Timestamps are in GMT/BST.
[0:00] <mikeeusa> yo TheSeeker a friend of mine used to use arrays as subroutines
[0:00] <mikeeusa> in perl
[0:00] <mikeeusa> and it worked
[0:00] <i2p_iip> <stefanboy> an object is a self-contained unit consisting of methods (functions) and member variables. you create a pattern (called a "class") and then create instances of that class and utilize the objects in your program
[0:00] <TheSeeker> perl ... *shudder*
[0:00] <i2p_iip> <stefanboy> much cleaner and more organized than most "top-to-bottom" languages
[0:01] <i2p_iip> <stefanboy> (but that's not to say top-to-bottom languages aren't good... C is still my first choice)
[0:01] <TheSeeker> idealy I would then have each node in it's own thread, with the nodes runnign themselves automatically isntead of having the 'main loop' do the work? that's somethign I can't grasp the concept of very well.
[0:02] <mikeeusa> perl ROX
[0:02] <mikeeusa> ROX
[0:02] <TheSeeker> I H8 perl.
[0:02] <kmag> no, it's generally a problem to have each object in its own thread
[0:03] <kmag> lots of threads bog down the system
[0:03] <TheSeeker> then how are you supposed to "make the objects interact" ?
[0:03] <TheSeeker> if not by some "main loop"
[0:03] <i2p_iip> <stefanboy> no need for threads for what you're doing
[0:04] <kmag> you still generally use a main loop in OO programs
[0:05] <kmag> think of OOP as a way to factor your code
[0:06] * Ash-Fox (HAL-9000@adm245.neoplus.adsl.tpnet.pl) Quit (Read error: 104 (Connection reset by peer))
[0:06] <kmag> instead of writing send_msg_to_node(sender, receiver, msg)
[0:06] <kmag> you write sender.send(msg, receiver)
[0:07] * Ash-Fox (HAL-9000@adm245.neoplus.adsl.tpnet.pl) has joined #FreeNET
[0:07] <kmag> you put all of the code dealing with nodes in one place
[0:07] <TheSeeker> all the code would be in one place... main() :P
[0:07] <kmag> and code not dealivng with nodes in another place :-P
[0:08] <mikeeusa> PRL RX!
[0:08] <mikeeusa> U R GEH!
[0:08] <kmag> OO Perl is interesting
[0:09] <TheSeeker> but this is a sim of a bunch of nodes, there is no 'other code' it's all about the nodes. I'm not writing some universally useful bit of code that can be re-used in a dozen places... I'm writing a proof of concept routing for freenet bit of code which won't be used ever again :P
[0:09] <kmag> so the next sim you write you will write from scratch?
[0:10] <TheSeeker> I very much doubt I'll have a "next sim" ...
[0:10] <TheSeeker> and even if I do, it's unlikely to have anythign to do with freenet, and will require completely different code anyway...
[0:11] <kmag> okay, I'm not trying to talk you into using an OO language
[0:11] <TheSeeker> I'm just not lazy enough to be a good coder, SORRY :P
[0:11] <kmag> just trying to explain it
[0:12] <kmag> I wish I understood Prolog better
[0:12] <TheSeeker> I'm pretty friggen lazy, but not lazy enough to figure out ways I can possibly save myself from ever having to write 'those 4 somewhat useful lines of code' ever again in my life...
[0:13] <kmag> I write most of my code in C or a small subset of C++
[0:13] <mikeeusa> TheSeeker, why you hate perl?
[0:13] <mikeeusa> *
[0:13] <mikeeusa> correction
[0:14] <mikeeusa> TheSeeker, why you hate perl mon?
[0:15] <kmag> mikeusa: ever try Python?
[0:16] <TheSeeker> have you seen what perl syntax looks like? it's NOT human readable stuf, no way to begin to understand what a line of it does if you don't have 5 lines of documentation to go along with it.
[0:17] * Hirvox (~hirvox@cs181124095.pp.htv.fi) Quit (Read error: 54 (Connection reset by peer))
[0:18] <TheSeeker> snippet of a random perl program found on the net:
[0:18] <TheSeeker> $shift="+8P,!`V,#\"`5C!RH!!/\\+/,\"!s(oh)+!()+\015+\\\@oh=~s///\"";
[0:19] <kmag> Perl is very utilitarian
[0:21] <kmag> I'm generally not a fan of dynamically typed languages
[0:21] <TheSeeker> perl is barely a step up from ASM as far as I'm concerned where readability and undestandability are concerned. I have trouble sometimes figuring out what I wrote in XBasic if I look at something long enough afet I wrote it ... and I tend to use fairly descriptive variables. How would anyone be able to pick up where anyone else left of in a perl app if the original coder were to die or soemthing?
