Timestamps are in GMT/BST.
[0:01] * Terminator2 is now known as Terminator3
[0:22] * brainZzZ (gibe@c-24-18-82-42.client.comcast.net) Quit (Client Quit)
[0:24] * kokje (~kokje@c-24-127-127-236.hsd1.ca.comcast.net) has joined #freenet
[1:48] * bryan_W (~is@69.151.14.170) Quit ("Leaving")
[1:53] * nick01 (~nick01@home-07031.b.astral.ro) has left #freenet
[2:22] * sleon|tuX (test@p54A175F3.dip.t-dialin.net) has joined #freenet
[3:02] * sleon|tuX (test@p54A175F3.dip.t-dialin.net) Quit (Read error: 131 (Connection reset by peer))
[3:13] * nextgens (~nextgens@sd440.sivit.org) has joined #freenet
[3:16] <nextgens> hi
[3:23] <kokje> Hello
[3:23] <kokje> is there any model for the processing time in freenet
[3:52] <kokje> anybody there
[4:00] <nextgens> kokje: what is "processing time" ?
[4:07] * e1ven (~e1ven@c-65-96-126-221.hsd1.ma.comcast.net) has joined #freenet
[4:29] * nextgens (~nextgens@sd440.sivit.org) Quit (Remote closed the connection)
[4:48] * sleon|tuX (test@p54A175F3.dip.t-dialin.net) has joined #freenet
[4:49] * sleon|tuX is now known as sleon|away
[4:58] <kokje> processing time is the time taken by freenet to process requests/inserts/queries
[5:31] * pittaman (~mistery_b@d54C32DBA.access.telenet.be) has joined #freenet
[5:49] * sleon|away is now known as sleon
[5:50] * sleon is now known as sleon|tuX
[5:53] * sleon_ (test@p54A175F3.dip.t-dialin.net) has joined #freenet
[6:07] * e1ven (~e1ven@c-65-96-126-221.hsd1.ma.comcast.net) Quit ("Leaving")
[6:08] * sleon|tuX (test@p54A175F3.dip.t-dialin.net) has left #freenet
[6:10] * NullAcht15 (~NullAcht1@dsl-213-023-244-056.arcor-ip.net) has joined #freenet
[6:12] * sleon_ (test@p54A175F3.dip.t-dialin.net) Quit ("Leaving")
[6:16] * oierw (~mathew@cpe-66-91-35-90.hawaii.res.rr.com) Quit (Read error: 145 (Connection timed out))
[6:18] * oierw (~mathew@cpe-66-91-35-90.hawaii.res.rr.com) has joined #freenet
[7:37] * NullAcht15 (~NullAcht1@dsl-213-023-244-056.arcor-ip.net) Quit (Remote closed the connection)
[7:39] <i2p_iip> <Sonax> Has anyone answered the guy who asked about the copyright situation of the logo etc. on suppoort? Should someon? (toad, sanity, i'm looking in you'r direction...)
[7:41] <toad_> ian was going to
[7:42] <i2p_iip> <Sonax> ... but saving the world through dancing got in the way, or? Anyway, how are you doing Toad? Is .7 progressing in any way?
[7:46] <sanity_> sonax: i did
[7:47] * d3wd3 (imnotreal@82.208.142.215) has joined #freenet
[7:48] <d3wd3> anybody here for assistance ?
[7:48] <i2p_iip> <Sonax> sanity: You answered him personally outside the list, or do i need to check my mail program?
[7:49] <i2p_iip> <Sonax> Or, put in another way, i found the question interesting and would love to hear you'r answer in general.
[7:50] <d3wd3> i don't get it.. how do i search this freenet ? can it work as a filesharing app ?
[7:50] <d3wd3> and what am i sharing to the network ?
[7:51] <d3wd3> oh, i just reaad the faq, it says it's not searchable
[7:52] <toad_> d3wd3: it's not searchable in general
[7:52] <d3wd3> then how do i get resources with it ? totally confuzed
[7:52] <sanity_> sonax: personally
[7:52] <d3wd3> sorry for my noobish attitude, i've used a lot of p2p apps but can't understand this one
[7:52] <toad_> d3wd3: you either a) browse it, using fproxy, or b) use frost for message boards and some files
[7:52] <toad_> d3wd3: through fproxy, it's essentially an anonymous WWW
[7:53] <toad_> d3wd3: through frost, it's a bunch of message boards which can have files attached
[7:53] <toad_> either way you can have files
[7:53] <sanity_> sonax: basically i said if its for profit we would ask for a % of profits, and he should avoid implying that we are endorsing any message he sends (like a t-shirt with "I love pirating music" or something)
[7:53] <toad_> frost is pseudo-searchable
[7:53] <d3wd3> so i have these bookmarks in 127.0.0.1 webpage
[7:53] <toad_> whereas fproxy has index pages
[7:53] <d3wd3> but they don't seem to load..
[7:53] <toad_> d3wd3: they will. freenet can take a while to get going
[7:54] <toad_> d3wd3: as in, it can take up to several days, even a week or two sometimes, to run as well as a well-established node
[7:54] <i2p_iip> <Sonax> d3wd3: And c) use FUQID to up/down -load files you find on frost.
