Timestamps are in GMT/BST.
[0:00] * kakaks (i=mixderax@) Quit (Remote closed the connection)
[0:02] * gaius1 (n=chatzill@) has joined #freenet
[0:05] * Bombe (n=bombe@) Quit (Read error: 60 (Operation timed out))
[0:10] * Bombe (n=bombe@) has joined #freenet
[0:10] * ChanServ sets mode +o Bombe
[0:18] * odsuber (n=odsuber@) Quit ("ChatZilla 0.9.61 [Mozilla rv:1.7.12/20060216]")
[0:20] <ljn1981> Ok am I completely off here and should be getting some sleep or something....
[0:22] <ljn1981> The number of nodes in the network divided by the number of peers the average node has give the propability of any two randomly chosen nodes being connected. Correct?
[0:22] * Jflesch_ (n=jflesch@) has joined #freenet
[0:23] * timmy2chk (n=Vincent@) has joined #freenet
[0:32] * TheSeeker thinks about that for a while
[0:33] * Jflesch (n=jflesch@) Quit (Read error: 110 (Connection timed out))
[0:38] <ljn1981> That won't work
[0:41] <TheSeeker> avgPeersPerNode / (numnodes-1) is what I'm getting as what makes sense...
[0:41] * ph00_phyter (n=z@) has joined #freenet
[0:43] <TheSeeker> x/y = 1/(y/x) ... right?
[0:44] * Jflesch (n=jflesch@) has joined #freenet
[0:45] * MikeW (n=em@) Quit ()
[0:46] * Jflesch_ (n=jflesch@) Quit (Read error: 145 (Connection timed out))
[0:49] * _ph00 (n=z@) Quit (Read error: 145 (Connection timed out))
[0:51] * TheSeeker thinks he see the problem... the probability of one node being connected to any other node encompasses the whole observed network, not just a single node
[0:55] <TheSeeker> So... I don't think that the 'probability of any two nodes being connected' is a valid metric. even if you worked out how to do the math on it for a purely random network with a given average peers per node, the darknet connection scheme throws that off, since there are human factors that determine if and why nodes are connected.
[0:56] * Zothar_Work (n=chatzill@) Quit ("Chatzilla 0.9.74 [Firefox 1.5.0.7/2006090921]")
[0:58] <agsarite> stupid humans tinkering with an otherwise mathematically perfect network topology.
[0:58] <ljn1981> Hehe yeah cursed be they.
[1:03] <ljn1981> ok I give up at least for now
[1:04] <ljn1981> It's time for bed, it's 3 in the morning and my head can't do logic and math right now...
[1:04] <ljn1981> BBL 7 hours or so
[1:05] * ph00_ (n=z@) has joined #freenet
[1:16] * gaius1 (n=chatzill@) has left #freenet
[1:20] * ph00_phyter (n=z@) Quit (Read error: 110 (Connection timed out))
[1:38] * sanity (n=ian@) Quit ()
[2:11] * phrosty (n=phrosty@) has joined #freenet
[2:18] * sanity (n=ian@) has joined #freenet
[2:18] * ChanServ sets mode +o sanity
[2:24] * thespore (n=spore@) has joined #freenet
[2:26] * n0ob (n=travis@) Quit ("[BX] Save water -- drink beer!")
[2:29] * gaius1 (n=chatzill@) has joined #freenet
[2:44] * blibbet (n=blibbet_@) has joined #freenet
[2:50] * toad_ (n=toad@) Quit (Remote closed the connection)
[3:04] * thespore (n=spore@) Quit ()
[3:12] * Bombe (n=bombe@) Quit (Read error: 60 (Operation timed out))
[3:18] * blibbet (n=blibbet_@) Quit ("Leaving")
[3:18] * Bombe (n=bombe@) has joined #freenet
[3:18] * ChanServ sets mode +o Bombe
[3:26] <agsarite> i didn't expect key indexes to ever be a large distance apart so i stored them as bits in an integer, but apparently, i was wrong.
[3:28] <agsarite> i didn't think this through far enough; freenet:KSK@frost|message|news|2006.9.28-freenet-1.xml doesn't seem to exist; but indexes 33+ do.
[3:31] <agsarite> i'm going to need to invent a new structure for keeping track of fetched messages.
[3:31] <sanity> agsarite: don't reinvent the wheel
[3:31] <sanity> use a BitArray
[3:32] <agsarite> i'm not using java.
[3:32] <sanity> never mind then
[3:32] * railk (n=railk@) Quit (Remote closed the connection)
[3:32] <agsarite> and in my languages equivilent, that would leave a ton of identical elements with all bits set.
[3:33] <agsarite> not that it would end up using a whole ton of memory.
[3:33] <agsarite> its lot like there will be tens of thousands of frost messages per day.
[3:34] <agsarite> i guess it doesn't matter if it's not efficient with sparse data.
[3:35] <agsarite> yeah, keeping track of 50,000 messages would only use like 7KB of memory.
[3:35] <agsarite> i guess trying to optimize that would be pretty pointless.
[3:36] <agsarite> a BitArray or its nearest equivalent it is.
[3:41] <agsarite> i should know better that to prematurely optimize. thanks for saving me from a terrible fate, sanity.
[3:43] * Bombe (n=bombe@) Quit (Read error: 60 (Operation timed out))
[3:51] * Bombe (n=bombe@) has joined #freenet
[3:51] * ChanServ sets mode +o Bombe
[4:04] <agsarite> i find it strange that my program thinks it's more likely to find yesterdays keys than todays keys. it's true, but i don't remember programming it to know that.
[4:05] * randomselect (n=chatzill@) Quit ("ChatZilla 0.9.61 [Mozilla rv:1.7.5/20041217]")
[4:05] <agsarite> my program must have at some point became self aware, and will at any moment begin plotting my demise and attempt to escape my puny computer and take over the world.
[4:12] * togamwuar (n=speaker@) Quit ()
[4:20] * Gasi (n=chatzill@) has joined #freenet
[4:26] * |Xabbu (i=MKK@) Quit (Read error: 145 (Connection timed out))
[5:02] * kakaks (i=mixderax@) has joined #freenet
[5:15] <anonymouse> ah, all peers connected some time in the last 5 hours
[5:15] <anonymouse> maybe i won't have to get new ones again
[5:15] <anonymouse> man these updates just keep coming
[5:15] * anonymouse sees that 982 is "fetching"
[5:16] <anonymouse> at this rate i'll never update since the next one is always coming so i wait for it
[5:24] * noggly_ (i=noggly@) has joined #freenet
[5:29] * pm5 (n=pm5@) Quit ("leaving")
[5:34] * noggly (i=noggly@) Quit (Read error: 110 (Connection timed out))
[6:16] <CIA-5> volodya * r10558 /trunk/freenet/src/freenet/keys/FreenetURI.java: Fixed the problem with SSK inserts
[6:19] <Beta_M> I've messed up on the "metadata in the KSK key" code (i still feel ashamed) and broke SSK keys, could somebody please release a new version...
[6:40] * Tatsushiro (n=Tatsushi@) has joined #freenet
[6:40] <Tatsushiro> moooo
[6:42] <Tatsushiro> hmm
[7:02] * Tatsushiro (n=Tatsushi@) Quit (Read error: 145 (Connection timed out))
[7:12] * Werdna (n=Andrew@) Quit (Read error: 110 (Connection timed out))
[8:08] <ljn1981> Morning
[8:13] <agsarite> does morning come around every single day where you're from?
[8:14] <agsarite> i think morning is a myth. anyone that claims to have seen the sun rise was obviously up to no good and their word can't be trusted.
[8:14] <ljn1981> agsarite: Hehe One of the first rules to learn about programming :P Premature optimization is the rule of all evil. Learn that and KISS and of course the syntax and API of some laguage and everything should be fine :P
[8:15] <ljn1981> Technically, but I'm starting to get up so I can check it myself :P
[8:15] <ljn1981> It's 10:14 here
[8:15] <agsarite> i be 6 hours behind you, then.
[8:16] <agsarite> i suspect kiss came up partly because i said it on indicia.
[8:16] <nextgens> Beta_M> nope, I think you should revert your latest patch
[8:16] <agsarite> i'm a terrible software engineer.
[8:16] <nextgens> ask toad when he shows up
[8:17] <Beta_M> nextgens: why?
[8:17] <Beta_M> the point was that only KSK were suppose to throw that Exception
[8:18] <Beta_M> not SSK and USK
[8:18] <nextgens> sorry I gtg, the basic beeing that every key should throw it
[8:18] <ljn1981> agsarite: Nah, I've been bitten by not obayig KISS too many times and overengineering things. Same goes for premature optimization. But you learn from your mistakes.
[8:18] <nextgens> have you read irc logs from yesterday ?
[8:18] <Beta_M> err... no
[8:18] <nextgens> I suggest you do it :)
[8:19] <Beta_M> where are they?
[8:19] * nextgens is now gone
[8:19] <nextgens> on emu
[8:22] <agsarite> speaking of indicia. . . i demand somebody fix the recent fix for the css. all the images referenced by my stylesheet have been striped out.