[0:21] <kmag> Sounds like you need to comment more
[0:21] <TheSeeker> XBasic and Euphoria both require declaration of variable types before use...
[0:22] <TheSeeker> though Euphoria is wacky and uses integers atoms and sequences and that's about it x.x
[0:23] <TheSeeker> The types: object, sequence, atom and integer are predefined. Variables of
[0:23] <TheSeeker> type object may take on *any* value. Those declared with type sequence must
[0:23] <TheSeeker> always be sequences. Those declared with type atom must always be atoms. Those
[0:23] <TheSeeker> declared with type integer must be atoms with integer values from -1073741824
[0:23] <TheSeeker> to +1073741823 inclusive.
[0:24] <TheSeeker> ack, linebreaks :/
[0:25] <TheSeeker> anyway... I was going to go to sleep a long time ago...
[0:26] <kmag> 31-bit integers?
[0:26] <mikeeusa> TheSeeker, i can read perl just fine
[0:26] <kmag> k. sleep well Seeker
[0:26] <mikeeusa> usually dont have probs editing other peeps perl
[0:28] <kmag> you can write messy code in any language and you can write readable, clean code in just about any language outside of bf, ook, and kin
[0:29] <kmag> I wonder how long it takes to learn bf sufficiently well to write a quine
[0:29] <kmag> and if a quine has ever been written in ook
[0:44] <kmag> so.. back to 1.5.0 JVMs... anyone running their node on one?
[0:48] <KenMan> i did for a while, worked okay without problems.
[0:48] <KenMan> In fact, I ran a stable node, and I ran a small cluster of local nodes used for testing things out. 1.5 never caused me any problems...
[0:50] <kmag> KenMan: did it claim to be using a 1.4.2 JVM?
[0:50] <KenMan> I can't remember. But that wouldn't surprise me. Where do you see that reported ? logs or web interface ?
[0:51] <kmag> stdout
[0:51] <KenMan> hmm... are you in windows or linux or ... ?
[0:51] <kmag> linux
[0:51] <KenMan> okay, gentoo ? or do you use start-freenet.sh ?
[0:51] <kmag> I do have a 1.4.2 JVM installed
[0:51] <kmag> start-freenet.sh
[0:52] <KenMan> okay. Edit that file and down at the bottom, right before the line that begins with java, add a new one that reads 'which java'
[0:52] <KenMan> that will tell you the location of the JVM that is being used. It may be pulling the wrong one, perhaps ?
[0:53] <kmag> I checked my path
[0:53] <KenMan> if you wanted to be certain, you could hardcode the path to the java binary on the line which invokes fred.
[0:54] <kmag> Freenet says something about "Found a 1.4.2 JVM!
[0:54] <kmag> "
[0:54] <KenMan> your path may not be the same path which is used inside the start-freenet.sh script
[0:54] <kmag> wow... I'm an idiot
[0:54] <KenMan> aren't we all, secretly ?
[0:55] <KenMan> path order, or what ?
[0:56] <kmag> it's invoked with /bin/sh, right?
[0:56] <kmag> #!/bin/sh
[0:56] <KenMan> t'would seem that way to me ;)
[0:57] <kmag> I only added my 1.5 JVM to my bash startup scripts
[0:57] <kmag> err... PATH
[0:57] <kmag> ... in my bash startup scripts
[0:57] <KenMan> right. So now you got it going with 1.5 ('the tiger') now ?
[0:58] <kmag> they should really get rid of that misleading println that says "Found a 1.4.2 JVM... invoking it"
[0:58] <KenMan> well, what version is reported on the web interface (Information page) ?
[0:59] <kmag> I was thinking that Fred was trying to make itself a daemon by creating a subprocess and blocking SIG_HUP
[0:59] <kmag> I haven't actually shut down my node
[0:59] <KenMan> oh, you are referring to the startup script. I think that is where it reports 1.4.2 . Try grabbing an updated start script. I don't know if it got changed or not.
[1:00] <kmag> oh
[1:00] <KenMan> All i know is that i only installed 1.5 onto the machine I used, and it worked flawlessly.
[1:01] <KenMan> If it was pronouncing 1.4.2 to be 'da bomb' , I must have ignored that out of habit. Humans are creatures of habit, if nothing else.
