the "stealing" they do is legal
[17:54] <plixed> you can call the tax legal "stealing" too
[17:54] <i2p_iip> <Newsbyte> legal stealing sounds like an oxymoron
[17:54] <i2p_iip> <Newsbyte> legal stealing sounds like an oxymoron
[17:54] <i2p_iip> <Newsbyte> legal stealing sounds like an oxymoron
[17:54] <i2p_iip> <Newsbyte> legal stealing sounds like an oxymoron
[17:55] * toad_ hopes that was just the bot messing up
[17:57] * sanity (~ian@81-178-94-146.dsl.pipex.com) has joined #freenet
[17:58] <toad_> sanity: hi, http://amphibian.dyndns.org/simulations2-fetchfrac0.025-100x25.png
[17:58] <toad_> observe the brown-red line above the purple line
[17:58] * pittaman (~mistery_b@d51A410CA.kabel.telenet.be) Quit (Client Quit)
[17:58] <toad_> that's the difference that passing estimators makes
[17:58] <toad_> i'm currently working on dumping the estimators, datastore etc, every 100k reqs...
[17:58] <toad_> apart from that we're just wasting time debating politics; hobx_ started it :|
[17:59] <sanity> well, passing estimators definitely appears to help
[18:00] <hobx> plixed: Those who deserve wealth is those who make it. Tax is stealing.
[18:00] <sanity> hobx: eat the rich
[18:00] <toad_> eww, you might catch something
[18:00] <hobx> Violence has ever been the tool of lesser men.
[18:01] <toad_> just take all their money ;)
[18:01] <plixed> sanity: if love that movie
[18:01] <plixed> s/if/i/
[18:01] <toad_> lesser?!?!?
[18:02] <i2p_iip> <Newsbyte> hmm, strange that sometimes, there's an echo of what one writes :-)
[18:02] <toad_> plixed: what movie? i know of a song..
[18:02] <i2p_iip> <Newsbyte> so, ermm...cvs adjusted, yet?
[18:02] <yonkel-cluster> it has been close to 48 hours and my node is still super slow......what do i do?
[18:03] <plixed> toad_: that one http://www.imdb.com/title/tt0092944/
[18:03] <i2p_iip> <Newsbyte> yonkel; reseed?
[18:03] <sanity> "eat the rich" is also a rather extreme socialist slogan
[18:03] <toad_> yonkel-cluster: hmmm
[18:04] <toad_> yonkel-cluster: what do you mean by super slow?
[18:04] <plixed> the movie is rather extreme too :-)
[18:04] <i2p_iip> <Newsbyte> not that I'm worthy of giving such a technical advice, ofcourse...
[18:04] <plixed> but ultimatly funny
[18:04] <yonkel-cluster> toad_: meaning it take for freakin ever to retrieve pages and suc
[18:05] <yonkel-cluster> does it clear node data every time you restart freenet?
[18:05] <yonkel-cluster> because i have restarted often because i have updated
[18:05] <toad_> yonkel-cluster: no
[18:05] <toad_> yonkel-cluster: how many connections according to Open Connections page?
[18:06] <yonkel-cluster> is that in advanced mode?
[18:06] * d-ArkAngel (~Rob@cpc2-midd2-5-0-cust16.midd.cable.ntl.com) has joined #freenet
[18:06] <d-ArkAngel> evening all
[18:06] <yonkel-cluster> hey
[18:06] <dli_> d-ArkAngel, not all :(
[18:06] <i2p_iip> <Newsbyte> evening
[18:06] <toad_> yonkel-cluster: yes, in advanced mode
[18:07] <i2p_iip> <Newsbyte> plix, you wouldn't be the one offering to translate freenetstuff in spanish, are you?
[18:07] <toad_> if you have less than 20 conns, you have problems; if you have more than 50, it's definitely caused by something else...
[18:07] <toad_> bbiab
[18:07] <d-ArkAngel> hey toad, what are the two sims running on my box? Not wanting to kill them, just wondering how much data you've got out of 48hours of them running :-)
[18:08] <yonkel-cluster> toad_: 7 conns
[18:08] <i2p_iip> <Newsbyte> angel, perhaps this one? http://amphibian.dyndns.org/simulations2-fetchfrac0.025-100x25.png
[18:09] <yonkel-cluster> toad_: 7 conns all outbound
[18:09] <i2p_iip> <Newsbyte> yonk, that sucks
[18:10] <yonkel-cluster> i2p_iip: well it might be a nat problem
[18:10] <i2p_iip> <Newsbyte> I know exactly how that feels; I've been there myself, and I suspect it's rather a shortcomming
[18:10] <dli_> i2p_iip, could it be a firewall problem?