[7:54] <i2p_iip> <Sonax> sanity: ok - thanks. (Good slogan, by the way :)
[7:54] <toad_> yeah, but you can get big files from any of the above means
[7:54] <d3wd3> so what is this frost ?
[7:54] <toad_> there are big files on some fproxy sites
[7:54] <toad_> http://jtcfrost.sf.net/
[7:55] <toad_> d3wd3: are you firewalled?
[7:55] <d3wd3> no, i'm active
[7:55] <toad_> d3wd3: you must be able to receive connections on the listenPort
[7:55] <d3wd3> i am
[7:55] <d3wd3> somebody should make a little flash animation for noobs explaining the concept and how to operate it
[7:55] <toad_> do you have any incoming connections yet?
[7:55] <toad_> web interface -> advanced mode -> open connections
[7:56] <toad_> d3wd3: it really isn't that hard. it's slow to bootstrap, that's the main problem.
[7:56] <d3wd3> there are 9
[7:56] <toad_> d3wd3: how many incoming?
[7:56] <toad_> and ian refuses to admit reality on the web page :)
[7:56] <d3wd3> the arrow is pointing from left to right
[7:56] <toad_> for fear that it might put people off! :|
[7:57] <d3wd3> ah, that arrow means idle
[7:57] <toad_> |When you first start Freenet your node will know very little about the network, meaning that it could take several minutes or longer to retrieve a web page. Please be patient since Freenet will learn how to find information more effectively over time and its speed will gradually improve with use.
[7:57] <toad_> its reliability too...
[7:57] <d3wd3> hmm, i have a constant 2.5kb/down 2.5kb/s up transfer
[7:58] <d3wd3> that's me working as a proxy ?
[8:00] <d3wd3> hmm
[8:00] <d3wd3> so this frost is like a frontend for the freenet network ?
[8:07] <d3wd3> damn, i'm confused again
[8:08] * FallingBuzzard (~srademach@207.152.112.129) has joined #freenet
[8:09] <d3wd3> SSK@Sc6qV~D6iFhaYord6HtbjJ8MaEYPAgM/YoYo//activelink.png
[8:09] <d3wd3> ?
[8:09] <d3wd3> that's a file that is transferred through my connection ?
[8:11] <d3wd3> ok so i clicked on a site in the bookmarks and it says:
[8:11] <d3wd3>
[8:11] <d3wd3> Couldn't retrieve key: SSK@9G4s%7ejLQJB7ALQg-v2q5xKAJy9YPAgM/CofE//
[8:11] <d3wd3> Hops To Live: 17
[8:11] <d3wd3> Error: Route Not Found
[8:11] <d3wd3> The request couldn't even make it off of your node. Try again, perhaps with the GPL to help your node learn about others. The publicly available seed nodes have been very busy lately. If possible try to get a friend to give you a reference to their node instead
[8:12] <d3wd3> are there some non freenet boards with freenet files uri ?
[8:14] <d3wd3> maybe this program is not for me ? :)
[8:17] <toad_> d3wd3: hmm?
[8:17] <toad_> d3wd3: retry
[8:17] <toad_> d3wd3: like i said, it'll suck to start with
[8:17] <toad_> it has to learn where stuff is, it has to get connected to lots of nodes
[8:18] <d3wd3> and what about frost ? it seems to load messages on boards but i can't see the
[8:18] <d3wd3> m
[8:18] <d3wd3> oh found them
[8:18] <toad_> d3wd3: let it run for a couple of days; as long as you have some connections it should be usable then
[8:18] <toad_> if not sooner
[8:18] <d3wd3> DAYS ? lol
[8:19] <toad_> as long as you have INCOMING connections, that is
[8:19] <toad_> d3wd3: like i said, it takes a while to learn
[8:19] <toad_> we are working on that, of course
[8:19] <d3wd3> you're a dev ?
[8:19] <toad_> yeah..
[8:20] <i2p_iip> <Sonax> d3wde: Alos, YoYo is currently not maintained. So no luck there, but TFE, CofE and FIND should all load.
[8:21] <toad_> FIND links to 5 other indexes
[8:22] <d3wd3> excuse me for not being so enthusiastic, but who would want to wait days for an website to open ? lol
[8:23] <d3wd3> so it seems to have dozens of connections
[8:23] <toad_> d3wd3: it won't take 48 hours for a website to opne
[8:24] <toad_> it will take 48 hours for your node to learn enough to open the website quickly
[8:24] <d3wd3> no ?
[8:24] <toad_> and the other websites too hopefully
[8:24] <d3wd3> well i'm on a 3.5 gb monthly cap and leaving it on like that would deplete that in a few days
[8:24] <toad_> well then how do you use ANY p2p?
[8:24] <i2p_iip> <Sonax> d3wd3: When The StormTroopers and BigBrother censor everything else, you might find it worth the wait (also, once you'r node get's integrated it should pick up speed, and finally, the devels (toad) are working to improve it overall.