[1:01] <kmag> alright... the web interface hasn't shown up yet... I keep my node very heavily loaded
[1:03] <kmag> I run Freenet on my old machine
[1:03] <kmag> a 266 MHz PII
[1:03] <kmag> it won't overclock to 300 MHz :-(
[1:03] <KenMan> glutton for punishment, huh ?
[1:03] <kmag> it fails to POST
[1:04] <KenMan> did you get the upgrade , to 256MB total RAM ?
[1:04] <kmag> 288 MB total RAM
[1:04] <kmag> 2 x 128 + 32
[1:04] <KenMan> oh, you went with the aftermarket hot-wad kit !
[1:05] <kmag> yea, it came wit 2 x 64
[1:05] <kmag> err... 2 x 32
[1:05] <KenMan> wait, i'm getting those confused with 8086 machines , back when there was only one maker. Never mind :p
[1:05] <KenMan> I was only off by a decade or so...
[1:06] <kmag> damn... that old workhorse is almost 7 years old
[1:07] <kmag> my laptop runs much hotter since I added an extra gig of RAM
[1:07] <kmag> I don't dare use it as a laptop any more
[1:07] <KenMan> as long as it don't burn yer nuts...
[1:11] <kmag> anyone out there running AMD64 machines?
[1:12] <kmag> or PPC64 Machines (in 64-bit mode, OS X still runs G5s in 32-bit mode)
[1:18] <kmag> KenMan. it always has been running on 1.5.0, but start-freenet.sh claims 1.4.2
[1:21] <KenMan> right. So fix it, or get used to not looking at it. Either way.
[1:23] <kmag> heh... what a useless echo
[1:24] <kmag> it always claims to use a 1.4.2 JVM
[1:29] * linagee (~linagee@netblock-66-245-229-151.dslextreme.com) Quit (Read error: 104 (Connection reset by peer))
[1:30] <kmag> anyway, it's late here
[1:30] * linagee (~linagee@netblock-66-245-229-151.dslextreme.com) has joined #freenet
[1:30] <kmag> thanks for your help
[1:30] * kmag (~kmag@pcp04402498pcs.nrockv01.md.comcast.net) Quit ("sleep time")
[2:59] * sanity (~ian@81-178-106-165.dsl.pipex.com) has joined #freenet
[3:26] <KenMan> hi sanity
[4:02] <KenMan> okay, i am running 1000 nodes, 12 routes, 12HTL, and 2000 item datastores. But it's gonna take a long time to reach a meaningful state...
[4:03] <KenMan> I'm logging all the datastores so I can watch the specialization for each node as it emerges.
[4:05] <KenMan> I'm capturing at least half of all the measurements I want to gather. Like connection lifetimes, peer success rates, etc...
[4:06] <KenMan> I believe it is possible to do what NGR strives for with a lot less effort.
[4:07] <KenMan> Rather than measuring how long each transfer takes, measure how many results a single peer returned over a fixed length time window.
[4:07] <KenMan> rank the peers this way, and favor getting new connections from the highest ranking peers.
[4:24] * Hory (~Miranda@82.76.81.56) has joined #FreeNet
[4:31] * Ash-Fox (HAL-9000@adm245.neoplus.adsl.tpnet.pl) Quit ()
[4:32] * hobx_ (~chatzilla@ankh.math.chalmers.se) Quit (Remote closed the connection)
[4:33] * d-ArkAngel (~Robert@213-131-104-86.onyx.net) has joined #freenet
[5:14] * sanity (~ian@81-178-106-165.dsl.pipex.com) Quit ()
[5:21] * sanity (~ian@81-178-106-165.dsl.pipex.com) has joined #freenet
[5:29] <TheSeeker> If my noad is overloaded, do I send a high-priority message to my peers telling them "hey, I'm overloaded" before or after they send me more requests?
[5:34] <TheSeeker> hmm, that sounds bad: Dec 13, 2004 12:23:00 AM (freenet.node.states.maintenance.Checkpoint, YThread-558, ERROR): unhandled throwable in Checkpoint: Polling and aggregation of diagnostics.: java.lang.NullPointerException
[5:55] <TheSeeker> I can't read the freenet log outside of a hex editor :( it's 538 MB after less than 24 hours...
[6:03] * nextgens (~nextgens@d213-103-219-226.cust.tele2.fr) has joined #freenet
[6:08] <TheSeeker> upgraded to 1.5.0-b64 from 1.5.0-b32c ... hope that help...