[18:10] <i2p_iip> <Newsbyte> short-way solution: reseed
[18:10] <yonkel-cluster> i have 2 nats and an ethernet bridge in between me and the internet
[18:11] <d-ArkAngel> look at the connections pics, if it shows any incomming as well then it's working ok...
[18:11] <yonkel-cluster> wait do i need to forward the listenport or the clientport?
[18:11] <i2p_iip> <Newsbyte> that is possible to, I had a NAT too. Since I made an extra hole and the reseeding at the same time, I'm not sure what did the trick exactly
[18:11] <d-ArkAngel> listenport
[18:12] <d-ArkAngel> and your ipAddress needs to be set as the outer most IP address
[18:12] <i2p_iip> <Newsbyte> yeah, not the 8888 or such
[18:12] * nextgens (~nextgens@d80-170-167-224.cust.tele2.fr) Quit ("Download Gaim: http://gaim.sourceforge.net/")
[18:12] <yonkel-cluster> ok let me double check
[18:13] <i2p_iip> <Newsbyte> if the list of nodes don't grow rather fast, try reseeding
[18:13] <i2p_iip> <Newsbyte> the open connections, I mean
[18:13] <yonkel-cluster> i2p_iip: how does reseeding work? and what is with the "<newsbyte>" thing?
[18:14] <d-ArkAngel> if you need to know what IP address you're seen as from the outside world you could look here. http://www.urgentclick.com/whats_my_ip_address.php
[18:15] <i2p_iip> <Newsbyte> and maybe you can find some hints on www.freenethelp.org too - or were you already aware of that wikisite?
[18:15] <i2p_iip> <Newsbyte> and maybe you can find some hints on www.freenethelp.org too - or were you already aware of that wikisite?
[18:15] <i2p_iip> <Newsbyte> and maybe you can find some hints on www.freenethelp.org too - or were you already aware of that wikisite?
[18:15] <d-ArkAngel> the the <newsbyte> thing is his nick name. The i2p_iip is a bot that sits in the other chat chanel and ports the messages back and forth so that everyone can chat
[18:15] <toad_> okay, restarted it with a slightly improved estimator passing
[18:15] <toad_> bbiab
[18:15] <toad_> yonkel-cluster: you must forward the listenPort
[18:16] <toad_> bbl
[18:17] <i2p_iip> <Newsbyte> three times, again. Wonder what's causing that (?)
[18:17] * FallingBuzzard (~srademach@207.152.112.129) Quit (Read error: 104 (Connection reset by peer))
[18:19] <yonkel-cluster> wait i think i fixed something
[18:19] <yonkel-cluster> still all outbound
[18:19] <yonkel-cluster> what do i do? rrrrrrr
[18:19] <d-ArkAngel> it might take a while for someone to make a connection to you once you've fixed it
[18:19] <plixed> afaik you don't get inbound conns unless your node sends out announcements
[18:19] <i2p_iip> <Newsbyte> did you try all the suggestions made?
[18:21] <plixed> or it will take it a very long time without them
[18:22] <d-ArkAngel> well since bidi is in then you should be able to get noticed even with only a few connections... but as you say it might take some time
[18:22] <yonkel-cluster> the problem seems to be with the ethernet bridge
[18:22] <i2p_iip> <Newsbyte> well, actually, with 20 active connections, it should augment rather rapidly. In 30 minutes, he could have 50+ open connections
[18:22] * sanity (~ian@81-178-94-146.dsl.pipex.com) Quit (Read error: 104 (Connection reset by peer))
[18:22] <plixed> my observation is: the first inbound conns will arrive after the "wake announcement" in the ticker happend
[18:22] <d-ArkAngel> you could try and connect to the port you have opened with telnet from a shell acount externaly and make sure that the connection really occurs
[18:23] * sanity (~ian@81-178-94-146.dsl.pipex.com) has joined #freenet
[18:24] <d-ArkAngel> also after any change to your conf file you'll need to stop and then start your node again
[18:25] <plixed> a nice goodie would be if freenet is able to determine if it's firewalled or not and say so on start page / general info
[18:26] <yonkel-cluster> the connection doesn't even seem to be made!