[8:25] <d3wd3> well i only download small things, and use local proxies, etc
[8:25] <toad_> well, i'm afraid we can't help australians very much
[8:25] <toad_> and new zealanders
[8:25] <toad_> and other people with absurd bandwidth caps :)
[8:26] <d3wd3> omg, 30k download and 15k up
[8:26] <d3wd3> what happened ?
[8:26] <toad_> we'd like to, but if you can't run the node a lot of the time you'll always have difficulty finding stuff
[8:26] <toad_> d3wd3: cool! try now on the sites
[8:26] <toad_> d3wd3: you can limit the bandwidth usage
[8:26] * greycat (rfc1413@wooledge.org) has joined #freenet
[8:26] <d3wd3> well 30k is the modem cap
[8:27] <d3wd3> it won't go faster than that ever
[8:27] <d3wd3> omg the site opened
[8:27] <toad_> you can limit the upload with the outputBandwidth setting
[8:27] <toad_> the download will be roughly the same as the upload if you're not actually using the node
[8:27] <d3wd3> so this is used for posting stuff without the fear of censorship..
[8:28] <d3wd3> or the mafia coming for you
[8:28] <toad_> d3wd3: or free web hosting... lots of possible uses
[8:28] <greycat> it also lets you read anonymously
[8:28] <d3wd3> free web hosting lol, like anyone would visit your site if it took that much time to load
[8:30] <toad_> d3wd3: it gets faster
[8:30] <toad_> d3wd3: that's what i've been trying to tell you
[8:31] <toad_> the node as a whole speeds up the longer it runs
[8:31] <toad_> it takes a few days to learn its way around after which it is tolerable
[8:31] <toad_> we are working on making it fast
[8:32] <greycat> (note: they've been working on making it fast for about 5 years now)
[8:32] <toad_> and hosting is a big deal for a lot of people - if you have a popular site with biggish files, you'll probably have to abandon it if you don't have significant revenue from it
[8:32] <toad_> greycat: :)
[8:32] <d3wd3> so let me understand, the data is forwarded through many ip's ?
[8:34] <toad_> yes
[8:34] <toad_> it is also cached on many IPs
[8:35] <d3wd3> hmm
[8:35] <d3wd3> but that increases internet traffic
[8:35] <toad_> the network decides where to cache stuff, partly based on how often it is requested, and partly based on the "key" of the data
[8:35] <d3wd3> it doubles, triples or makes the overall traffic a few times in size
[8:35] <toad_> d3wd3: yes, but it is cached many places
[8:36] <d3wd3> well this for ppl with good internet connections, not me
[8:36] <toad_> quite possibly :|
[8:36] <toad_> what's 3.5GB/mo in terms of bytes/sec?
[8:36] <d3wd3> nooo
[8:37] <d3wd3> you can only download+upload a total ammount of 3.5 gbytes a month
[8:37] <toad_> 1400 bytes per second :<
[8:37] <d3wd3> the speed is 30 kbytes/s
[8:37] <toad_> not good!
[8:37] <d3wd3> what goes over that you have to pay
[8:37] <d3wd3> 1 cent/mbyte
[8:37] <d3wd3> this subscription is 19$/month
[8:37] <toad_> d3wd3: sure but any really distributed p2p system will want you to run continuously and upload stuff
[8:38] <greycat> $19 a month is a third of what I pay.
[8:38] <d3wd3> the unlimited one is 29$/month but it's on a 5 mbit line with like 100 users on it :)
[8:38] <d3wd3> the pings are horrible, and the speeds too
[8:38] <toad_> freenet's architecture means you have to be able to run it more or less continuously... leaches don't get good performance
[8:38] <toad_> just like leaches don't get good performance on bittorrent - except that freenet doesn't just download ONE file
[8:38] <toad_> it caches stuff for many files
[8:39] <toad_> and your internet connection is only sufficient for you to leach unfortunately
[8:40] <toad_> we have had modem users occasionally
[8:40] <toad_> but even they could use more than 3.5GB/month
[8:40] <greycat> psssst.... it's spelled "leech"...
[8:40] <toad_> 1400bytes/second is 1/6th of what you can get symmetrically over a modem
[8:41] <greycat> no, more like 1/2.
[8:41] <toad_> greycat: 33600*2 = 67200
[8:41] <toad_> 67200/8 = 8400
[8:41] <greycat> a standard consumer grade modem can't go over 33kbps, and 1400 bytes/sec over a modem is 14kbps
[8:42] <toad_> you can get 8.4kB/sec over a modem, IF you use both upload and download at once
[8:42] <greycat> oh, you're adding both together.
[8:42] <toad_> yes, which is what freenet would do
[8:42] <toad_> but you can get 4.2kB/sec uncompressed download over a modem
[8:42] <greycat> is his bandwidth limit on *total* upload+download, or is it just based on downloads, or just uploads?