[6:08] <TheSeeker> helps
[6:22] <TheSeeker> already crashed :(
[6:23] <TheSeeker> I think I may need to reboot my PC ... it's been a while...
[6:37] * Hory (~Miranda@82.76.81.56) Quit (Read error: 110 (Connection timed out))
[6:39] * kevlAway (~kevloral@CZ1-RAS-8-u-0179.du.onolab.com) Quit ("Peace and Protection 4.22")
[6:48] * Hory (~Miranda@82.76.81.56) has joined #FreeNet
[7:14] * pupok (~r00t@81-178-106-165.dsl.pipex.com) has joined #freenet
[7:18] * nextgens is now known as nextgens|Confere
[7:21] * sdogi (~java@84-50-18-71-dsl.kjj.estpak.ee) Quit (Read error: 104 (Connection reset by peer))
[7:25] * sdogi (~java@84-50-19-119-dsl.kjj.estpak.ee) has joined #freenet
[7:27] * guido^pe (~unknown@dsl-213-023-244-084.arcor-ip.net) has joined #freenet
[7:31] * toad_ (toad@toad-with-underline.active.supporter.pdpc) has joined #freenet
[7:31] * ChanServ sets mode +o toad_
[7:32] <TheSeeker> Hi Toad, want a rediculously huge logfile generted in less than a day's uptime, terminating in: Dec 13, 2004 12:23:00 AM (freenet.node.states.maintenance.Checkpoint, YThread-558, ERROR): unhandled throwable in Checkpoint: Polling and aggregation of diagnostics.: java.lang.NullPointerException ?
[7:42] * bllx (~leot@host217-44-61-175.range217-44.btcentralplus.com) has joined #freenet
[7:42] <bllx> can anyone tell me where Freenet bugs are reported?
[7:44] <TheSeeker> http://mail.freenetproject.org/mailman/listinfo/support/
[7:45] <bllx> thanks, I know about that, but there was a page somewhere with live listings of freenet probs
[7:46] <TheSeeker> oh?
[7:46] <bllx> yes, it had an os X prob detailed, that I have too
[7:46] <bllx> maybe it was someone's private page, but I didn't think so
[7:52] <bllx> aah, freenethelp.org is what I was looking for
[7:59] <bllx> bye
[7:59] * bllx (~leot@host217-44-61-175.range217-44.btcentralplus.com) has left #freenet
[8:08] * tyesinclair (~tyesincla@170.252.64.1) has joined #freenet
[8:08] * tyesinclair (~tyesincla@170.252.64.1) has left #freenet
[8:19] * robilad (~topic@mpiat2305.ag2.mpi-sb.mpg.de) has joined #freenet
[8:25] * sdogi_ (~java@84-50-20-74-dsl.kjj.estpak.ee) has joined #freenet
[8:38] * sdogi (~java@84-50-19-119-dsl.kjj.estpak.ee) Quit (Read error: 110 (Connection timed out))
[8:48] * FallingBuzzard (~srademach@207.152.112.129) has joined #freenet
[8:49] <TheSeeker> toad_: do you know why Freenet tends to report and increasingly large number of keys in my datastore that don't exist on my hard drive?
[8:54] * toad_ (toad@toad-with-underline.active.supporter.pdpc) Quit (Remote closed the connection)
[8:58] <TheSeeker> guess not :P
[9:00] * hobx_ (~chatzilla@ankh.math.chalmers.se) has joined #freenet
[9:12] * nextgens|Confere (~nextgens@d213-103-219-226.cust.tele2.fr) Quit (Read error: 110 (Connection timed out))
[9:41] * greycat (~wooledg@192.35.79.70) has joined #freenet
[9:48] <i2p_iip> <Sonax> Hi greycat. Why can't i get your flog lately?
[9:48] <greycat> I haven't been running my node for a few days.
[9:48] <greycat> (and therefore haven't been inserting)
[9:48] <i2p_iip> <Sonax> that would explain that.
[10:18] * Freddy2 (~holahola@ibis-13.ala.um.es) has joined #freenet
[10:18] <Freddy2> hi
[10:19] * Freddy2 (~holahola@ibis-13.ala.um.es) Quit (Client Quit)
[10:23] * kevloral (~kevloral@CZ1-RAS-8-u-0179.du.onolab.com) has joined #freenet
[10:23] <kevloral> g'afternoon all
[11:26] * sdogi_ (~java@84-50-20-74-dsl.kjj.estpak.ee) Quit (Read error: 104 (Connection reset by peer))
[11:27] * sdogi (~java@84-50-22-161-dsl.kjj.estpak.ee) has joined #freenet
[11:41] * fx0 (metricuk@ACCB7D60.ipt.aol.com) has joined #freenet
[11:45] * fx0 (metricuk@ACCB7D60.ipt.aol.com) Quit ()
[11:56] * fwef (~kevloral@CZ1-RAS-8-u-0179.du.onolab.com) has joined #freenet
[11:56] * kevloral (~kevloral@CZ1-RAS-8-u-0179.du.onolab.com) Quit (Nick collision from services.)