[18:26] <yonkel-cluster> yeah but in the meantime.....i cannot get my node to work
[18:26] <d-ArkAngel> have you managed to forward any other ports to the same machine successfully?
[18:27] <yonkel-cluster> never tried
[18:27] <yonkel-cluster> should i reconfig and see if it picks another listenport and then try that?
[18:27] <d-ArkAngel> probably won't help
[18:27] <d-ArkAngel> it's not likely to be the port number
[18:28] <d-ArkAngel> so you've setup both NAT's to forward correctly?
[18:28] <yonkel-cluster> well i think so
[18:28] <yonkel-cluster> but there is a problem!
[18:28] <yonkel-cluster> i have this ethernet bridge in the way
[18:28] <yonkel-cluster> it connects my one router to my wlan
[18:28] <yonkel-cluster> i think that is the problem
[18:29] <plixed> can't, or it wouldn't be a bridge
[18:29] <d-ArkAngel> can you log into the NAT machines? or are they dedicated hardware boxes?
[18:29] <yonkel-cluster> yes and yes
[18:29] <yonkel-cluster> they are linksys routers
[18:29] <yonkel-cluster> one is the wrt54g
[18:30] * lolo-laptop (~lostlogic@68.251.84.226) Quit ("Client exiting")
[18:30] <d-ArkAngel> why do you have them both running NAT?
[18:30] <yonkel-cluster> ummmmm
[18:30] <yonkel-cluster> i dunno
[18:30] <d-ArkAngel> surely you can't be using the whole 192 range of IP's
[18:30] <yonkel-cluster> never really thought about it
[18:30] <yonkel-cluster> how do i turn it off?
[18:31] <d-ArkAngel> are you sure the second one is running NAT?
[18:31] <yonkel-cluster> i am not sure if the 1st one it
[18:31] <d-ArkAngel> I don't know how you turn it off I don't know enough about your setup...
[18:32] <d-ArkAngel> do you have machines between the two nat boxes?
[18:32] <plixed> so this wlan-thingy is not a bridge but a nat-router?
[18:33] <yonkel-cluster> here: here is a picture of my network
[18:33] <yonkel-cluster> ftp://www.bluefacemonkey.com/magen-net.jpg
[18:33] <d-ArkAngel> I only see one NAT layer there....
[18:33] <d-ArkAngel> from Family winXP can you ping your machine?
[18:34] <yonkel-cluster> good question....answering it means going upstairs
[18:34] <yonkel-cluster> dammit
[18:34] <d-ArkAngel> or from any of the other machines.
[18:34] <yonkel-cluster> wait i doubt it
[18:34] <yonkel-cluster> one sec i have an idea
[18:34] <d-ArkAngel> from the cluster, of the laptop
[18:39] <d-ArkAngel> also are you running any firewall software on the machine that's got the node running on it?
[18:56] * zorton (zorton@circle.xyxx.com) has joined #freenet
[18:56] <zorton> i'm hoping to improve my freenet performance a bit
[18:56] <zorton> first: should I be running the stable or unstable network?
[18:56] <d-ArkAngel> well that kinda depends.
[18:57] <d-ArkAngel> stable is probably you best bet if you want it to "just work"
[18:57] <zorton> which has the most content thought?"
[18:57] <d-ArkAngel> probably stable (in theory it's got the most nodes too, but you can't really tell with freenet ;-) )
[18:57] <zorton> heh
[18:58] <zorton> well i've got a stable node up and running on a shell machine
[18:58] <zorton> and I setup a ssh tunnel to the shell and fire up firefox to connect to the node
[18:58] <zorton> however it runs terrible
[18:58] <d-ArkAngel> and is it accepting incomming connection properly?