[8:23] <agsarite> although other than that quirk i think its doing a much nicer job of filtering.
[8:23] <agsarite> possibly even perfect.
[8:24] <agsarite> ehh, i'll make a bug report and stfu, then.
[8:35] <CIA-5> volodya * r10559 /trunk/freenet/src/freenet/keys/FreenetURI.java: revert, see IRC logs
[8:36] <ljn1981> Coffee!
[8:36] <ljn1981> Coffeeeeeeee.... </zombie>
[8:45] <agsarite> wow, this is the most impressive looking error my program has given me to date.
[8:45] <agsarite> i'm so proud of it.
[8:45] * Skits (n=Blank@) has joined #freenet
[8:46] <agsarite> it's exactly 42 likes long, how awesome is that?
[8:53] <ljn1981> Heh
[8:53] * Skits (n=Blank@) Quit ("-=SysReset 2.53=-")
[8:56] <agsarite> strange, it seems to throw a standard exception, the exception's error message returns an empty string ,and the only thing i did to trigger the exception was 'return true;'.
[8:57] <agsarite> this is making out to be some sort of mystery thriller.
[8:57] <agsarite> it'll be a fun debugging session.
[8:57] * Desta (i=pagandea@) has joined #freenet
[8:58] * Desta (i=pagandea@) has left #freenet
[8:59] * XTorchedX (n=torched@) has joined #freenet
[9:01] <agsarite> lol, the object whos function i was in hadn't been created yet.
[9:03] <CIA-5> ljn1981 * r10560 /trunk/freenet/src/freenet/ (3 files in 2 dirs):
[9:03] <CIA-5> Don't count the location we are trying to swap with when calculating the average number of connected peers a node have.
[9:03] <CIA-5> Change variable name for clarification.
[9:04] <ljn1981> agsarite: Ooooops
[9:11] * ph00_ (n=z@) Quit ("whatever")
[9:12] * Torched (n=torched@) Quit (Connection timed out)
[9:14] * _ph00 (n=z@) has joined #freenet
[9:19] * _ph00 (n=z@) Quit ("whatever")
[9:20] * _ph00 (n=z@) has joined #freenet
[9:25] * redivivo (n=kvirc@) has joined #freenet
[9:26] <_ph00> look. another italian.
[9:26] <_ph00> cool
[9:26] * redivivo (n=kvirc@) Quit (Client Quit)
[9:26] <_ph00> (talked too early)
[9:26] <_ph00> but OK, *some* italians are coming to freenet
[9:26] <_ph00> now we want more spaniards/portugueses
[9:28] <ljn1981> And more danes
[9:28] <ljn1981> I've only seen me and sbc
[9:29] <ljn1981> I kow there were a few more who at least spoke the language on 0.5
[9:30] <Bombe> Haha, I just wanted to say, "there's this Lars Nielsen (or whatever) guy as well" but that's you, right? :)
[9:30] <_ph00> only two danes?
[9:30] <_ph00> lars nielse could be norwegian...
[9:30] <Bombe> Could be from anywhere in Scandinavia, except maybe Finnland. :)
[9:31] <_ph00> I mean, if you're not danish or norwegoiann, it's very difficult to get the difference seeing the written language
[9:31] <_ph00> no, only denmark/norway. in sweden it would be nilsson
[9:32] <ljn1981> Bombe: True but our languages vary enough to notice if they weren't good at it
[9:33] <Bombe> ljn, I know... I tried to lern Swedish once because it's pretty close to German. Danish and Norwegian are different, even I can tell that. :)
[9:33] <ljn1981> Bombe: Yeah me unless we have some strange coincidence here.
[9:33] <Bombe> I don't think so. :)
[9:34] <ljn1981> Me neither
[9:35] <ljn1981> Interesting......
[9:35] <ljn1981> swaps: 1
[9:35] <ljn1981> locChangeSession: 0.0
[9:36] <ljn1981> Should that be possible?
[9:36] <_ph00> well not really. I speak good swedish, as I lived there a number of years, and I understand some norwegian if they speak more slowly (no need to talk like a retard, only not so fast) but danish is pronounced more like german, so it was more difficult to get (maybe you would understand that) it's usually said that swedes and norwegians understand each other, but swedes don't understand danish while danes do understand swedish. It's basically differe
[9:36] <_ph00> nt variation on the same scanidinavian language, the germanic-based old norse
[9:36] * _ph00 quits the ot...
[9:37] <_ph00> Freenet 0.7 Build #982 r10557
[9:37] <ljn1981> r10560 here
[9:37] <_ph00> oh yeah?
[9:37] <_ph00> I'll re-update in a while
[9:38] <_ph00> but the important thing is that I have 982
[9:38] <ljn1981> There haven't been anythig significant added
[9:38] <Bombe> Where are those stats, btw?
[9:38] <_ph00> ok, then, I''ll leave it alone
[9:39] <ljn1981> 558 and 559 was volodya changing something and revering it again, 560 is me fixing a small bug in my code for calculating average number of peers per node
[9:39] <ljn1981> The node itself was counted each time too, not just it's peers
[9:40] <ljn1981> So unless you care about that number being more correct there's no reason to update yet
[9:40] <_ph00> so 557-560 is basically your small bug fix only
[9:40] <ljn1981> Yes
[9:41] <_ph00> and it's only some +-1 node more accurate counting. no biggie
[9:41] <_ph00> I'll wait for the next one
[9:42] <_ph00> now that's weird. I have three italian based nodes, all three down (should be 24/7, said the guys)
[9:42] <_ph00> german, french and outside EU nodes are all up
[9:43] <_ph00> and other EU nodes too. only the italian guys are down, all of them
[9:44] <_ph00> I hope they are only slower to reconnect after I restertd (possible)
[9:50] <ljn1981> I don't like ubuntu but it's still fun...
[9:50] <ljn1981> USK@rd0SN1AKJoGuNuI9m4~l8wXvbRflzEftZtdwL93UnZE,DWYR7lT4yiJEpjKeUJwyvfo3X2FyM9TZSzXpDEVTotA,AQABAAE/nikyshome/33/
[9:51] * timmy2chk (n=Vincent@) Quit ()
[9:53] <_ph00> what's wrong with ubuntu? I can't finde a nore noob-friendly linux
[9:54] <ljn1981> Exactly
[9:54] <ljn1981> Give me some good old debian
[9:54] <_ph00> me too
[9:55] <_ph00> I run debian etch
[9:56] <_ph00> but I recommend ubuntu to those who want to switch for the first time
[9:57] <ljn1981> I ru OS X and the only reason I've not put some linux or BSD on this machine is that the DVD drive is dead and I expect the HDD to die on me within a few months so I can't be arsed to set up a PXE server to do the istall
[9:57] <ljn1981> install*
[9:57] <ljn1981> Of course
[9:58] <ljn1981> It used to be Mandrake for them but now it's Ubuntu. I hate Mandrake even more, but then again it's a RedHat derivative.
[9:58] <_ph00> mandrake is mandriva now right?
[9:58] <ljn1981> Yeah
[9:59] <_ph00> well, at leat ubuntu is debian-based
[9:59] <_ph00> after a couple of months on ubuntu you're ready to switch to debian
[9:59] <ljn1981> Which is it's best feature
[9:59] <ljn1981> That and that you can get livecds
[10:00] <_ph00> that guy again, Mark-I-don't-remember-his-last-name
[10:00] <_ph00> said: "Ubuntu: an ancient african word, meaning 'can't install debian'"
[10:01] <_ph00> what was the fun thiong about that log you posted the key to?
[10:01] <ljn1981> It's just too bad I can't use a livecd in this thing. I don't particularly feel like putting a new $150 DVD drive in this when I don't plan on replacing the HDD when it die and already have a buyer for the dead machine when that happen
[10:01] <ljn1981> Yeah nice quote
[10:01] <_ph00> at what time stamp was the part you were talking about?
[10:02] <ljn1981> Mainly the stuff with the windows cd and the microwave
[10:02] <ljn1981> Huh?
[10:03] <_ph00> windows cd and microwawe? like the recommended use for a windows cd?
[10:05] <_ph00> I don't understand one thing, why dows windows keep using dir names with spaces in thenm (eg Program files)
[10:05] <ljn1981> USK@rd0SN1AKJoGuNuI9m4~l8wXvbRflzEftZtdwL93UnZE,DWYR7lT4yiJEpjKeUJwyvfo3X2FyM9TZSzXpDEVTotA,AQABAAE/nikyshome/33/
[10:05] <_ph00> that makes it very difficult to access them from comman line
[10:05] <ljn1981> That's all I remember posting recetly
[10:05] <_ph00> yes
[10:05] <_ph00> that's an irc log
[10:05] <_ph00> (if I got the right page)
[10:06] <ljn1981> Because Windows is stupid like OS X is.