[11:56] * fwef is now known as kevloral
[12:03] * d-ArkAngel (~Robert@213-131-104-86.onyx.net) Quit (Read error: 104 (Connection reset by peer))
[12:11] * Ash-Fox (HAL-9000@adm245.neoplus.adsl.tpnet.pl) has joined #FreeNET
[12:27] * nextgens (~nextgens@d80-170-148-141.cust.tele2.fr) has joined #freenet
[12:30] * nextgens (~nextgens@d80-170-148-141.cust.tele2.fr) has left #freenet
[12:30] * nextgens_ (~nextgens@d80-170-148-141.cust.tele2.fr) has joined #freenet
[12:33] * moskau23 (~Miranda@dsl-082-082-239-085.arcor-ip.net) has joined #freenet
[12:36] * nextgens_ is now known as nextgens
[12:43] <TheSeeker> hmm, I guess it wasn't frost that was directly causing the keys reported vs actual datastore files problem... I haven't run Frost at all this start and have 530ish more keys reported on the histogram pages than have keys in my datastore directory...
[12:48] * _6128_ (~boutinde@195.25.40.253) has joined #freenet
[12:48] <_6128_> hello
[12:52] * Hirvox (~hirvox@cs181124095.pp.htv.fi) has joined #freenet
[12:57] * tyesinclair (~tyesincla@170.252.64.1) has joined #freenet
[12:58] * tyesinclair (~tyesincla@170.252.64.1) has left #freenet
[13:05] * nextgens_ (~nextgens@d80-170-148-141.cust.tele2.fr) has joined #freenet
[13:12] <KenMan> TheSeeker: your counts are probably related to failed insertions. I understand there are a lot of those.
[13:13] * nextgens (~nextgens@d80-170-148-141.cust.tele2.fr) Quit (""ya mum quits IRC"")
[13:13] * nextgens_ is now known as nextgens
[13:14] * nextgens (~nextgens@d80-170-148-141.cust.tele2.fr) has left #freenet
[13:14] <KenMan> you've almost got me curious enough to run a node, in order to debug. But first, you should investigate logging options.
[13:14] * nextgens (~nextgens@d80-170-148-141.cust.tele2.fr) has joined #freenet
[13:15] <TheSeeker> why would freenet count keys as being in the datastore for a failed insert?
[13:15] * kevloral is now known as kevlAway
[13:16] <KenMan> a bug, perhaps. As in, let's count this new key being inserted, before it finishes. Oops, it failed.
[13:16] <TheSeeker> and... what logs do you mean? I have the 538 MB log for that one day's operation before crashing, but it was at the default logging level, so I don't know if that would be useful...
[13:18] <KenMan> you can set logLevelDetail to select logging for certain parts of the code, so you don't create an avalanche.
[13:35] <TheSeeker> as if I knew what parts of the code I'd need to log, or even how to do that :P
[13:41] * d-ArkAngel (~Rob@cpc2-midd2-5-0-cust16.midd.cable.ntl.com) has joined #freenet
[13:49] * pittaman (~mistery_b@d51A410CA.kabel.telenet.be) has joined #freenet
[14:02] * sanity (~ian@81-178-106-165.dsl.pipex.com) Quit ()
[14:14] * nextgens (~nextgens@d80-170-148-141.cust.tele2.fr) Quit ("bbl")
[14:15] * hobx_ (~chatzilla@ankh.math.chalmers.se) Quit (Remote closed the connection)
[14:54] * hobx_ (~chatzilla@ankh.math.chalmers.se) has joined #freenet
[15:02] * sdogi_ (~java@84-50-18-136-dsl.kjj.estpak.ee) has joined #freenet
[15:14] * sdogi (~java@84-50-22-161-dsl.kjj.estpak.ee) Quit (Read error: 110 (Connection timed out))
[15:33] <KenMan> okay, after letting this largish sim run for a long time, most nodes are basically specializing around a single point in keyspace. Some have more than one point, but generally they all have developed one dominant point.