[18:58] <zorton> not sure how I would be able to tell
[18:59] <zorton> the machine is not behind a firewall in anyway if tha'ts what you mean
[18:59] <d-ArkAngel> got to the node frontpage, then click switch to advanced mode
[18:59] <d-ArkAngel> then on the menu on the left hit Open Connections
[19:00] <zorton> kay
[19:00] <d-ArkAngel> that should show you how many connections your node has
[19:00] <d-ArkAngel> Connections open (Inbound/Outbound/Limit) 84 (46/38/200)
[19:00] <d-ArkAngel> is mine for example
[19:00] <i2p_iip> <Newsbyte> angel, did you re-install a new seednodes.ref yet?
[19:00] <i2p_iip> <Newsbyte> I mean, yonkel
[19:00] <d-ArkAngel> nope
[19:01] <d-ArkAngel> ok right ;-)
[19:01] * yonkel-cluster (~yonkeltro@pcp04664407pcs.wilog501.pa.comcast.net) Quit (Client Quit)
[19:01] <d-ArkAngel> he went to try and ping his machine from elsewhere on the netwrok, I think he's not got the NAT forwarding right.
[19:02] <i2p_iip> <Newsbyte> it shouldn't be all that of a problem, it's supposed to wrok with nat, now
[19:02] <zorton> Connections open (Inbound/Outbound/Limit) 39 (0/39/200)
[19:02] <zorton> looks like inbound isn't work too well
[19:02] <d-ArkAngel> yeah, is there a firewall on the box your running it on?
[19:03] <zorton> doubtfull
[19:03] <i2p_iip> <Newsbyte> it's all on www.freenethelp.org! ;-)
[19:03] <zorton> wrong IP in the config file
[19:04] <d-ArkAngel> that'll be the ticket
[19:04] <plixed> zorton: try telneting to you freenbox box on the ListenPort
[19:04] <d-ArkAngel> you can use a DNS name, if the box is at risk of moving IP's
[19:05] <d-ArkAngel> you'll need to stop, and then restart the node before it picks up changes in the config file.
[19:05] <zorton> she connects but the port closes right away
[19:05] <zorton> Connections open (Inbound/Outbound/Limit) 44 (34/10/200)
[19:05] <d-ArkAngel> that's looking better
[19:06] <d-ArkAngel> should improve performance a little (more connections is always better)
[19:06] <d-ArkAngel> (untill your box can't cope any more of course :-) )
[19:07] <zorton> however I think I have problems with firefox
[19:07] <d-ArkAngel> other things you can do to help performance would be give it a larger store, the more stuff on your node, the more chance other nodes will want to connect to you, and the better connection to the network will evolve
[19:07] <zorton> if I open a bunch of tabs to firefox and try browsing freenet it seems to lock up
[19:07] <zorton> however I have set the http,max connect parm to 500 or so
[19:07] <d-ArkAngel> have you made the tweaks to increase the number of simultanious file downloads?
[19:08] <d-ArkAngel> and the max per site ones as well?
[19:08] <i2p_iip> <Newsbyte> wasn't there something said about having to change the max allowed of open connections on firefox?
[19:08] <i2p_iip> <Newsbyte> right. well, that can't be it, then
[19:09] <d-ArkAngel> because to firefox all freesites look like the same server, so you need to jump the connections to server up as well as the general maximum connections
[19:10] <zorton> any documentation or a wiki on this?
[19:10] <d-ArkAngel> yup somewhere :-)
[19:10] <zorton> heh
[19:10] <i2p_iip> <Newsbyte> certainly.wiki at www.freenethelp.org
[19:10] <zorton> well if you know the param I could tweak it
[19:10] <zorton> Newsbyte: searching there now
[19:11] <d-ArkAngel> I'm using
[19:11] <d-ArkAngel> network.http.max-connections : 1148
[19:11] <d-ArkAngel> network.http.max-connections-per-server : 1024
[19:11] <d-ArkAngel> network.http.pipelining : true
[19:11] <d-ArkAngel> network.http.pipelining.maxrequests : 100
[19:11] <d-ArkAngel> in about:config
[19:12] <dli_> d-ArkAngel, do I need to set those for konqueror?
[19:12] <d-ArkAngel> all they do is speed things up a little, you don't need to set them.
[19:12] <i2p_iip> <Newsbyte> you can do a search on keyword(s) too
[19:12] <i2p_iip> <Newsbyte> you can do a search on keyword(s) too
[19:13] <d-ArkAngel> it's because there's soo much lag between freenet on each http request that it's far better to do them in parallel than in series as would be the case on the WWW
[19:13] <zorton> I think I may be running into a limit of ssh as well
[19:13] <zorton> i'm connecting to the freenet web interface through a ssh proxy to my shell machine as well
[19:14] <d-ArkAngel> where is the node? out on the open internet?