[8:42] <toad_> he said total :(
[8:43] <d3wd3> no not 1400 bytes/s
[8:43] <d3wd3> 30 kbytes/s
[8:43] <greycat> toad computed that 3.5 GB/month is about 1400 bytes/sec
[8:44] <toad_> d3wd3: freenet doesn't use continually 100%, but you can approximate it to that
[8:44] <toad_> d3wd3: because it fetches stuff for other nodes as well as your local leeching requests
[8:44] <toad_> d3wd3: therefore, you have to divide the 3.5GB/month into the number of seconds in a month, and you end up with 1400 bytes/sec
[8:45] <greycat> my calculation matches that, to 2 significant digits
[8:45] <d3wd3> ah :)
[8:45] <toad_> even if it turns out to only use 25%, it'd be a peak of 4*1400=5.6kB/sec
[8:45] <d3wd3> well surely i won't use it
[8:45] <toad_> combined for upload and download
[8:45] <greycat> freenet just isn't designed for metered connections
[8:45] <toad_> come back when you have a real internet connection :)
[8:46] <toad_> and I apologize for freenet's initial slowness and unreliability, it is a problem of the current architecture, and it DOES get better
[8:51] <d3wd3> problem is i can't get a better connection
[8:51] <d3wd3> it's 110$ for broadband
[8:51] <d3wd3> 512kbps
[8:54] * mazzanet (~mazzanet@mazzanet.user) Quit (Read error: 54 (Connection reset by peer))
[8:54] <d3wd3> i know that sucks
[8:54] <d3wd3> it's hell for a filesharing addict
[8:56] <greycat> where is that?
[8:59] <d3wd3> romania
[9:00] <toad_> :(
[9:00] <toad_> so that's what you have now?
[9:00] <toad_> 30kB/sec is broadband, it's just metered broadband
[9:01] <toad_> there are people who run freenet occasionally rather than continually... it's just not officially endorsed, and will produce weak performance
[9:01] <toad_> but it is possible
[9:03] <toad_> ... as you have seen
[9:05] <toad_> freenet probably isn't the best thing for what would normally be considered filesharing at the moment though
[9:05] <toad_> have a look around, it will improve while it is on, and it won't have to start from scratch if you stop it and then start it again
[9:18] * verl (verl@h155n2fls33o877.telia.com) has joined #freenet
[9:19] <toad_> hi verl
[9:20] * goatee (~goatee@ip216-239-88-210.vif.net) Quit (Read error: 110 (Connection timed out))
[9:24] * toad_ tries a growing-whacky-sim...
[9:37] * goatee (~goatee@ip216-239-88-249.vif.net) has joined #freenet
[9:39] <i2p_iip> <Sonax> Toad: Are you wasting you time playing The Sims, when you should be coding?!? :)
[9:46] * goatee_ (~goatee@ip216-239-88-86.vif.net) has joined #freenet
[9:51] <toad_> Sonax: no
[9:57] * goatee (~goatee@ip216-239-88-249.vif.net) Quit (Read error: 60 (Operation timed out))
[10:13] * Hory (~Miranda@82.76.81.56) has joined #FreeNet
[10:18] * sleon|tuX (test@p54A175F3.dip.t-dialin.net) has joined #freenet
[10:20] * ced_ (~chatzilla@d213-103-71-240.cust.tele2.fr) has joined #freenet
[10:22] <ced_> whats the status of freenet now?
[10:22] <greycat> stable build 5102 exists and runs. 0.7 is in progress.
[10:23] <ced_> 5102 runs or crawl?
[10:24] <ced_> some versions used to be incredibely slow
[10:28] <toad_> 5102 is similar performance-wise to 5100
[10:33] <ced_> does the windows installer use the bz2 seednodes?
[10:35] <sanity_> toad: how is the growing sim stuff doing?
[10:43] * ced_ (~chatzilla@d213-103-71-240.cust.tele2.fr) Quit (Remote closed the connection)
[10:45] <toad_> sanity_: well, the subset networks are not necessarily fully connected
[10:46] <toad_> sanity_: this messes inserts and probably requests up
[10:46] <toad_> it is caused by me not doing bidirectional connections
[10:47] <sanity_> ah yes, just got your email
[10:51] <toad_> Average hops for greedy routing: 9.07311
[10:51] <toad_> seems to have risen slightly for 4000...
[10:52] <toad_> Cycle 0: Average path length: 1.003798269677147, Success probability: 0.23695, inserts: 0.6205
[10:52] <toad_> hmmm
[10:52] <toad_> why would any insert fail if the graph is fully connected, and has 50 nodes?
[10:58] <toad_> sanity_: i can just do what we do on the real network, and decrement the HTL when we RNF - but then the insert hasn't gone to the full 10 nodes
[10:58] <toad_> it biases the results
[11:00] <toad_> ah
[11:00] <toad_> if i use an initial network of 100, then all inserts succeed (with bidi)
[11:00] <toad_> it's just that some nodes are so well connected that they can't send a request 10 hops on a network of 50 nodes!
[11:00] <toad_> okay, now we let it bootstrap until it reaches success probability of say 90%
[11:01] <toad_> then we start adding nodes
[11:01] <toad_> that is, if it will actually bootstrap for me...
[11:06] <toad_> bbiab
[11:08] <toad_> this is using the 3-estimator NGR
[11:09] <sanity_> the 3-estimator NGR didn't work before, did it?
[11:28] * goatee__ (~goatee@ip216-239-81-79.vif.net) has joined #freenet
[11:36] * goatee_ (~goatee@ip216-239-88-86.vif.net) Quit (Read error: 60 (Operation timed out))
[11:49] * d3wd3 (imnotreal@82.208.142.215) Quit ("Ever stop to think, and forget to start again?")