[15:33] <KenMan> This might be attributed to the rate of new connections I am permitting. In other words,
[15:33] <d-ArkAngel> are they all different points? or are there duplicates?
[15:34] <KenMan> if I turned up the limit on that rate, most nodes might not converge one just a single point quite so heavily.
[15:34] <KenMan> They are all unique specializations.
[15:34] <KenMan> keyspace coverage is sufficiently uniform
[15:34] <d-ArkAngel> so there's essentialy no redundancy developing then...
[15:35] <KenMan> no, there most certainly is. The average number of instances is above ten
[15:35] <KenMan> 145899 unique keys, 2000000 storage slots
[15:35] <d-ArkAngel> through overlap between specilisations, rather than "backup" specilisations then.
[15:36] <KenMan> yeah, that's how it looks
[15:36] <KenMan> i think that through manipulating the rate of new connections I can control the overlap and focus - that's my intuition
[15:37] <d-ArkAngel> and idea how the specilisations are spreading out over the graph? (similar specilisation nodes "near" each otheretc )
[15:38] <KenMan> right now, the average number of queries for each dropped connection is 6799.3 , and successes is 913.779
[15:38] <KenMan> the overall success for the full network is only ~40% , but that's 1000x12x12x2000
[15:39] <d-ArkAngel> 913 out of 6799?
[15:39] <KenMan> i don't yet have the ability to examine nodes and their peers, to compare specializations. I'll get there.
[15:39] <d-ArkAngel> what shape network is that?
[15:39] <KenMan> 1000nodes, 12 routes each, 12 HTL max, and 2000 keys in each datastore
[15:40] <KenMan> 913/6799 = 13% pSuccess for the dropped peers. I'm using poor success as the criteria for selecting which peer to drop.
[15:41] <d-ArkAngel> wouldn't that be bad in the real network?
[15:41] <KenMan> what do you mean, in more detail ? i don't understand
[15:42] <d-ArkAngel> since your node would hate to learn about new nodes.... (since you don't know what to route them then their psuccess is bound to be lower than anyone elses while you learn about them...)
[15:42] <d-ArkAngel> also it would mean nodes have a predudice against talking to new nodes that have just joined the network.
[15:42] <KenMan> no, it's not bound to be bad. I 'protect' them for the first 1000 queries, or something like that.
[15:43] <KenMan> the worst performing peer that got dropped had 4.5% 123 out of 2726 queries
[15:44] <KenMan> well, peer management gets really complicated in a hurry, if you start examining it. I haven't given it a tremendous amount of attention yet, but it deserves it.
[15:44] <d-ArkAngel> indeed
[15:44] <d-ArkAngel> I can't say I've got any answers, just trying to help poseing questions
[15:45] <KenMan> I'm sure when I start manipulating it in detail, things will undergo a dramatic change... which is kind of scary.
[15:45] <KenMan> I love sharing questions and analysis!
[15:45] <d-ArkAngel> there needs to be a balance between good for the network, and good for the local node I suppose.
[15:45] <KenMan> this thing I'm running is still extremely simple. And I want to keep it that way if possible ;)
[15:46] <d-ArkAngel> because anarchy doesn't scale as a form of govenmnet, but then neither does comunism :-)
[15:46] <d-ArkAngel> Hence we need a captitalist mentality :-)
[15:47] <KenMan> well good for the group -vs- good for the individual... hmm. In this simple model, approximately 1/12 requests are locally originating. Coincidentally there are 12 peers.
[15:47] <KenMan> so each node gets a 'fair share' by coincidence.
[15:47] <d-ArkAngel> :-)
[15:47] <d-ArkAngel> gotta love coincidence
[15:47] <KenMan> well, okay. Maybe i set 12 and 12 on purpose, but you get my drift .
[15:48] <d-ArkAngel> indeed
[15:49] <KenMan> let me check on this: For 10K requests, there are about 108373 node query ops netwide.
[15:49] <KenMan> subtract 10K for the originating nodes, and that gives us an average hop-length (for succeeeds AND fails) of 9.8HTL , i think.
[15:50] <KenMan> so the requests are visit 10.8 nodes, on average. The max is 13 (orig + 12 hops).
[15:51] <KenMan> There are issues to address - my simple choice for an estimator, rate of connection changes, and something else...
[15:51] <KenMan> can't recall my morning (mental/unwritten) notes at the moment :(
[15:52] <d-ArkAngel> :-)
[15:52] <KenMan> say , did you get multithreaded sims going ? and have machines to make use of it ?