[19:14] <zorton> yes
[19:14] <zorton> limited to listening from 127.0.0.1
[19:14] <d-ArkAngel> and how paranoid are you?
[19:15] <zorton> not terrabily
[19:15] <zorton> better way to do it?
[19:15] <zorton> i'm putting the node on a shell machien for performance and firewall issues
[19:15] <i2p_iip> <Newsbyte> it's not paranoia if they are really after you! ;-)
[19:15] <zorton> every location I can get to the internet from allows ssh but not neccessarly freenet connections
[19:15] <d-ArkAngel> well you could always set it up to listen for connections via your IP address, and then conect
[19:16] <zorton> oh sure
[19:16] * kers (~kers@hd5e25c8f.gavlegardarna.gavle.to) Quit (Read error: 104 (Connection reset by peer))
[19:16] <d-ArkAngel> directly to the node rather than use the ssh tunnel, if you think that's really a limitation.
[19:16] <i2p_iip> <Newsbyte> if you search for 'browser' or 'open connections' or 'firefox' or something, you might find it quicker in the wiki
[19:16] <d-ArkAngel> he's already got that set Newsbyte
[19:16] <i2p_iip> <Newsbyte> I do remember *something* is said about it, anyway
[19:17] <d-ArkAngel> zorton: do you have a proxy server setup?
[19:17] <zorton> a regular http proxy server?
[19:17] <d-ArkAngel> yeah
[19:17] <zorton> yep
[19:18] <zorton> same thing, ssh tunnel to a diffrent machine
[19:18] <zorton> firefox is my "secure" browser
[19:18] <d-ArkAngel> but you've got localhost in the exceptions right?
[19:18] <i2p_iip> <Newsbyte> well, idem for all other questions you may have (and I'm returning to play my game :-)
[19:18] <i2p_iip> <Newsbyte> stupid dragons destroyed my castle :-/
[19:19] <zorton> as far as firefox's proxy configuation goes yep
[19:19] <d-ArkAngel> (just because those network.http settings have couterparts for proxy connections, but if the connections aren't going through the proxy then it's not a problem.
[19:21] <zorton> might be the problem
[19:21] <zorton> i've got the tunnel set to debug
[19:21] <zorton> and i'm only seeing about 7 or 8 channels
[19:21] <d-ArkAngel> here's an interesting thought... you *could* run freenet on your local host. assuming you've got decent bandwithd connection, and use the SSH port forwarding to allow connections the other way... port forward the incomming freenet conenctions, rather than your http conenctions to the node...
[19:21] * pupok (~r00t@81-178-94-146.dsl.pipex.com) Quit (Read error: 104 (Connection reset by peer))
[19:21] <zorton> is interesting idea though
[19:21] * sanity (~ian@81-178-94-146.dsl.pipex.com) Quit (Read error: 54 (Connection reset by peer))
[19:22] <zorton> I could just give up and see if I can run freenet on the localhost :)
[19:22] <d-ArkAngel> just a thought tho, might be better of leaving the node running for a day or so now you've got the incomming connections working properly.
[19:22] <d-ArkAngel> and see if that improves the situation
[19:22] * sanity (~ian@81-178-94-146.dsl.pipex.com) has joined #freenet
[19:22] <zorton> well the problem i'm pretty certain is in firefox
[19:22] <d-ArkAngel> freenet "should" work without the need for incomming connections, since the connections are bi-directional these days anyway.
[19:23] * pupok (~r00t@81-178-94-146.dsl.pipex.com) has joined #freenet
[19:23] <d-ArkAngel> but it seems to still function better being able to accept connections as well (get integrated into the network better)
[19:24] <plixed> d-ArkAngel: you will always need some nodes that are able accept incoming connections or you will not have a network at all
[19:24] * kers (~kers@hd5e25c8f.gavlegardarna.gavle.to) has joined #freenet
[19:25] <d-ArkAngel> indeed, but nodes will function in situations where incomming conections are not feasable
[19:25] <zorton> Connections open (Inbound/Outbound/Limit) 27 (0/27/128)
[19:25] <zorton> bummer, dosen't look like that's working :(
[19:25] <d-ArkAngel> for example the ones that people run at a Uni that have T1 internet conenctions, but are behind a firewall beyond their control :-)
[19:26] <d-ArkAngel> it's certainly better to have the connections, but in the absence of the ability to do it it's better than no node at all :-)
[19:27] <plixed> d-ArkAngel: the freenet uses UDP instead of TCP it would be possible to eleminate the need for portforwarding.