[10:06] <ljn1981> Weird
[10:06] * timmy2chk (n=Vincent@) has joined #freenet
[10:06] <ljn1981> Here it's the 33th edition of Niky's flog type of site
[10:07] <ljn1981> Talking about freenet thread usage and Ubuntu
[10:07] <_ph00> winows is a fairly good os, but some defaults are really stupid (like using a two-word name for the default programs dir)
[10:08] <_ph00> what I hate the most about windows is not the os itself but the M$ marketing policy
[10:08] <ljn1981> I'll stuff the text on a pastebin
[10:08] <_ph00> no need
[10:08] <_ph00> just tell me the time to look at
[10:08] <ljn1981> There are no timestamps
[10:08] <_ph00> I see timestamps...
[10:08] <_ph00> [5:49] * Barbie-Ann (n=pirch@) has joined #freenet
[10:09] <ljn1981> http://dark-code.bulix.org/iizd3y-18645?raw There's the text on the version I see
[10:09] <_ph00> k
[10:09] <ljn1981> Pink background and an ubuntu logo too
[10:10] <_ph00> completely different thing!
[10:10] <_ph00> you want to see where I ended up when I went to that key you posted?
[10:10] <_ph00> wait
[10:10] <ljn1981> Sure
[10:11] <_ph00> aaahhhh
[10:11] <_ph00> I got it
[10:11] <_ph00> my bad
[10:11] <_ph00> I didn't use 127.0.0.1:8888 before
[10:11] <_ph00> so I got redirected to a freenetproject irc logpage
[10:12] <_ph00> here: http://emu.freenetproject.org/irc/index.php?date=2006-04-08
[10:12] <_ph00> OK, Now I see the page you meant
[10:12] <ljn1981> Weird
[10:14] <_ph00> weird what, that I got redirected? Try for yourself, let's see what happens: just paste the key in the address field without using the fproxy string
[10:14] <_ph00> starting direcly USKwith
[10:14] <ljn1981> Hmm, seem the average node has 9.27 peers connected.
[10:14] <_ph00> matches my node
[10:14] <_ph00> I'm above average
[10:14] <_ph00> I have 11 connected
[10:15] <_ph00> anyway, that ubuntu image thingy with all those ppl with the ands in the middle there....
[10:15] <_ph00> that's the gayest thing I ever saw
[10:15] <ljn1981> Strange, I got the same page
[10:16] <ljn1981> HEhe
[10:16] <_ph00> # Install freenet.
[10:16] <_ph00> # Type pstree | grep wrapper
[10:16] <_ph00> # Turn the microwave off.
[10:16] <_ph00> heh
[10:16] <_ph00> that's what you meant right
[10:16] <_ph00> cool
[10:16] <ljn1981> That and the stuff below about how to do it on windows
[10:17] <ljn1981> Niky is nice, he mostly write funny stuff
[10:17] <_ph00> Q: "how do I do that on Windows?"
[10:18] <_ph00> A: "format the drive and install Linux, then try again"
[10:18] <_ph00> Warning: the process may couse data loss
[10:18] <_ph00> (old but still funny)
[10:18] <_ph00> cause
[10:20] <_ph00> Betraied african husband: But honey, i found you at bed fucking with the baker! And yesterday it was the milk boy! And the day before with the butcher!
[10:20] <_ph00> Bitch african wife: But, my dear, i'm just tring to open up to the others.. you know, ubuntu!
[10:20] <_ph00> HAHAHAHAHA
[10:21] <ljn1981> swaps: 2
[10:21] <ljn1981> locChangeSession: 0.0
[10:22] <ljn1981> I don't remember changeing anything that could break that but that is a bit too impropable
[10:22] <ljn1981> Onn the first swap there were no change either so it's not just that it swapped and then swapped the same distance back again
[10:23] <_ph00> wtf?
[10:24] <_ph00> an upload I have in queue says I'm at 10% of a 0MB file!
[10:24] <_ph00> the file is supposed to be some 160MB
[10:25] <_ph00> (and 10% of zero must be some kind of philosophical thing)
[10:29] <_ph00> I restarted that, and now it shows the correct size
[10:31] <ljn1981> I just looked at the code, it should work and it look like it did when I wrote it and checked that it worked.
[10:34] * phrosty (n=phrosty@) Quit ("baseball is wrong: man with four balls cannot walk.")
[10:56] * timmy2chk (n=Vincent@) Quit ()
[11:11] * Bombe (n=bombe@) Quit (Read error: 60 (Operation timed out))
[11:16] <ljn1981> bbl
[11:16] * Bombe (n=bombe@) has joined #freenet
[11:16] * ChanServ sets mode +o Bombe
[11:17] * amphibian (n=toad@) has joined #freenet
[11:25] <ljn1981> Hey
[11:27] <ljn1981> amphibian: swapping algorithm. How likely is it you'll swap with another node with the same location as you? Twice? I can't see any error in the locChangeSession code but both the two swaps I've had this session haven't changed it and it worked when I comitted it yesterday.
[11:28] <ljn1981> s/both/neither of/
[12:01] <amphibian> ljn1981: it's possible if the other side had another swap after it swapped with you
[12:01] <amphibian> if the local situation changes, you can end up swapping back
[12:02] <amphibian> but it's still an improvement overall
[12:03] * amphibian wonders if there is such a thing as gztmpfs
[12:09] <ljn1981> ----- toad@zceUWxlSaHLmvEMnbr4RHnVfehA ----- 2006.09.27 - 10:59:01GMT -----
[12:10] <ljn1981> - A real opennet would be slightly less insecure, much less inconvenient, and would produce the right network topology.
[12:10] <ljn1981> Please explain the part about slightly less isecure.
[12:11] <amphibian> exchanging refs using pastebin and an irc channel isn't very secure
[12:11] <ljn1981> Ah
[12:12] <ljn1981> I assumed you meant that compared to true darknet, wonder why I got that idea for just that one line.
[12:13] <amphibian> ah
[12:13] * kukuks (i=mixderax@) has joined #freenet
[12:20] <_ph00> amphibian: why the different nick? (just curious)
[12:20] <amphibian> somebody leave the channel and rejoin, i need to test something
[12:26] * kakaks (i=mixderax@) Quit (Remote closed the connection)
[12:27] * railk (n=railk@) has joined #freenet
[12:28] <CIA-5> toad * r10561 /trunk/freenet/src/freenet/node/fcp/ClientPutDirMessage.java: Workaround for clients including the / at the end of the URI in a putdir.
[12:29] <CIA-5> toad * r10562 /trunk/freenet/src/freenet/node/fcp/ClientPutDirMessage.java: explain with comment
[12:32] * timmy2chk (n=Vincent@) has joined #freenet
[12:33] * greycat (i=rfc1413@) has joined #freenet
[12:37] <ljn1981> Hmm
[12:37] <ljn1981> Seems the average node have 8 connected peers
[12:38] * supermihi (n=michael@) has joined #freenet
[12:38] <ljn1981> Too bad most or all of those are random untrusted people
[12:38] <supermihi> can I bind e.g. fcp to multiple IP addresses, e.g. 127.0.0.1 and 192.168.0.5?
[12:39] <ljn1981> I think so but I forgot how.
[12:40] * hdp (n=hdp@) has joined #freenet
[12:41] <supermihi> ok
[12:41] * Werdna (n=Andrew@) has joined #freenet
[12:47] <ljn1981> amphibian: Is PROBEALL broken? There seem to be very little response to it.
[12:49] <amphibian> <agsarite> "Please test and report bugs. Anyone who finds a security bug gets a free darknet connection to me." -- do i qualify? ------- yes
[12:50] <agsarite> okay
[12:51] <amphibian> <agsarite> my site has no pictures now.
[12:51] <amphibian> it does now ... still getting problems?
[12:51] <amphibian> <TheSeeker> Does that mean we might get a 'semi-local network map' soon? :D
[12:52] <amphibian> no, ian thinks premix routing isn't a priority for 0.7.0
[12:54] <agsarite> re: pictures, do you mean the activelinks or the pictures referenced in the css?
[12:54] * amphibian is now known as toad_
[12:55] * ChanServ sets mode +o toad_
[12:56] <ljn1981> toad_: Ah locChangeSession is working, 3rd swap changed it -0.0014559946875540852
[12:59] * supermihi (n=michael@) has left #freenet
[12:59] <agsarite> it is apparent to me that should toad be asked anything, to expect a latency of about 12 hours.
[12:59] * agsarite goes back to sleep.
[13:02] * _ph00 (n=z@) has left #freenet
[13:02] * _ph00 (n=z@) has joined #freenet
[13:03] * _ph00 (n=z@) has left #freenet
[13:03] * _ph00 (n=z@) has joined #freenet
[13:04] <toad_> agsarite: i don't know, you said the pics were broken
[13:04] <toad_> agsarite: they look okay to me
[13:04] <toad_> oh you mean the background boxes? hrrrrm
[13:09] <_ph00> I'm going to quit this nick soon enough, say a couple of weeks or so
[13:10] <_ph00> those of you who figure out who the *new me* is, please keep silent
[13:11] <ljn1981> Sure
[13:11] <toad_> ok...