[12:08] <toad_> sanity_: indeed.. and it didn't work this time either :(
[12:10] <sanity_> what abuot just the single estimator?
[12:10] <toad_> doesn't look good either
[12:10] <toad_> but maybe
[12:12] <toad_> does not seem to be learning fast if at all...
[12:13] <toad_> considering there are only 100 nodes this is somewhat disturbing...
[12:14] <toad_> and why is the average path length 0.9?
[12:16] <toad_> there may be something about the whole network that is not true of the subset network...
[12:17] <toad_> it does seem to be slowly learning...
[12:17] <toad_> sanity_: the whole network is fairly strongly connected; each node N is connected to N-1 and N+1 in addition to having many crosslinks
[12:18] <toad_> this is not generally true of the subset
[12:19] <toad_> it may be that the way we constructed the whole network generates some unexpected properties that make routing really easy
[12:19] <toad_> which are not preserved in the subset?
[12:20] <toad_> with 3 estimators, the psuccess hovered around 0.17
[12:20] <toad_> with one estimator, it hovers around 0.12
[12:20] <toad_> it might be slowly climbing, but if so, it's very slow
[12:21] <toad_> sanity_: ideas?
[12:22] <sanity_> hmmm
[12:22] <toad_> :|
[12:23] <sanity_> and you are growing the network using the method we discussed? (ie. by following links)
[12:24] <toad_> i start off by activating one random node
[12:24] <toad_> i keep a list of border nodes
[12:24] <toad_> when i activate a node, i add all its inactive connected nodes to the border node list
[12:24] <toad_> then when i want to expand, i pick a random border node and activate this
[12:24] <sanity_> to the top or bottom of the list?
[12:24] <sanity_> oh, a random one... ok
[12:25] <toad_> i do this 99 times after the first random node, so the inserts work tolerably
[12:25] <toad_> then i run a load of requests
[12:25] <toad_> i'm still in that phase
[12:25] <sanity_> yeah, i guess maybe the smaller networks don't have the right property...
[12:25] <toad_> running requests on 100 node subset
[12:26] <toad_> i doubt that a 500 or 1000 node subset would have the right property even
[12:26] <sanity_> how many connections does each node have on-average in the full size network?
[12:26] <toad_> but i haven't tried it yet...
[12:26] <toad_> each node has 10 connections
[12:26] <toad_> each connection is bidirectional
[12:26] <toad_> there are 4000 nodes in the full size network
[12:27] <toad_> in the subset, it's probably a LOT less...
[12:27] <sanity_> so on average, a subset of 400 nodes would only have one connection each?
[12:28] <sanity_> a subset of 100 would probably only have 1 connection each (since we know they must have at least one to be activated)
[12:28] <sanity_> we need to keep the number of per-node connections the same as the network grows
[12:29] <sanity_> try putting the inactive nodes at the top of a stack, and pulling the ones to activate off the bottom
[12:29] <sanity_> that would effectively do a depth-first traversal, and would try to activate all a node's links before moving on to the next one
[12:30] <sanity_> s/depth/breadth
[12:30] <toad_> hmmm
[12:30] <toad_> so
[12:30] <toad_> make a queue
[12:30] <toad_> activate one node
[12:31] <toad_> push its links on the top of the stack
[12:31] <toad_> then to activate another node, pull it from the bottom?
[12:31] <toad_> and then push that one's links on the top... ahhh
[12:31] <toad_> i see
[12:35] <toad_> Average conns per node: 2.44
[12:35] <toad_> that's with the current code
[12:36] <toad_> that ought to be enough :|
[12:43] <sanity_> that depends on where those connections go
[12:43] <sanity_> try the breadth first thing
[12:43] <toad_> do you think that the breadth first thing is plausible
[12:43] <toad_> ?
[12:43] <toad_> or is it a matter of we don't care whether it's plausible at this stage?
[12:44] <toad_> Average conns per node: 2.26
[12:45] <sanity_> the simulation isn't plausible right now, people don't make friends by calculating a 1/d distance :-)
[12:45] <toad_> more or less same number of conns per node...
[12:45] <sanity_> really?
[12:45] <sanity_> per active node?
[12:45] <toad_> wait...
[12:46] <toad_> bug :)
[12:46] <toad_> i'm not completely convinced about my bidi setup algo...
[12:46] <toad_> many times it gets stuck on the last node...
[12:48] <toad_> Average conns per node: 2.66
[12:48] <toad_> argh
[12:48] <sanity_> how many nodes are active with 2.66 connections?
[12:49] <sanity_> that would imply about 1/4 of the nodes are active, assuming 10 connections per node in the full network?
[12:49] <toad_> 100 nodes active
[12:49] <sanity_> out of 4000?