[15:52] <KenMan> i know toad sat upon some of your ideas, but did you keep going anyways ?
[15:52] <d-ArkAngel> I've only got dual opterons to run it on
[15:53] <d-ArkAngel> I did get some work done, but I'm in the process of moving house so free time is kinda scarse at the moment so I've not had a chance to work on it for a couple of weeks now...
[15:54] <KenMan> that's right, now i remember. So how is the move going ?
[15:54] <d-ArkAngel> well we got all boxed up and ready to move and the people we were buying from pulled out on us. and we ended up with nowhere to move too!
[15:54] <KenMan> oh christ. Now I remember that too! Damn!
[15:55] <d-ArkAngel> so we've had to buy something else... in theory we move on the 23rd! as long as these lot don't mess us about :-)
[15:55] <KenMan> is it comparable for your needs ? I mean, it sounds harsh to have to move so quickly.
[15:55] <d-ArkAngel> very different, but just as nice.
[15:56] <KenMan> good. I hope it works out... sounds like a messy situation (emotionally).
[15:57] <d-ArkAngel> well that's life isn't it :-)
[15:57] <KenMan> a well-adjusted attitude can bring you through a lot of troubles...
[15:57] <d-ArkAngel> as I keep saying "everything will work out one way or another in the end" :-)
[15:57] <KenMan> yup. that's my belief.
[15:58] <d-ArkAngel> as long as I can get broadband and a place to park my 'puter I'm happy :-)
[15:58] <KenMan> just the bare essentials, huh? :)
[15:58] <d-ArkAngel> indeed
[15:59] <d-ArkAngel> a freezer and a microwave are always a bonus :-)
[15:59] <KenMan> if there are things in them, anyway
[16:13] <KenMan> one of the things that concerns me involves node-turnover and network bootstrapping. My sim creates a static set of nodes, and they all magically acquire a nearly full set of routes from the git-go.
[16:14] <KenMan> But you have to start somewhere, and just go from there. So eventually I'll get it added in.
[16:18] <d-ArkAngel> Toads sims were quite interesting when he had them growing the performance was HORRIBLE
[16:18] <d-ArkAngel> be started it with a group of 10 fully connected nodes.
[16:20] <d-ArkAngel> you had to add nodes VERY slowly to prevent the performance degradation being vast.
[16:22] <KenMan> yes, the rate of change needs to be controlled. But even if you slam many nodes together all at once, given enough time to recover, it should be possible.
[16:23] <KenMan> http://mywebpages.comcast.net/jkcorson/FreeNet/N590DS.png - also, 591-599
[16:24] <KenMan> comments appreciated. Not much to say , i suppose. Except that they ARE specialized in some shape or form ...
[16:25] <KenMan> i don't know who is connected to whom, but these are effectively 10 random nodes from the network. Unlikely to be connected to one another.
[16:34] * greycat (~wooledg@192.35.79.70) Quit ("This time the bullet cold rocked ya / A yellow ribbon instead of a swastika")
[16:41] * Hirvox (~hirvox@cs181124095.pp.htv.fi) has left #freenet
[16:49] * Hirvox (~hirvox@cs181124095.pp.htv.fi) has joined #freenet
[16:49] <Hirvox> http://onionnetworks.com/products/swarmstream/
[16:54] * FallingBuzzard (~srademach@207.152.112.129) Quit (Read error: 54 (Connection reset by peer))
[17:00] * pittaman (~mistery_b@d51A410CA.kabel.telenet.be) Quit (Client Quit)
[17:14] * Hory (~Miranda@82.76.81.56) Quit ("CyberLore.net - Recommendations on the best games, freeware and websites.")
[17:18] * moskau23 (~Miranda@dsl-082-082-239-085.arcor-ip.net) Quit (Read error: 110 (Connection timed out))
[17:19] * mikeeusa (~r00t@ool-18be3a49.dyn.optonline.net) Quit (Read error: 110 (Connection timed out))
[17:21] * mikeeusa (~r00t@ool-18be3a49.dyn.optonline.net) has joined #freenet
[17:36] * sanity (~ian@81-178-106-165.dsl.pipex.com) has joined #freenet
[17:38] * sanity (~ian@81-178-106-165.dsl.pipex.com) Quit (Client Quit)
[17:41] * Hory (~Miranda@82.76.81.56) has joined #FreeNet
[17:46] * sanity (~ian@81-178-106-165.dsl.pipex.com) has joined #freenet
[18:07] * guido^pe (~unknown@dsl-213-023-244-084.arcor-ip.net) Quit ()
[18:07] * guido^pe (~unknown@dsl-213-023-244-084.arcor-ip.net) has joined #freenet
[18:29] * Hory (~Miranda@82.76.81.56) Quit ("CyberLore.net - Recommendations on the best games, freeware and websites.")