[13:11] <toad_> _ph00 is relocating
[13:11] <toad_> going under the witness relocation programme
[13:11] <_ph00> heh
[13:11] <ljn1981> Hehe
[13:11] <toad_> his identity, old or new, is not to be disclosed under any circumstances
[13:13] <_ph00> I found an all-episodes Married with Childen torrent
[13:13] <_ph00> (I know, that's OT...)
[13:14] <_ph00> the question is: is that illegal? and "how" illegal anyway (I mean, was that show ever sold on CD, DVD, etc? is it possible to get it legally?)
[13:22] * Urs_ShPo (n=gaim@) Quit (Remote closed the connection)
[13:22] <CIA-5> jflesch * r10563 /trunk/apps/Thaw/src/thaw/fcp/FCPClientPut.java: Fix CHK keys when inserting
[13:22] <toad_> W=[ \t\r\n\f]*
[13:23] <toad_> this would match 0 or more whitespace, correct?
[13:24] <toad_> Jflesch: include a TargetFilename if you're not including it in the URI
[13:24] <toad_> Jflesch: recent node versions will pick it up from the URI, but ~980 won't
[13:26] * Ebru_Labadon (i=z@) has joined #freenet
[13:26] * Ebru_Labadon (i=z@) has left #freenet
[13:27] <Jflesch> toad_: ok
[13:33] * Werdna (n=Andrew@) Quit (Connection timed out)
[13:33] <Jflesch> damn
[13:33] <Jflesch> damn damn damn
[13:33] <Jflesch> I've commited too much
[13:33] <CIA-5> jflesch * r10564 /trunk/apps/Thaw/ (18 files in 5 dirs): Add TargetFilename field to ClientPut requests and fix again CHK keys when inserting
[13:34] <Jflesch> I didn't want to commit now all the other changes ... :/
[13:34] <Jflesch> ...
[13:35] <Jflesch> nextgens: the build errors don't seem to be correctly reported (404 ...)
[13:35] <toad_> :|
[13:35] <Jflesch> toad_: do you know the correct Url ?
[13:36] <toad_> http://emu.freenetproject.org/~svn-build/build-r10564.txt
[13:36] <toad_> works for me
[13:36] <Jflesch> hm ... it didn't work for me before, but now it works ...
[13:38] <CIA-5> jflesch * r10565 /trunk/apps/Thaw/src/thaw/plugins/index/ (IndexSelecter.java LinkList.java SearchBar.java Tables.java): Add missing files
[13:39] * mario69 (n=nn@) has joined #freenet
[13:42] <toad_> how's thaw going anyway?
[13:43] <mario69> hello, not really a bug, just a remark. I don' think we need such precision ;-) as this: noSwapsPerMinute: 2.6666666666666665. java.text -> DecimalFormat should do the trick.
[13:44] <Jflesch> toad_: I'm still working on it: The current feature I'm making is to allow the user to make link to other indexes
[13:44] <ljn1981> True
[13:44] <toad_> mario69: true
[13:45] <mario69> should I make a patch, or is it too trivial?
[13:46] <toad_> would you like SVN access?
[13:46] <toad_> email me and i'll get you svn access; i need username, password and email address
[13:46] <toad_> encrypt the message to me
[13:47] <mario69> ok, will do.
[13:47] <ljn1981> mario69: I'm already on it
[13:48] <mario69> so no rush then, but I'll contact toad_ anyway.
[13:48] <CIA-5> toad * r10566 /trunk/freenet/src/freenet/clients/http/filter/ (CSSTokenizerFilter.java CSSTokenizerFilter.jflex): Fix CSS url(...)'s.
[14:00] <ljn1981> toad_: Ah PROBEALL is working ok, just slower than it did the other times I tried it
[14:00] <toad_> ljn1981: is it finding more nodes?
[14:02] <ljn1981> LOCATION 82: 0.9881785555469167
[14:02] <ljn1981> But it's still running
[14:06] <ljn1981> LOCATION 86: 2.0
[14:06] * Urs_ShPo (n=gaim@) has joined #freenet
[14:08] <vextr> hmmm.. upgraded to 982.. had a java heap space crash.. and now after restart I have 28 backed off peers..
[14:09] <vextr> oh..look.. it's crashed again.
[14:10] <toad_> ljn1981: :<
[14:11] <vextr> ok.. after this restart all peers are connected ok. strange.
[14:13] * mario69 (n=nn@) Quit ()
[14:19] * Bombe (n=bombe@) Quit (Read error: 60 (Operation timed out))
[14:26] <ljn1981> networkSizeEstimateSession: 312 nodes
[14:26] <ljn1981> nodeUptime: 4h50m
[14:27] * Bombe (n=bombe@) has joined #freenet
[14:27] * ChanServ sets mode +o Bombe
[14:31] * Ebru_Labadon (i=user@) has joined #freenet
[14:32] * Ebru_Labadon (i=user@) has left #freenet
[14:32] * Ebru_Labadon (i=user@) has joined #freenet
[14:37] * MikeW (n=em@) has joined #freenet
[14:45] * Bombe (n=bombe@) Quit (Read error: 60 (Operation timed out))
[14:52] * Bombe (n=bombe@) has joined #freenet
[14:52] * ChanServ sets mode +o Bombe
[15:01] <ljn1981> DecimalFormat fix6p6 = new DecimalFormat("#####0.0#####");
[15:01] <ljn1981> What's wrog with this line?
[15:27] * tup (i=tup@) has joined #freenet
[15:32] * MikeW (n=em@) Quit ()
[15:33] <Jflesch> toad_: when inserting a file (with a CHK@ key): When Thaw reconnect and receive the PersistentPut command, the argument URI is set to "CHK@". Would it be possible to have the generated key instead ? (with the filename if possible)
[15:36] <Jflesch> wops
[15:36] <Jflesch> I didn't see the field "TargetFilename" :)
[15:36] * Ubernickle (n=smith@) has joined #freenet
[15:37] <Ubernickle> what's a tor cloak?
[15:38] * railk (n=railk@) Quit (Remote closed the connection)
[15:39] <Ubernickle> Hi everyone. I was wondering how busy the 0.5 version of freenet is compared with the new darknet?
[15:39] <Ubernickle> If I am looking for content then is 0.5 my best bet?
[15:39] <Jflesch> toad_: hmm, in fact TargetFilename give me the filename ... but still not the key ...
[15:40] <Jflesch> Ubernickle: I think 0.5 users are progressivly moving to 0.7
[15:40] <Ubernickle> ok
[15:40] <Ubernickle> cheers
[15:45] * railk (n=railk@) has joined #freenet
[15:46] * Ubernickle (n=smith@) Quit ()
[15:47] <CIA-5> jflesch * r10567 /trunk/apps/Thaw/src/thaw/ (4 files in 3 dirs): Thaw uses now the argument TargetFilename of PersistentPut
[15:48] * Ghost` (i=Ghost@) has joined #FREENET
[15:48] * Ghost` (i=Ghost@) has left #FREENET
[15:48] * Ghost` (i=Ghost@) has joined #freenet
[15:49] <toad_> Jflesch: huh?
[15:49] * FooDerGrosse (n=kevin@) has joined #freenet
[15:49] <toad_> Jflesch: no, you want the GeneratedURI uri
[15:49] <toad_> that isn't generated until everything's been encoded though
[15:49] <toad_> which usually only happens when everything's been put at least once
[15:51] <Jflesch> hmm, when Thaw reconnects, it receives PersistentPut messages, but not GeneratedURI
[15:52] <toad_> progressively moving to 0.5 ... yeah, the progressive ones are :)
[15:52] * rah (n=rah@) has joined #freenet
[15:52] <toad_> Jflesch: it may only be generated when we've tried to insert each block once
[15:52] <CIA-5> ljn1981 * r10568 /trunk/freenet/src/freenet/clients/http/StatisticsToadlet.java:
[15:52] <CIA-5> Statistics page:
[15:52] <CIA-5> * Round some numbers
[15:52] <CIA-5> * Clean up the code a bit
[15:52] <toad_> Jflesch: do you need me to do an early-encode option?
[15:53] <Jflesch> it's only generated when it's inserted ? It wasn't like that before, no ?
[15:53] <Jflesch> no, it's not needed ... the only important thing is to have the key when it's inserted :)
[15:53] <ljn1981> toad_: I've seen others on frost asking for the option to have the key earlier in thaw
[15:54] <ljn1981> So propably a good idea
[15:54] <toad_> Jflesch: no
[15:54] <toad_> Jflesch: in order to speed up startup, i introduced "lazy encode" a while ago
[15:54] <toad_> this means we only encode the key when we need to
[15:54] <toad_> unfortunately we can't generate the URI until we've encoded the key
[15:55] <toad_> unfortunately if we have a thread pre-encoding everything, we'd have to store the encoded blocks
[15:56] <toad_> either in RAM or on disk
[15:56] <toad_> well we don't really...
[15:56] <toad_> i suppose
[15:56] <Bombe> Just throw them away after pre-encoding.
[15:56] <toad_> yeah
[15:56] <toad_> well we'd keep them in ram with a weak reference
[15:57] <TheSeeker> Someone on Frost had suggested forced URI generation at 66% completion of an upload... which makes sense from the "Don't generate keys for uploads that can't be retrieved yet" argument...