[19:27] <d-ArkAngel> zorton: doh, and it looked so promising :-)
[19:27] <plixed> s/the/if/
[19:27] <zorton> well that's the local node
[19:27] <zorton> running on the local machine
[19:27] <zorton> funny thing is I can telnet to the beast from the big bad internet
[19:27] <d-ArkAngel> olixed.... ummm no it wouldn't UDP would FORCE the need for port forwarding...
[19:28] <toad_> woah
[19:28] <toad_> Overall success ratio: 0.9876
[19:28] <toad_> Overall success ratio: 0.9872
[19:28] <toad_> Overall success ratio: 0.9876
[19:28] <toad_> Overall success ratio: 0.9896
[19:28] <d-ArkAngel> nice what's that with?
[19:28] <toad_> Overall success ratio: 0.9876
[19:28] <toad_> Overall success ratio: 0.9896
[19:28] <plixed> you are cheating, toad ;-)
[19:28] <toad_> Overall success ratio: 0.9908
[19:28] <toad_> Overall success ratio: 0.982
[19:28] <toad_> Overall success ratio: 0.994
[19:28] <toad_> Overall success ratio: 0.9912
[19:28] <toad_> these obviously aren't the averaged ones
[19:28] <d-ArkAngel> they those long runners on my box?
[19:28] <toad_> no
[19:28] <toad_> i'm not using the remotes at the moment
[19:28] <d-ArkAngel> so what stats are they?
[19:28] <toad_> because one of them is down
[19:29] <toad_> (not yours)
[19:29] <toad_> these are with estimator passing enabled
[19:29] <toad_> i'm not sure i've checked that in
[19:29] <d-ArkAngel> mines got 400x25 and 400x50 that have been running for a little over 48hrs now :-)
[19:29] <d-ArkAngel> that you set going sat night.
[19:29] <plixed> d-ArkAngel: no. all nat-implementations hold the udp-port open for a short time to wait for possible replies, all you need to do is to send regularly udp packets out and the port is open
[19:29] <d-ArkAngel> :-)
[19:30] <toad_> http://amphibian.dyndns.org/simulations2-fetchfrac0.025-100x25.png
[19:30] <toad_> observe the graph!
[19:30] <d-ArkAngel> yeah, but that wouldn't allow new servers to connect to you...
[19:30] <toad_> now i read backlog
[19:33] <d-ArkAngel> and I'm fairly certain that not all nat-implementations have wait gates on UDP connections
[19:34] <plixed> ok, many or most
[19:35] <d-ArkAngel> and you'd still need port forwards to accept new connections
[19:35] <d-ArkAngel> because they only allow related trafic through, not any trafic on that port
[19:35] <toad_> <d-ArkAngel> network.http.max-connections : 1148
[19:35] <toad_> <d-ArkAngel> network.http.max-connections-per-server : 1024
[19:35] <toad_> <d-ArkAngel> network.http.pipelining : true
[19:35] <toad_> <d-ArkAngel> network.http.pipelining.maxrequests : 100
[19:35] <toad_> <d-ArkAngel> in about:config
[19:35] <toad_> woah
[19:36] <d-ArkAngel> so I like tabs :-)
[19:36] <toad_> unfortunately, the stellar performance achieved by passing estimators comes at a massive performance cost on the simulations... maybe there are ways to improve it though
[19:36] <d-ArkAngel> cost as in slowdown?
[19:38] <toad_> <zorton> if I open a bunch of tabs to firefox and try browsing freenet it seems to lock up
[19:38] <toad_> <zorton> however I have set the http,max connect parm to 500 or so
[19:38] <toad_> well, there's a limit on the node side as well
[19:38] <plixed> d-ArkAngel: do you know skype? this thingy is able to accept UDP packets without portforwarding. i am not completly shure how it works
[19:38] <toad_> because fproxy downloads cost threads...