[18:33] * Sugadude (~Sugadude@82.143.204.119) Quit (Remote closed the connection)
[18:34] * Sugadude (debian-tor@wg55.waag.org) has joined #freenet
[18:43] * Ash-Fox (HAL-9000@adm245.neoplus.adsl.tpnet.pl) Quit (Nick collision from services.)
[18:43] * Ash-Fox (HAL-9000@adk16.neoplus.adsl.tpnet.pl) has joined #FreeNET
[19:00] * sanity (~ian@81-178-106-165.dsl.pipex.com) Quit ()
[19:22] * sanity (~ian@81-178-106-165.dsl.pipex.com) has joined #freenet
[19:22] * sanity (~ian@81-178-106-165.dsl.pipex.com) Quit (Client Quit)
[19:31] * sanity (~ian@81-178-106-165.dsl.pipex.com) has joined #freenet
[19:34] * guido^pe (~unknown@dsl-213-023-244-084.arcor-ip.net) Quit (Read error: 113 (No route to host))
[19:40] * d-ArkAngel (~Rob@cpc2-midd2-5-0-cust16.midd.cable.ntl.com) Quit (Read error: 104 (Connection reset by peer))
[19:41] * sanity (~ian@81-178-106-165.dsl.pipex.com) Quit ()
[19:43] <KenMan> It's time to confess: Who shot the toad ?
[19:50] <i2p_iip> <Sonax> KenMan: When was he last seen (working)?
[20:00] * sanity (~ian@81-178-106-165.dsl.pipex.com) has joined #freenet
[20:08] <sdogi_> how is it going with the freenet guys? when does it take over the internet ?:)
[20:18] * sanity (~ian@81-178-106-165.dsl.pipex.com) Quit ()
[20:21] <KenMan> toad was last seen >working< many days ago. I think he popped in once or twice over the past 4 days though. I hope he didn't hop onto a poison lily pad.
[20:22] <KenMan> He had the flu sometime last week, maybe he is still recovering from that :(
[20:23] * sanity (~ian@81-178-106-165.dsl.pipex.com) has joined #freenet
[20:25] * sanity (~ian@81-178-106-165.dsl.pipex.com) Quit (Client Quit)
[20:35] <KenMan> sdogi_: freenet's opportunity to dominate the internet has long since passed. now it's just a question of whether we can claim any real estate at all :o
[20:45] * Sugadude (debian-tor@wg55.waag.org) Quit (Remote closed the connection)
[20:45] <KenMan> http://mywebpages.comcast.net/jkcorson/FreeNet/ukhist1000x12x12x2000.png
[20:46] <KenMan> this is a histogram of the global set of keys stored in the sim network, for anyone who is interested.
[20:46] <KenMan> It does not count redundant key copies, just a single instance of each key retained by the network.
[20:46] * Sugadude (~Sugadude@static-64-115-210-23.isp.broadviewnet.net) has joined #freenet
[20:49] <KenMan> although it has narrow peaks and valleys, i would call this coverage relatively uniform. Comparing the lowest to highest is only about 2:1...
[21:10] * nextime (~nextime@danex.i-m-c.it) Quit ("nextime has no reason")
[21:17] * Elly (~Elly@ool-182c3b26.dyn.optonline.net) has joined #freenet
[21:18] * hobx_ (~chatzilla@ankh.math.chalmers.se) Quit (Remote closed the connection)
[21:19] <sdogi_> let me get this straigth, socks proxy server is like nat right? i set up some socks server, connect to it with some program and it binds ports on the server and forwards stuff to me
[21:19] <sdogi_> ?
[21:59] * Elly is now known as Elly|afk
[22:08] * nextime (~nextime@danex.i-m-c.it) has joined #freenet
[22:15] * xena (~xena@espians.com) Quit (Client Quit)
[22:18] * Ribs (~freenode@riblet.plus.com) Quit (Read error: 60 (Operation timed out))
[22:50] * Elly|afk (~Elly@ool-182c3b26.dyn.optonline.net) Quit (Read error: 110 (Connection timed out))
[23:25] * i2p_iip (~changate@host.teitel.net) Quit ("Leaving")
These logs were automatically created by Jay Oliveri with his gimp hapi on irc.freenode.net.