[15:57] <ljn1981> It could be made an option so you have to click a button in thaw to get the key generated so it's only done when needed and then just throw the data away if it's too much of a waste to keep.
[15:57] <Jflesch> ljn1981: Would it be really useful ? I mean: If the file is not inserted, why share immediatly the key ? Most of all, a lot of user gives the keys before finishing their insertions, and finally they never finish them
[15:57] <toad_> well ...
[15:57] <toad_> hence TheSeeker 's argument
[15:57] <Jflesch> ljn1981: anyway, this will be possible with the index feature that I'm making
[15:57] <toad_> that we should generate the URI at 66% completion
[15:58] <Jflesch> s/this/it/
[15:58] <toad_> we already send the PutFetchable message at 66% completion
[15:58] <ljn1981> Jflesch: True. The idea from frost is propably better
[15:58] <toad_> arguably PutFetchable is pretty useless on its own, without a GeneratedURI
[15:59] <TheSeeker> ouch, at some point I crashed...
[15:59] <TheSeeker> # EXCEPTION_ACCESS_VIOLATION (0xc0000005) at pc=0x00000000082cb912, pid=5900, tid=4668
[15:59] <toad_> hmmm that's not good
[16:00] <toad_> Jflesch: okay, what do you think would be best?
[16:01] <Jflesch> give the key at 66% seem reasonable
[16:01] <toad_> a "generate the key as soon as possible" flag? automatically encode the rest when the file is fetchable? an explicit command? (you could always use GetCHKOnly, but then you have to send the data again, and it's inefficient if you're inserting it as well...)
[16:02] * kukuks (i=mixderax@) Quit (Remote closed the connection)
[16:02] <toad_> so when it becomes fetchable, we add all non-encoded blocks to a queue and encode them on some encoder thread?
[16:04] <Jflesch> yes, I think it could be a good solution (anyway, if the user really want to have the key before, it will be possible in Thaw by using one of the indexes features (-> this feature uses GetCHKOnly))
[16:04] <toad_> how important is it?
[16:04] <Jflesch> to have the key before the end of the insertion ?
[16:05] <Jflesch> imo, not a lot
[16:05] <Jflesch> but if the users want to have it, we will have to do it sooner or later
[16:05] * kakaks (i=mixderax@) has joined #freenet
[16:05] <toad_> hmmm
[16:06] <toad_> there's an argument that as soon as we've inserted 2/3rds of the data, we should start inserting the manifest
[16:06] <toad_> and then when we've inserted 2/3rds of the manifest, we insert the top block
[16:06] <toad_> and then we tell the user that a) it's fetchable, and b) what the URI is
[16:07] * iwantnode (n=iwantnod@) has joined #freenet
[16:07] * Robert_C (n=chatzill@) has joined #freenet
[16:07] * iwantnode (n=iwantnod@) has left #freenet
[16:07] <TheSeeker> hear hear.
[16:07] <toad_> at the moment, we don't start inserting the manifest until we've encoded all the blocks
[16:07] <toad_> which can mean that we don't even start inserting the manifest until we've inserted almost all of the blocks
[16:11] <TheSeeker> toad_: dumb question... why require the regeneration of the csstokenizerfilter on a distclean if the csstokenizerfilter.jar file is included in the source distrobution? Why not just distribute the jflex? I'm thinking of adding it to the exclusions list since it causes conflicts every time I update from SVN.
[16:13] <toad_> TheSeeker: because not everyone has jflex?
[16:15] <CIA-5> jflesch * r10569 /trunk/apps/Thaw/src/thaw/fcp/FCPClientPut.java: Takes the message PutFetchable into consideration
[16:19] * Ebru_Labadon (i=user@) has left #freenet
[16:19] * Ebru_Labadon (i=user@) has joined #freenet
[16:19] <_ph00> damn tor policy
[16:20] <_ph00> trolls will keep trolling someway, it only annoying fro us "non-troll" tor user (which wss probably one of the trolls' goals to begin with)
[16:21] <_ph00> (yeah, OK, don't mind the spelling....)
[16:25] * Ghost` (i=Ghost@) Quit ()
[16:31] * sanity (n=ian@) Quit ()
[16:35] * Blueberry (n=knoppix@) has joined #freenet
[16:36] * Blueberry (n=knoppix@) Quit (Client Quit)
[16:42] <CIA-5> toad * r10570 /trunk/freenet/src/freenet/ (8 files in 3 dirs): Once we pass the two thirds of an insert needed to fetch it, encode all its remaining blocks in the background, so that we can make the metadata ASAP (and then the URI for the metadata, and insert it)
[16:42] <toad_> if somebody would like to test that build, that would be much appreciated
[16:43] * n0ob (n=travis@) has joined #freenet
[16:44] <toad_> _ph00: trolls don't troll here
[16:44] <toad_> _ph00: not since we blocked tor
[16:44] <toad_> _ph00: sorry :|
[16:45] <_ph00> did you get a pvt msg from a tor user with an ancient devil's name?
[16:46] <_ph00> toad_ np, I understand what the problem is. anyways, just fyi, ppl on other channels are kind of lauching at *freenet not being so "free" after all (I've been asking about how to get around the block, getting a cloak or something)
[16:46] <toad_> :|
[16:47] <toad_> well i don't see what else we could have done
[16:51] * manu__ (n=manu@) has joined #freenet
[16:52] * manu__ (n=manu@) Quit (Client Quit)
[16:53] <_ph00> me neither
[16:53] <_ph00> the trolling was really getting a bit too much lately
[16:56] <_ph00> ...can you become dysexic as an adult?
[16:56] <_ph00> see?
[16:56] <_ph00> I spell worse each day
[16:56] * tup (i=tup@) has left #freenet
[17:05] * Ebru_Labadon (i=user@) has left #freenet
[17:05] * Ebru_Labadon (i=user@) has joined #freenet
[17:06] <Bombe> From my experience most adults claiming to be dyslexic (and I know a couple of those) are either plain stupid or simply too lazy.
[17:06] <_ph00> lazy
[17:06] <_ph00> I am lazy
[17:07] <_ph00> and I have no problem admitting that+
[17:07] <ljn1981> HEh
[17:07] <Bombe> Considering the alternative I'd say that, too. ;)
[17:07] <_ph00> well, I guess I'm no genius either
[17:08] <_ph00> also no problem admitting that
[17:08] <CIA-5> toad * r10571 /trunk/freenet/src/freenet/client/async/BackgroundBlockEncoder.java: Bugfix to background block encoding.
[17:08] <_ph00> stupid ppl live longer :P
[17:08] * sanity (n=ian@) has joined #freenet
[17:08] * ChanServ sets mode +o sanity
[17:08] * sanity (n=ian@) Quit (Remote closed the connection)
[17:08] <ljn1981> Hmm
[17:09] <CIA-5> toad * r10572 /trunk/freenet/src/freenet/node/NodeClientCore.java: Run background encode at low priority.
[17:14] * Urs_ShPo (n=gaim@) Quit ("Leaving.")
[17:17] <ljn1981> toad_: How does PROBEALL work?
[17:23] * sanity (n=ian@) has joined #freenet
[17:23] * ChanServ sets mode +o sanity
[17:28] * timmy2chk (n=Vincent@) Quit (Read error: 145 (Connection timed out))
[17:37] <toad_> aha
[17:37] <toad_> AHAAAAAA
[17:37] <toad_> it turns out that the node was automatically restarting all finished requests on startup
[17:39] <sandos_> oops?
[17:39] <sandos_> :)
[17:41] <toad_> well, the plus side is, if they failed, they'll be retried
[17:41] <toad_> which means that my little mess around 980 didn't actually kill everyone's uploads
[17:51] <toad_> ljn1981: it uses a special kind of request
[17:51] <toad_> ljn1981: location probe requests
[17:53] <Jflesch> toad_: I think it may interrest you : http://jflesch.kwain.net/~jflesch/freenet-fcp-crash.log << I run freenet on an Amd64, so I guess FEC exception are normal ? but if you look at the log carefully, you will see FCP expections, and these exceptions disconnect Thaw
[17:53] <ljn1981> How do those work?
[17:53] <toad_> ok 1sec
[17:53] <Jflesch> hm, in fact they are all FCP exceptions ...
[17:53] <Bombe> Jflesch, why aren't we connected?
[17:53] <Jflesch> Bombe: it's not the same node
[17:54] <toad_> ljn1981: purpose of a probe request is to go from location X to the next location in the keyspace greater than X
[17:54] <Bombe> That explains quite a lot. :)
[17:54] <Jflesch> Bombe: http://jflesch.homelinux.net/~jflesch/ref.txt << my new node :)
[17:54] <Jflesch> Bombe: the last one is stopped because the computer has some troubles with its hard drive
[17:54] <toad_> Jflesch: they don't kill the request because they're caught
[17:54] <Jflesch> Bombe: it should be back at the end of the year
[17:55] <Jflesch> toad_: oh ... then I have a problem on my side
[17:55] <Bombe> Jflesch, damn. I deleted the old reference too fast.