[19:38] <d-ArkAngel> if a real node disconnects from a peer, does it lose the estimator info it has for that node, or does it story it somewhere?
[19:38] <toad_> it's 24-36 simultaneous HTTP downloads
[19:39] <d-ArkAngel> plixed. you can ONLY accept packats if you have SENT a packet to that server recently.
[19:39] * kers (~kers@hd5e25c8f.gavlegardarna.gavle.to) Quit ("Leaving")
[19:39] <d-ArkAngel> OR if there is a port forward rule that tells the server where to send it.
[19:39] <toad_> <i2p_iip> <Newsbyte> stupid dragons destroyed my castle :-/ - what are you playing Newsbyte?
[19:40] <toad_> <d-ArkAngel> freenet "should" work without the need for incomming connections, since the connections are bi-directional these days anyway. - i think that it's mainly a bootstrapping problem...
[19:40] <plixed> d-ArkAngel: with udp it seems to be possible to circumvent this limitation
[19:40] <toad_> but we should do something about it eventually...
[19:41] <d-ArkAngel> plixed. trust me that you can't bypass that limitation
[19:41] <toad_> <plixed> d-ArkAngel: the freenet uses UDP instead of TCP it would be possible to eleminate the need for portforwarding. - not necessarily, it depends on the firewall. there is no single way to bypass ALL firewalls.
[19:42] <toad_> yeah, it's 14 seconds per cycle on 100x25
[19:42] <toad_> which is pretty crap
[19:42] <d-ArkAngel> ooooow
[19:42] <plixed> toad_: correct. but for many or most is seems to be possible
[19:43] <plixed> well, the skype docs explicitly state that portforwarding is not needed and i observed direct udp connections to another one without any portforwaring.
[19:43] <d-ArkAngel> yes, but that's because your box connected to them
[19:43] <d-ArkAngel> they didn't connect to you
[19:43] <d-ArkAngel> so that's an outgoing connection.
[19:43] <toad_> plixed: I agree that UDP is an interesting possibility
[19:43] <toad_> but there are others
[19:44] <plixed> there was no portforwarding on the other side, so i shouldn't be able to connect to him
[19:44] <toad_> e.g. routing a connection message via the request chain
[19:44] <toad_> and having them connect to you
[19:44] <zorton> well i'm trying to switch teh shell machine over to allowing a direct web browser connection to the front end on port 8888 from my IP
[19:44] <toad_> http://amphibian.dyndns.org/simulations2-fetchfrac0.025-100x25.png
[19:44] <zorton> however when I try and load the address http://shell.machine.around:8888 nothing is returned
[19:44] <toad_> nice graph, no? the red/brown line...
[19:44] * 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!")
[19:45] <toad_> zorton: you started loading a bunch of stuff, and now it won't load anything else, because it ran into the connection limit in fproxy?
[19:45] <d-ArkAngel> indeed
[19:45] <toad_> e.g. you are loading TFE?
[19:45] <zorton> I don't know where the bottleneck is
[19:45] <toad_> zorton: that's my suspicion - have you been fetching lots of stuff?
[19:46] <toad_> in lots of windows, all simultaneously, via fproxy?
[19:46] <zorton> here is the setup: me -> ssh_tunnel_to_thy_shell -> 127.0.0.1:8888 on shell -> freenet node
[19:46] <d-ArkAngel> how much data store are you providing zorton?
[19:47] <toad_> okay, next....
[19:47] <zorton> 256M
[19:47] <toad_> add an averaged success ratio based on the last 10,000 requests, and only the last 10,000 requests
[19:47] <zorton> bummer
[19:47] <zorton> thought it was set higher
[19:47] <toad_> that will take up approx 10,000 bits = 1kB
[19:48] <d-ArkAngel> assuming there's no overheads... :-)
[19:49] <zorton> well anyway, done for now, i'll be back tommarow for more fun :)
[19:49] <d-ArkAngel> yeah, I need a good nights sleep before tomorrow, I've got an auction on tue-thu this week.
[19:50] <d-ArkAngel> So I think I'll turn in for the night now as well.