[12:49] <toad_> and no it wouldn't
[12:49] <sanity_> why not?#
[12:50] <toad_> that's 100 nodes, each one has an average of 2.66 connections to other nodes within the 100
[12:50] <toad_> it does seem to have done something different though
[12:50] <toad_> there's a bunch of nodes with 10 conns first
[12:51] <toad_> and some with smaller numbers
[12:51] <toad_> and a lot of 1's
[12:51] <sanity_> that is to be expected
[12:51] <toad_> not the same pattern as with the random growth algo
[12:51] <toad_> which had no nodes with 10
[12:52] <sanity_> i know, that is what you would expect with a breadth-first growth pattern
[12:52] <toad_> success probability is slightly higher than before
[12:52] <sanity_> i suppose the vast majority of nodes only had 1 connection?
[12:52] <toad_> and MAY be rising
[12:52] <toad_> sanity_: no, the average was about the same
[12:53] <toad_> with breadth first the majority of nodes have 1 connection
[12:53] <sanity_> average what?
[12:53] <toad_> average number of active conns
[12:53] <sanity_> that is what i mean
[12:54] <sanity_> you would imagine that the data would get cached on one of the well connected nodes initially, and would therefore be easy enough to find...
[12:54] * d3wd3 (imnotreal@82.208.142.215) has joined #freenet
[12:54] <d3wd3> hey, isn't there something i can do to make freenet faster ? like get an up to date node file or something
[12:55] * goatee_ (~goatee@ip216-239-81-26.vif.net) has joined #freenet
[12:55] <sanity_> d3wd3: just leave your node running
[12:55] <toad_> wb d3wd3 ...
[12:55] <toad_> sanity_: he has a metered connection :<
[12:55] <d3wd3> hehe, i decided to use it after all :)
[12:55] <sanity_> oh bugger
[12:56] <d3wd3> Error: Route Not Found
[12:56] <d3wd3> Attempts were made to contact 2 nodes.
[12:56] <d3wd3> 2 cleanly rejected.
[12:56] <sanity_> toad: and you are sure you are only inserting and requesting from active nodes?
[12:56] <d3wd3> that means the site is down ?
[12:56] <toad_> d3wd3: how many connections do you have?
[12:56] <toad_> d3wd3: no it means your node can't talk to enough of the network to find it
[12:56] <d3wd3> hmm, i have about 5 free sites pending to open, and the config page is really laggy !
[12:57] <toad_> sanity_: reasonably sure...
[12:57] <d3wd3> why is it lagging to talk to localhost ??
[12:57] <greycat> missing a definition of localhost in /etc/hosts?
[12:57] <toad_> d3wd3: go to Open Connections; how many conns do you have open?
[12:57] <greycat> IPv6 vs. IPv4?
[12:58] <d3wd3> toad_ i'm waiting for that to open !
[12:58] <toad_> d3wd3: you have too many windows open?
[12:58] <greycat> web browser configuration, refusing to open more than {2,4} connections to 127.0.0.1 at once?
[12:58] <toad_> sanity_: does not look particularly promising...
[12:58] <d3wd3> hmm, why do i need to close all windows so i can get access to the main page ?
[12:58] <sanity_> toad: hmmm, its strange
[12:58] <d3wd3> hmm, how can i change that in firefox ?
[12:58] <toad_> d3wd3: because your browser (and fred) limit the number of connections they use at once
[12:59] * sleon|tuX (test@p54A175F3.dip.t-dialin.net) Quit (Read error: 60 (Operation timed out))
[12:59] <greycat> d3wd3: web browser config. tell it to open more connections to each server.
[12:59] <toad_> d3wd3: you can change that
[12:59] <sanity_> toad: can you just let it grow to full size?
[12:59] <greycat> if it's a mozilla/firefox browser, try about:config
[12:59] <toad_> hmm?
[12:59] <toad_> sanity_: hmm?
[12:59] <d3wd3> Connections open (Inbound/Outbound/Limit) 37 (25/12/200)
[12:59] <sanity_> toad: the network
[12:59] <toad_> sanity_: meaning?
[13:00] <toad_> how big would you like full size to be? how long should it be small for?
[13:00] * goatee__ (~goatee@ip216-239-81-79.vif.net) Quit (Read error: 60 (Operation timed out))
[13:00] <d3wd3> toad_ how can i increase the max number of connections to 127.0.. in firefox ?
[13:00] <sanity_> toad: it should grow steadily until all 4000 nodes are active
[13:01] <d3wd3> so 25/12 is good toad_ ?
[13:01] <toad_> d3wd3: 37 total is okay
[13:01] <toad_> sanity_: that's possible but why would it help?
[13:01] <toad_> and it would take a long time...
[13:01] <d3wd3> 1 question..
[13:02] <d3wd3> i guess the program cannot use an http proxy, right ?
[13:02] <d3wd3> because of it must be connectable
[13:02] <sanity_> toad: i need to think about it a bit
[13:03] <toad_> d3wd3: hmm?
[13:03] <d3wd3> can i make the transfers through an http proxy ?
[13:03] <toad_> no
[13:03] <toad_> why would you need to?
[13:03] <d3wd3> i have a proxy that is not metered by the isp
[13:03] <d3wd3> it's local
[13:03] <d3wd3> and very fast
[13:03] <toad_> ah
[13:03] <toad_> :|
[13:04] <toad_> well no, not at present
[13:04] <d3wd3> so how can i change the max connections to 127.0.0.1 in firefox ?