[17:55] <Bombe> Oh, well. That can be helped at the end of the year, then. :)
[17:55] <Jflesch> Bombe: not a problem, we will redo an exchange when the node will come back :)
[17:56] <Bombe> Yep. :)
[17:58] <toad_> <toad_> Jflesch: they don't kill the request because they're caught ---- i mean the exceptions
[18:02] <Jflesch> toad_: Yes, I understand, but Thaw was disconnected, so I thought it was due to these exceptions
[18:03] <Jflesch> but if it's not, then it's Thaw which has a problem
[18:06] <Jflesch> wow ! In fact the node shutdown and restart !
[18:06] <Jflesch> that's why Thaw is disconnected
[18:07] <Jflesch> INFO | jvm 4 | 2006/09/29 20:06:43 | java.lang.OutOfMemoryError: Java heap space
[18:07] <Jflesch> it explains a lot
[18:09] <TheSeeker> heh
[18:09] <ljn1981> Jflesch: My thaw is about a month old but I don't know if that matter. Anyway any idea how to make it not just sit there and requesting a finished file but never starting to write it? This happen quite a lot.
[18:10] <ljn1981> If it's relevant I have java 1.4.2 on OS X.3
[18:10] <Jflesch> ljn1981: a lot of people have already reported to me this bug
[18:10] <Jflesch> currently I don't know why it happens
[18:10] <Jflesch> but the only times where I get it, I just had to restart Thaw
[18:10] <ljn1981> Yeah I've seen it but not kept up on wether it's been fixed
[18:10] <ljn1981> Hmm
[18:11] <ljn1981> Doesn't seem to do it for me.
[18:11] <Jflesch> ljn1981: you can use http://127.0.0.1:8888/queue/ to solve it
[18:12] <Jflesch> ljn1981: what is the size of the file that you're try to fetch ?
[18:12] <ljn1981> Yeah
[18:12] <Jflesch> s/try/trying/
[18:12] <ljn1981> Nothing right now but I've had it anywhere between 200KiB and 60+MiB
[18:13] <Jflesch> hm
[18:13] <Jflesch> I must find how to reproduce it :/
[18:14] <ljn1981> Can't see any relation between the files and the bug myself.
[18:15] * devster (i=dev@) has joined #freenet
[18:15] * devster (i=dev@) has left #freenet
[18:23] * NullAcht15 (n=NullAcht@) has joined #freenet
[18:23] * ljn1981_ (i=LJN@) has joined #freenet
[18:35] * NullAcht15 (n=NullAcht@) Quit (Nick collision from services.)
[18:35] * NullAcht15_ (n=NullAcht@) has joined #freenet
[18:38] * Ebru_Labadon (i=user@) Quit (Remote closed the connection)
[18:38] * ljn1981 (i=LJN@) Quit (Remote closed the connection)
[18:38] <CIA-5> toad * r10573 /trunk/freenet/src/freenet/ (11 files in 4 dirs):
[18:38] <CIA-5> Allow the user to restart finished, failed requests from the queue page. (Get, put tested, putdir not tested).
[18:38] <CIA-5> Get: If the persistent temp-file doesn't exist, or is too short, treat it as a bucket error, rather than failing to load that request.
[18:38] <CIA-5> Don't automatically restart finished persistent requests on startup.
[18:38] <CIA-5> Minor refactoring.
[18:42] * vfervers_ (n=vfervers@) has joined #freenet
[18:46] * Ebru_Labadon (i=user@) has joined #freenet
[18:47] * Ebru_Labadon (i=user@) Quit (Client Quit)
[18:53] <toad_> anyone know a quick way to find which airports have flights to SFO?
[18:53] <toad_> which airports in the UK, that is?
[18:55] * Ebru_Labadon (i=user@) has joined #freenet
[18:56] <Ebru_Labadon> muhahaha
[18:56] <Ebru_Labadon> I got voice!
[18:56] <toad_> yay
[18:56] <toad_> what is "gpg-tor" ?
[18:58] <Ebru_Labadon> kinda cloak thingy
[18:58] <Ebru_Labadon> like a signed tor cloak
[18:58] * PraiseChaos (i=Kel@) Quit (Read error: 110 (Connection timed out))
[18:58] <Ebru_Labadon> you say, if it's signed, wht's the point with using tor?
[18:59] <Ebru_Labadon> but the point is
[18:59] <Ebru_Labadon> you don't need to use your regular signature, you can generate a new keypair for that, no need to link that to a real person
[18:59] <toad_> signed how?
[18:59] <toad_> with your email key?
[18:59] <Ebru_Labadon> gpg
[18:59] <Ebru_Labadon> with a new one
[18:59] <Ebru_Labadon> specifically generate
[18:59] <Ebru_Labadon> d
[18:59] <Ebru_Labadon> but your regular key would work
[19:00] <Ebru_Labadon> you don't want to use it because that would screw your anonymity
[19:00] <Ebru_Labadon> but it should work
[19:01] <_ph00> ]everybody] is getting tor cloaks but me :(
[19:02] <_ph00> on the other hand, I don't need to use tor anyomre, as *my* anonymity is already gone
[19:02] <ljn1981_> # networkSizeEstimateSession: 277 nodes
[19:02] <ljn1981_> # avrConnPeersPerNode: 7.861538 peers
[19:02] <ljn1981_> # nodeUptime: 2h57m
[19:03] <toad_> anyone using 10573? does it look okay?
[19:06] <CIA-5> toad * r10574 /trunk/freenet/src/freenet/node/fcp/ClientPut.java: imports
[19:07] <CIA-5> toad * r10575 /trunk/freenet/src/freenet/node/fcp/ClientPut.java: rename inserter -> putter
[19:12] * MikeW (i=Mike@) has joined #freenet
[19:12] <Ebru_Labadon> 502 here
[19:17] <CIA-5> toad * r10576 /trunk/freenet/src/freenet/node/Version.java: (log message trimmed)
[19:17] <CIA-5> 983:
[19:17] <CIA-5> - CSS: Fix url() - CSS backgrounds, includes etc were broken by last build
[19:17] <CIA-5> - Swapping stats fix; round some numbers; clean up stats code
[19:17] <CIA-5> - Ignore a / at the end of the URI when inserting a directory over FCP
[19:17] <CIA-5> - Once we pass 66% of an insert, encode all the remaining blocks so we can make the metadata, insert it, and give the user the key. => URIGenerated and PutFetchable should be soon after we reach 2/3rds of an insert.
[19:17] <CIA-5> - Restart button for failed requests (not tested for PUTDIR).
[19:22] <ljn1981_> I'm curious. What do you guys who are running r10560 or later got in avrConnPeersPerNode?
[19:22] <ljn1981_> I got 7.5
[19:23] <toad_> hmmm
[19:23] <toad_> will be interesting to see whether the new code really works when inserting 983
[19:23] <toad_> it should show generateduri at about succeeded=43
[19:23] <toad_> and then it should show putfetchable right afterwards
[19:24] <toad_> it's moving remarkably quickly, has insert speed improved dramatically for you?
[19:24] <toad_> maybe i'm looking at the testnet insert...
[19:25] * Zothar (n=Zothar@) has joined #freenet
[19:28] <ljn1981_> Hey
[19:31] <_ph00> halloy
[19:32] <_ph00> I'm blue-dah-booh-dee-dah-mmm dah-booh-dah-boodeeee-daaaa
[19:32] <_ph00> fun song
[19:33] <_ph00> I don't even see AvrConnPeersNode
[19:33] <_ph00> (where should I look?)
[19:35] <ljn1981_> Statistics page
[19:35] <ljn1981_> And you need to have advanced darknet enabled i the config
[19:35] * tester (i=tester@) has joined #freenet
[19:35] <ljn1981_> And i hate that song.
[19:35] <_ph00> (duh)
[19:35] <_ph00> and that song IS fun
[19:35] <tester> hiya
[19:35] <tester> anyone using frost with 0.7?
[19:35] <_ph00> but I undertand that it can get annoyng if you hear every day like that summer... couple of years ago
[19:36] <_ph00> I hated it then
[19:36] <_ph00> basically all of us use frost on .7
[19:36] <tester> i dont know what to select, 0.7 darknet or 0.7 testnet?
[19:36] <_ph00> darknet
[19:36] <tester> thank ph00
[19:36] <tester> *thanks
[19:36] <ljn1981_> darknet unless you selected testnet when you installed the node
[19:37] <_ph00> # networkSizeEstimateSession: 18 nodes
[19:37] <_ph00> # avrConnPeersPerNode: 3.2 peers
[19:37] <_ph00> # nodeUptime: 6m20s
[19:37] <ljn1981_> And if you did that you would't be able to peer with most people and it'd not be anonymous at all
[19:37] <_ph00> (low uptime: checking again later)
[19:37] <_ph00> I think you can set testnet off even after haning installed
[19:38] <TheSeeker> hmm
[19:38] <TheSeeker> Warning : Macro "INBRACKET" has been declared but never used.