[19:50] <plixed> good night :-)
[19:50] * dli_ (~tst@anna-bhat.uchicago.edu) Quit (Read error: 110 (Connection timed out))
[19:50] <d-ArkAngel> night all
[19:52] <toad_> seeya
[19:52] * sanity (~ian@81-178-94-146.dsl.pipex.com) Quit (Read error: 104 (Connection reset by peer))
[19:52] <d-ArkAngel> (one last comment to cause trouble before I go) Is estimator passing vunerable to exploits of people passing false estimators?
[19:52] <toad_> yes, therefore we have to impose limits on the level of specialization displayed
[19:52] <toad_> and be skeptical about claimed experience levels
[19:53] <toad_> e.g. we only use 200 bits out of 1000 on a binary running average
[19:54] <d-ArkAngel> cool. I'll leave you to it. Sorry I've not got this threaded version running yet. (one thing, in the single threaded sim you might be able to improve performance by using a non synchronised version of LRUqueue
[19:55] <d-ArkAngel> as the synchronisation isn't really needed in situations where it's only called by a single synchronised object, or run in a single threaded environment...
[19:57] * sanity (~ian@81-178-94-146.dsl.pipex.com) has joined #freenet
[19:59] <d-ArkAngel> toad you want these 400x25 sims leaving running?
[20:00] <d-ArkAngel> what's with the _ on the end of everyone's name? is it a fashion statement? :-)
[20:02] * hobx_ (~chatzilla@ankh.math.chalmers.se) Quit (burroughs.freenode.net irc.freenode.net)
[20:03] <d-ArkAngel> night all
[20:03] * d-ArkAngel (~Rob@cpc2-midd2-5-0-cust16.midd.cable.ntl.com) Quit (Read error: 54 (Connection reset by peer))
[20:07] * hobx_ (~chatzilla@ankh.math.chalmers.se) has joined #freenet
[20:07] * sanity (~ian@81-178-94-146.dsl.pipex.com) Quit (Read error: 104 (Connection reset by peer))
[20:12] * sanity (~ian@81-178-94-146.dsl.pipex.com) has joined #freenet
[20:14] <toad_> d-ArkAngel: hmmm
[20:14] <toad_> yes, if poss
[20:14] <toad_> bbl
[20:16] * toad_ running 100x25 simulations with HTL of 12,11,10,9,8,7,6
[20:17] <toad_> overnight
[20:21] <toad_> okay
[20:21] <toad_> 12 ... 1
[20:22] <toad_> bbl
[20:22] * sanity (~ian@81-178-94-146.dsl.pipex.com) Quit (Read error: 54 (Connection reset by peer))
[20:22] * sanity (~ian@81-178-94-146.dsl.pipex.com) has joined #freenet
[21:32] * guido^pe (~unknown@dsl-213-023-245-024.arcor-ip.net) has joined #freenet
[21:41] * Sugadude (~Sugadude@the.dogs.bollo.cx) has joined #freenet
[22:03] * guido^pe (~unknown@dsl-213-023-245-024.arcor-ip.net) Quit (Remote closed the connection)
[22:22] * icekey (~yosukey@cat.cs.mu.OZ.AU) has joined #freenet
[22:22] <icekey> Hello
[22:23] <icekey> We've just setup a mini freenet to test out something that we're developing
[22:23] <icekey> and noticed that it takes about 50 seconds to insert a file
[22:23] <icekey> on a small number of nodes
[22:24] <icekey> anyone know why it takes so long?
[22:32] * icekey (~yosukey@cat.cs.mu.OZ.AU) Quit (Remote closed the connection)
[22:33] * icekey (~yosukey@cat.cs.mu.OZ.AU) has joined #freenet
[22:42] <icekey> hmm
[22:46] * yonkeltron (~yonkeltro@pcp04664407pcs.wilog501.pa.comcast.net) Quit (Read error: 104 (Connection reset by peer))
[23:00] * Zv2S_ (~rfc1413@squirrel.owl.de) has joined #freenet
[23:00] * Zv2S (~rfc1413@squirrel.owl.de) Quit (Read error: 104 (Connection reset by peer))
[23:00] * Zv2S_ is now known as Zv2S
[23:01] * icekey (~yosukey@cat.cs.mu.OZ.AU) Quit ("Leaving")
These logs were automatically created by Jay Oliveri with his gimp hapi on irc.freenode.net.