[13:04] <toad_> sanity_: i tried to run active-nodes-only greedy routing on the active network. i got a stack overflow.
[13:04] <toad_> in other words, it's not 1/d.
[13:04] <toad_> d3wd3: about:config
[13:05] <toad_> d3wd3: look for network.http
[13:05] <toad_> stuff about max connections
[13:05] <d3wd3> ok, changed from 54 to 1000 :)
[13:05] <sanity_> toad: ok, so the way we are growing the network is wrong
[13:05] <d3wd3> better make it maxint just to be sure :)
[13:05] <toad_> sanity_: interestingly enough, it does this even if i do random growth
[13:06] <toad_> d3wd3: there are a few of those settings
[13:06] <toad_> sanity_: yeah...
[13:06] <sanity_> toad: well, people form small world networks which grow gradually - how do they do it?
[13:07] <d3wd3> ok, changed them all to 30000 :)
[13:07] <toad_> sanity_: actually, it may be that the network isn't all that wrong...
[13:07] * sleon|tuX (test@p54A15705.dip.t-dialin.net) has joined #freenet
[13:07] <toad_> what happens is that we get infinite loops while doing greedy routing
[13:07] <toad_> in the main network, the n+1 and n-1 links ensure that this doesn't happen
[13:08] <toad_> (greedy routing does not do backtracking)
[13:08] <d3wd3> ahh, so you are also good with algorithms toad ?
[13:08] <d3wd3> do you wear glasses ?
[13:08] <toad_> sleon|tuX: XP 2800+ with 1.5G of RAM and 3 IDE disks in LVM
[13:08] <sanity_> toad: perhaps we need to grow the network along the short links only
[13:09] <toad_> sanity_: huh?
[13:09] <toad_> hmmm
[13:09] <toad_> so we start with 500, then we add 499, then we add 501, ... ?
[13:09] <toad_> that seems grotesquely artificial...
[13:09] <sanity_> or start with 1, then add 2, then add 3...
[13:10] <sanity_> the whole simulation is grotesquely artificial
[13:10] <d3wd3> toad_ do you look like the guys on http://oldweb.uwp.edu/academic/mathematics/usaco/ ?
[13:11] <sanity_> d3wd3: no, we look much geekier than them
[13:11] <d3wd3> can't imagine much geekier than that
[13:11] <sanity_> toad: the point is we need to grow the network in a way that it retains its small world characteristic as it grows
[13:12] * Hory (~Miranda@82.76.81.56) Quit (Read error: 145 (Connection timed out))
[13:13] <toad_> Average hops for greedy routing: 4.98342
[13:13] <toad_> yay, it didn't overflow
[13:13] <toad_> Cycle 0: Average path length: 4.823134578811947, Success probability: 0.8939, inserts: 1.0
[13:13] <d3wd3> toad how old are you ?
[13:13] <d3wd3> 32 ? :)
[13:14] <toad_> d3wd3: 24
[13:14] <toad_> Cycle 0: Average path length: 5.06345113645918, Success probability: 0.892025, inserts: 1.0
[13:14] <toad_> Cycle 2: Average path length: 3.8773799033352025, Success probability: 0.936225, inserts: 1.0
[13:14] <toad_> Cycle 4: Average path length: 3.745376880422382, Success probability: 0.942275,inserts: 1.0
[13:15] <toad_> it'll start adding nodes when it reaches 95%
[13:16] <toad_> hmm, i wonder why it's falling...
[13:16] <toad_> Cycle 14: Average path length: 3.637923740017981, Success probability: 0.94545,inserts: 1.0
[13:17] <toad_> just wobbling maybe.. /me restarts with 90% threshold
[13:17] <toad_> Average conns per node: 5.46
[13:17] <toad_> Average hops for greedy routing: 5.10851
[13:18] <toad_> sanity_: average path length is falling, but psuccess seems to be falling with it
[13:19] <sanity_> as the network grows?
[13:19] <toad_> as the network stands still...
[13:19] <sanity_> strange
[13:21] <toad_> Average conns per node: 5.14
[13:21] <toad_> Average hops for greedy routing: 5.18452
[13:26] <toad_> hmmm
[13:26] * toad_ finds a bug...
[13:29] <d3wd3> so can a freenet site die out only if the original poster's node dies out ?
[13:29] <toad_> d3wd3: no, a freenet site dies out if nobody requests it
[13:29] <toad_> or if it's a DBR, if the author stops inserting it daily
[13:29] <greycat> (or whatever the DBR's expiration period is)
[13:30] <greycat> daily is merely the default on all the insertion tools
[13:30] <d3wd3> i get a lot of Error: Data Not Found
[13:30] <greycat> there's a lot of data that was once inserted, but which is now unfindable/gone
[13:30] <toad_> sanity_: i suspect that the fact that usually when i start the sim up, it hangs close to the end, tells me something...
[13:30] <d3wd3> you have a freenode sim ? :)
[13:35] * d3wd3 (imnotreal@82.208.142.215) Quit ("The beatings will continue until morale improves.")
[13:40] <toad_> Average conns per node: 4.54
[13:40] <toad_> Average hops for greedy routing: 5.07841
[13:40] <toad_> found a bug...