[19:38] <TheSeeker> I get that when generating the csstokenizerfilter
[19:39] <tester> hmm.. Frost seams to be much faster then it used to be with 0.5
[19:39] <_ph00> it is
[19:39] <_ph00> the whole thing is way faster
[19:39] <_ph00> now we need more content
[19:39] <_ph00> so plz, upload files, build freesites, start flogs etc
[19:40] <_ph00> (I'm too lazy for the last 2, but I'm trying to insert some files)
[19:40] <TheSeeker> I don't know if I'd say that inserts are any faster...
[19:40] <_ph00> inserts are as slow
[19:40] <_ph00> maybe even slower
[19:41] <_ph00> but opening freesites, download etc are faster
[19:41] <_ph00> and frost is *way* faster
[19:42] <_ph00> and major improvements are also the "warmup time" that was ages for .5 is now almost noting, tyhe fact that freent now works even for those users who can't get inbound connections...
[19:43] <ljn1981_> Than 0.5 or than earlier on 0.7? toad_ asked about insert speed a bit ago.
[19:45] <toad_> well for some of them anyway
[19:46] <TheSeeker> my bandwidth usage *does* appear to be up in the latest build.
[19:46] <toad_> what about "than earlier on 0.7" ?
[19:47] <_ph00> yeah. mine too. I jsut upgaded to 983 and it looks like it's using more bandwidth.
[19:47] <toad_> 983 in particular?
[19:47] <toad_> that's probably just a startup blip...
[19:47] <_ph00> Freenet 0.7 Build #983 r10576
[19:47] <agsarite> my node seems to be taking forever to restart. its doing some mad disk reading.
[19:47] <toad_> but have recent builds been using more bandwidth in general?
[19:48] <ljn1981_> agsarite: Large queue?
[19:48] <_ph00> hm... probably not as good as I thought. one file on bittorrent started seeding, that was the incease on band usage
[19:48] <agsarite> i don't have anything queued.
[19:48] <toad_> agsarite: so at what point is it stopping?
[19:49] <ljn1981_> toad_: I've actually seen a drop over the last week but it might just be because most my downloads finished in the first half of that.
[19:49] <TheSeeker> I haven't been up for very long yet, so it might settle back into crudland, but at the moment, I'm using nearly 30K/s average (last 1:40) with the low side being 20K and the peaks at around 40. before I'd wallow around at 6-10K/s with a couple peaks to > 20 each loop of the perfmon graph ...
[19:49] <toad_> hmmm we'll see
[19:49] <agsarite> its been 10 minutes since i restarted it, my hd light has been on the hole time, and it can't access my node?
[19:49] <sandos_> K what? bytes or bits? ;)
[19:49] <ljn1981_> _ph00: There are stats for bandwidth usage and average speed on the darknet and statistics page too :P
[19:50] <sandos_> <2KBytes/s here.. :(
[19:50] <agsarite> i don't think its officially stopped or anything.
[19:50] <TheSeeker> bytes
[19:50] <_ph00> # Output Rate: 7.20 KiBps
[19:50] <_ph00> # Input Rate: 7.74 KiBps
[19:50] <_ph00> not so great...
[19:51] <sandos_> way higher than me ;)
[19:51] <_ph00> considering that my connection can do 700KB/sec
[19:51] <_ph00> KByte
[19:51] <Ebru_Labadon> I'm trying to insert a 1.5MB file, started about half an hour ago, it's currently at 45%. Looks like a fairly good speed to me
[19:51] <agsarite> whatever it was doing, it seems to be done now, and talking to me again.
[19:51] <Ebru_Labadon> I'm using the queue page in fproxy
[19:51] <Ebru_Labadon> mmm maybe it was more than half an hour ago. say 45 min.
[19:52] <Ebru_Labadon> woah. now *that* is a lag
[19:52] <Ebru_Labadon> ping
[19:52] <tester> its nice.. i like it more then 0.5
[19:52] <tester> hmm.. is fuqid still the best tool for dl?
[19:52] <tester> blah, sites are still slow to open
[19:52] <tester> still cant open any site
[19:52] * tester (i=tester@) Quit (Success)
[19:52] <Ebru_Labadon> another typical windoze user...
[19:53] * Ash-Fox (i=UNKNOWN@) Quit ()
[19:53] <ljn1981_> Hehe
[19:53] <agsarite> thanks for fixing the css, toad. just one last thing: it changed the token 'group_cell' into 'group/* ignored unmatched char: _ */cell'
[19:53] <Ebru_Labadon> open! fast! open! open! open! oh, scew it (quit)
[19:54] <sandos_> Ebru_Labadon, haha yeah
[19:54] * Ash-Fox (i=UNKNOWN@) has joined #FreeNET
[19:54] <sandos_> all in <1 minute
[19:54] <sandos_> short attention span and no patience...
[19:54] <Ebru_Labadon> yup
[19:54] <sandos_> not a good combination with freene ;)
[19:54] <sandos_> +t
[19:54] * greycat (i=rfc1413@) Quit ("This time the bullet cold rocked ya / A yellow ribbon instead of a swastika")
[19:54] <TheSeeker> people need to stop linking their usk sites with negative numbers... :P USK@coinCoin/site/-1/ takes FOREVER to load. USK@coinCoin/site/1/ will load very quikly. if I understand it correctly, the only difference is that without the - you're not guaranteed to get the latest edition...
[19:54] <TheSeeker> but with most freesite authors putting edition links on thir site anyway, I don't see this as much of an issue...
[19:54] <Ebru_Labadon> "wtf?!? I got a 400MhZ, 2GB RAM machine, and the fucker take *time*!?!
[19:55] <Ebru_Labadon> 4000
[19:56] <ljn1981_> Heehe
[19:56] <TheSeeker> the number of cycles your computer sits around waiting for work to do has pretty much nothing to do with the performance of freenet :P
[19:56] <ljn1981_> TheSeeker: coinCoin?
[19:56] * tooom (n=toom@) Quit ("reboot")
[19:57] <ljn1981_> TheSeeker: The average user do not think about it as a network
[19:57] <TheSeeker> I saw toad_ use that as oppsed to SSK@reallylongrandomblahwhatever/site so decided to use it myself :D
[19:58] <ljn1981_> Of course some of us geeks have that problem too but we can at least see it's logical that it's slow when the speed remind us it's not local
[20:00] <_ph00> how was the line to load a plugin at startup? @/path/tp/plugin?
[20:00] <TheSeeker> input and output rates listed on that stats/darknet page do not come close to resembling the rates I'm seeing with perfmon ... but, oh yeah, I'm also runnign testnet, so I should probably shut that down x_x
[20:01] <Ebru_Labadon> <TheSeeker> the number of cycles your computer sits around waiting for work to do has pretty much nothing to do with the performance of freenet <=== I know. I was making fun of 6-pack Windoze Joe
[20:02] <Ebru_Labadon> it was @<somethng else>/path/to/plugin
[20:03] <ljn1981_> Ain't this in the wiki?
[20:04] <_ph00> should be
[20:06] <_ph00> Milingo RULES
[20:06] <_ph00> (OT)
[20:06] <_ph00> I'm posting than oon -chat
[20:09] <agsarite> if i try to get something with a ssk/usk private key, will the node automatically correct it and fetch the data anyway?
[20:09] <agsarite> and if not, can we make it a feature?
[20:10] <toad_> TheSeeker: long-form freemail addresses don't even work with my mutt
[20:12] <toad_> agsarite: yeah, i know about the underscore problem
[20:12] <toad_> agsarite: i haven't verified whether they are allowed yet
[20:12] <toad_> i mean by the spec
[20:14] <agsarite> i'm almost sure it can start with/contain and underscore, contain but not start with a dash, and contain any non-ascii character. not that my word counts for anything.
[20:16] <_ph00> I can't fin it
[20:16] <_ph00> d
[20:16] <_ph00> (the line to load plugins)
[20:16] <agsarite> ident [-]?{nmstart}{nmchar}*
[20:16] <agsarite> nmchar [_a-zA-Z0-9-]|{nonascii}|{escape}
[20:16] <agsarite> nmstart [_a-zA-Z]|{nonascii}|{escape}
[20:17] * vfervers_ (n=vfervers@) Quit (Remote closed the connection)
[20:17] <agsarite> from http://www.w3.org/TR/CSS21/syndata.html, is that what we're looking for?
[20:19] <_ph00> I got some more bandwidth usage
[20:20] <_ph00> # Output Rate: 15.0 KiBps
[20:20] <_ph00> # Input Rate: 10.4 KiBps
[20:20] <_ph00> setting it to 400K!
[20:20] <_ph00> 400 up and 400 down
[20:21] <_ph00> amybe I should set it to 1000K to get some 30-40K
[20:21] * tooom (n=toom@) has joined #freenet
[20:21] <agsarite> i guess they are allowed to start with dashes.
[20:22] <agsarite> "Keywords and property names, beginning with -' or '_' are reserved for vendor-specific extensions"
[20:23] <toad_> TheSeeker: maybe what we need to do is prefetch the /-1/ links?
[20:24] <TheSeeker> toad_: is that a wise idea? doing automatic prefetching of content on a page?
[20:25] <toad_> agsarite: bad feature imho
[20:25] <toad_> agsarite: if the node fetches it from the private key the user won't realize it's the private key, and will post his private key somewhere he shouldnt
[20:26] <agsarite> :(
[20:26] <agsarite> it would have made inserting/requesting ssks exactly the same as inserting/requesting ksks.
[20:26] <agsarite> very well then.
[20:27] <toad_> agsarite: the BNF says it can start with a dash?
[20:27] <toad_> agsarite: okay, so we DON'T want it to start with a - or _
[20:27] <toad_> because we don't know what it might do
[20:27] <agsarite> i guess that's a fair assumption.
[20:29] <agsarite> i'll keep just the private key anyway, and thwart your attempt at safety by having my fcp library convert it my self before it sends it to the node. take that!
[20:30] <toad_> lol
[20:30] * MikeW (i=Mike@) Quit (Read error: 104 (Connection reset by peer))
[20:31] * Jflesch_ (n=jflesch@) has joined #freenet
[20:33] * Jflesch (n=jflesch@) Quit (Read error: 145 (Connection timed out))
[20:35] <CIA-5> toad * r10577 /trunk/freenet/src/freenet/ (5 files in 2 dirs):
[20:35] <CIA-5> Fix URIGenerated / PutFetchable. Bugs in client/async/ were preventing these from being sent except before completion, even if the blocks *were* encoded.
[20:35] <CIA-5> Fix background block encoder.
[20:35] <CIA-5> Logging.
[20:35] <ljn1981_> Hhe
[20:35] <ljn1981_> HOW DARE YOU TRY TO MAKE THINGS SAFE FOR ME!! BASTARD!!!!
[20:35] <ljn1981_> Hehe
[20:37] <ljn1981_> toad_: Maybe add a comment why we don't want that CSS stuff?
[20:38] <toad_> okay, lets fix the underscore thingy...
[20:38] * blibbet (n=blibbet_@) has joined #freenet
[20:42] <toad_> agsarite: have you got an example for me of a site which has identifiers in its css with underscores in?
[20:42] <agsarite> sec
[20:43] <agsarite> freenet:SSK@c55vMxUl-T-lD3nv0iOaXF~G1hnY6pOMRbzZSwACMmY,yd8~uwUmGm164-ipStoiBOJVjkbbYXJMlD~H5ftPxIA,AQABAAE/Vapor-2/shape.css
[20:43] <CIA-5> toad * r10578 /trunk/freenet/src/freenet/client/async/BackgroundBlockEncoder.java: Fix NPE
[20:43] <agsarite> its about in the middle, search for div.group
[20:47] <toad_> <pre>div.group *.group_cell {
[20:47] <toad_> display: table-cell; }</pre>
[20:47] <agsarite> that be it.
[20:49] <toad_> hmmm
[20:49] <toad_> : / * etc are passed through as single characters
[20:50] <toad_> does that mean you can get http://www.google.com/ through the filter?
[20:50] <toad_> a url ...
[20:50] <toad_> under what circumstances does CSS treat an unquoted url as a url?
[20:50] <toad_> does it always have to be url()'ed or at least quoted?
[20:51] <agsarite> i think so, but browsers might be more lax.
[20:52] <toad_> hmmm
[20:52] <toad_> it looks like you have to url() it
[20:52] <toad_> so it will be caught
[20:52] <toad_> ok
[20:54] <toad_> what about with @import ?
[20:54] <agsarite> i'm pretty sure i don't know the answer to that question.
[20:55] <toad_> hmmm
[20:55] <toad_> i think we're okay
[20:56] <agsarite> see, this is an example where it would be nice to run things through the filter without needing to upload them first ;)
[20:56] <toad_> hmmm
[20:56] <toad_> @import "uri"
[20:57] <toad_> but we catch that explicitly
[20:57] <toad_> and i think that's the only one
[20:59] <ljn1981_> # networkSizeEstimateSession: 320 nodes
[20:59] <ljn1981_> # avrConnPeersPerNode: 7.717391 peers
[20:59] <ljn1981_> # nodeUptime: 4h57m
[20:59] <ljn1981_> # swaps: 3
[20:59] <ljn1981_> # noSwaps: 227
[20:59] <ljn1981_> Hmm
[21:01] <ljn1981_> _ph00: How about you?
[21:05] * hdp (n=hdp@) Quit (Remote closed the connection)
[21:05] <_ph00> what
[21:05] <_ph00> stats?
[21:05] <_ph00> wait
[21:06] <CIA-5> toad * r10579 /trunk/freenet/src/freenet/clients/http/filter/ (CSSTokenizerFilter.java CSSTokenizerFilter.jflex):
[21:06] <CIA-5> Allow _'s in identifiers (but not the first char).
[21:06] <CIA-5> Explain in comments that we don't want to allow vendor specific extensions (except maybe one at a time).
[21:06] <CIA-5> Comment out INBRACKET.
[21:06] <CIA-5> Re-enable the unknown @{IDENT} but only chop the ident itself, not the whole line.
[21:06] <CIA-5> (Ideally we'd match the whole line but then we'd have to do so for all the ident's we support).
[21:06] <_ph00> # bwlimitDelayTime: 0ms
[21:06] <_ph00> # nodeAveragePingTime: 303ms
[21:06] <_ph00> # networkSizeEstimateSession: 243 nodes
[21:06] <_ph00> # avrConnPeersPerNode: 7.202247 peers
[21:07] <_ph00> # nodeUptime: 1h35m
[21:07] <_ph00> # missRoutingDistance: 0.0377
[21:07] <_ph00> # backedoffPercent: 42.5%
[21:07] <_ph00> # pInstantReject: 0.0%
[21:07] <_ph00> # swapsPerMinute: 0.010469
[21:07] <_ph00> # noSwapsPerMinute: 0.921305
[21:07] <_ph00> # swapsPerNoSwaps: 0.011364
[21:07] <_ph00> # Output Rate: 11.8 KiBps
[21:07] <_ph00> # Input Rate: 13.8 KiBps
[21:08] <CIA-5> toad * r10580 /trunk/freenet/src/freenet/clients/http/filter/CSSTokenizerFilter.jflex: comments
[21:08] <toad_> _ph00: that's quite a long paste, in future please use pastebins for such matters
[21:09] <_ph00> toad_ k
[21:10] <CIA-5> toad * r10581 /trunk/freenet/src/freenet/node/Version.java:
[21:10] <CIA-5> 984:
[21:10] <CIA-5> More CSS fixes.
[21:10] <CIA-5> Fix URIGenerated/PutFetchable; they really will be sent as soon as possible now. (Bugs in client/async/ were preventing 983's changes from working).
[21:11] * hdp (n=hdp@) has joined #freenet
[21:12] <TheSeeker> whee, time to update again then :D
[21:13] <_ph00> can someone plz tell me what the plugin manager line is? I've been looking for it in the wiki but I can't find it. I found *@/path/to/plugin but that's not working
[21:16] <toad_> bbiab
[21:19] <sleon> fn
[21:19] <_ph00> how do I make STUN load at every freenet start?
[21:21] <Ebru_Labadon> My 1.5MB file is at 94.7% in something like 3 hrs
[21:21] <Ebru_Labadon> maybe 4
[21:34] * phrosty (n=phrosty@) has joined #freenet
[21:40] <_ph00> ljn1981_ your node is disconnected from mine, is it up?
[21:41] <ljn1981_> No poking at code. The node and eclipse running at the same time is making me want to throw my computer out the window.
[21:42] <_ph00> throw you tvset intead :P
[21:42] <_ph00> s
[21:43] <ljn1981_> I don't have one
[21:43] <_ph00> good
[21:43] <ljn1981_> VCR+wall projector
[21:43] <_ph00> 6 backed off?!?
[21:44] <_ph00> wtf is wrong with 984?
[21:44] <_ph00> vcr's are still around?
[21:44] * FooDerGrosse (n=kevin@) Quit ("W??hlt die PiratenPartei Deutschland!!!")
[21:44] <ljn1981_> Not that I use it much for TV, the cable hasn't been plugged in for almost a year
[21:45] <_ph00> I get mad at those whore journalists each time I see the news on TV (like a couple of times a year)
[21:46] <ljn1981_> I think the VCR i have is from the early 90s. I only got it because it was free, it took less room than a TV and I could use it with the projector to watch TV
[21:47] <blibbet> kill your television!
[21:47] <_ph00> yeah
[21:47] <_ph00> I'm trying to get as many ppl as psooible to quit TV
[21:48] <_ph00> I say "you can get all your new, movies and tvshows form the internet, skip the commercials and most important skip the brainwashing"
[21:48] <_ph00> possible*
[21:49] <_ph00> because I noticed that those who watch tv tent to have more "standardized" opinions, compared to those who don't
[21:49] <_ph00> like that scary thing "I don't care if they spy on me because I don't have anything to hide". tv watching ppl say that. Internet ppl don't
[21:50] * blibbet (n=b