[13:41] <toad_> looks v. promising
[13:42] <toad_> network growing, path length falling or stable or rising slightly, nodes being added, psuccess rising
[13:47] * toad_ runs one with random-growth in parallel
[13:47] * MatriXica (~Matrixica@83-97-75-249.evolva.ro) has joined #freenet
[13:47] <MatriXica> hello
[13:47] <toad_> hi
[13:48] <MatriXica> i already asked u i think
[13:48] <MatriXica> here free shell ?
[13:48] <toad_> not here no
[13:48] <toad_> this is freeNet
[13:48] <toad_> which is an anonymous distributed data store
[13:48] <MatriXica> i see
[13:48] <MatriXica> but i will get an free shell instead ?
[14:01] * MatriXica (~Matrixica@83-97-75-249.evolva.ro) Quit ()
[14:02] <sleon|tuX> toad_, wow, better then my workstation
[14:15] * thelema (~thelema@193.220.86.4) has joined #freenet
[14:19] * thelema (~thelema@193.220.86.4) Quit (Client Quit)
[14:20] * Hory (~Miranda@82.76.81.56) has joined #FreeNet
[16:04] * FallingBuzzard (~srademach@207.152.112.129) has left #freenet
[16:27] * pittaman (~mistery_b@d54C32DBA.access.telenet.be) Quit (Client Quit)
[16:49] * greycat (rfc1413@wooledge.org) Quit ("This time the bullet cold rocked ya / A yellow ribbon instead of a swastika")
[17:22] * Hory (~Miranda@82.76.81.56) Quit ("CyberLore.net - Recommendations on the best games, freeware and websites.")
[17:47] * sleon|tuX (test@p54A15705.dip.t-dialin.net) Quit (Read error: 60 (Operation timed out))
[18:05] * jokern (~bjorn@213.184.217.12) has joined #freenet
[18:06] <toad_> Cycle 1343: Average path length: 4.27305791757594, Success probability: 0.996675, inserts: 1.0
[18:06] <toad_> Nodes active: 1442
[18:06] <toad_> hrrm
[18:06] <toad_> how is that even possible?
[18:08] <toad_> is that even possible?!
[18:08] <toad_> four hops to span a network of 1,442 nodes with a pSuccess of 99.67% !
[18:10] <toad_> i have some issues building the original network though...
[18:11] <toad_> Average hops for greedy routing: 8.19785
[18:11] <toad_> Nodes active: 1464
[18:11] <toad_> Keeping 1464 keys
[18:11] <toad_> Cycle 1365: Average path length: 4.299724034119418, Success probability: 0.9965, inserts: 1.0
[18:11] <toad_> Load: max: 3039, min: 15, avg: 146
[18:17] <Elly> my node is down again
[18:17] <toad_> Elly: :<
[18:18] <Elly> toad_: It took my network down completely
[18:18] <Elly> no DNS resolving, couldn't even get to my router from any of my boxen
[18:18] <toad_> ouch
[18:18] <Elly> when I did /etc/init.d/freenet stop it stopped doing that
[18:18] * Elly shrugs
[18:19] <toad_> sounds like greycat's problem
[18:19] <toad_> he's never really managed to fix it...
[18:19] <Elly> I don't know what to do about it
[18:19] <Elly> Every time it happens freenet is on and killing freenet stops it
[18:19] <Elly> but my net sniffer (ethereal) never shows anything amiss
[18:21] * jokern (~bjorn@213.184.217.12) Quit ("Leaving")
[19:20] * goatee_ is now known as goatee
[22:12] * oierw (~mathew@cpe-66-91-35-90.hawaii.res.rr.com) Quit (Read error: 145 (Connection timed out))
[22:41] -lilo- [Global Notice] Good morning to all. URGENT NEWS ITEM on the website. THE SERVERS WILL BE SHUTTING DOWN IN FIVE MINUTES AND WILL BE DOWN FOR THE NEXT THREE DAYS. Please view the latest freenode news on http://freenode.net/news.shtml .... massive apologies for this situation. :(
[22:41] <kokje> is there any model for freenet which specifies the average processing time taken by the node
[22:42] -lilo- [Global Notice] Repeating, please check network status on http://freenode.net/news.shtml immediately! Thanks. :(
[22:43] -lilo- [Global Notice] Apologies for the problems with the website, please try http://freenode.net/news.shtml again.... repeating, urgent information on site. Thanks.
[22:44] * kokje (~kokje@c-24-127-127-236.hsd1.ca.comcast.net) Quit (Remote closed the connection)
[22:45] -lilo- [Global Notice] The site should be working now. Apologies for the inconvenience. Again, please reload: http://freenode.net/news.shtml
[22:52] -lilo- [Global Notice] Please check http://freenode.net/news.shtml one more time.... thanks!
[23:09] -lilo- [Global Notice] Hi all. Apologies for the surprise you received earlier, and I hope you don't mind that we got April Fool's Day out of our system a bit earlier than some. Happy April, and thank you for using freenode. :)
These logs were automatically created by Jay Oliveri with his gimp hapi on irc.freenode.net.