Timestamps are in GMT/BST.
[0:05] * louisb (n=lou@CPE0013f70131cb-CM0012c91006bc.cpe.net.cable.rogers.com) Quit (Read error: 110 (Connection timed out))
[0:10] * Ash-Fox (i=UNKNOWN@edj117.neoplus.adsl.tpnet.pl) Quit (Nick collision from services.)
[0:11] * Ash-Fox (i=UNKNOWN@eco138.neoplus.adsl.tpnet.pl) has joined #FreeNET
[0:20] * Hory (n=Miranda@82.78.27.85) Quit ("http://CyberLore.net - Recommendations on the best games, freeware and websites.")
[0:39] * sleon|tuX (i=test@e180060224.adsl.alicedsl.de) has joined #freenet
[1:04] * nextime (n=nextime@213-140-6-96.ip.fastwebnet.it) Quit ("nextime has no reason")
[1:25] * nextime (n=nextime@213-140-6-96.ip.fastwebnet.it) has joined #freenet
[1:33] * Ash-Fox (i=UNKNOWN@eco138.neoplus.adsl.tpnet.pl) Quit (Remote closed the connection)
[1:33] * Ash-Fox (i=UNKNOWN@eco138.neoplus.adsl.tpnet.pl) has joined #FreeNET
[2:13] * sanity (n=ian@81-178-94-218.dsl.pipex.com) Quit ()
[2:21] * louisb (n=lou@CPE0013f70131cb-CM0012c91006bc.cpe.net.cable.rogers.com) has joined #freenet
[2:27] * TheBishop_ (n=bishop@port-212-202-175-197.dynamic.qsc.de) has joined #freenet
[2:45] * gathers (n=gathers@c-9746e255.434-1-64736c11.cust.bredbandsbolaget.se) Quit ("L??mnar")
[3:19] * louisb (n=lou@CPE0013f70131cb-CM0012c91006bc.cpe.net.cable.rogers.com) Quit (Read error: 110 (Connection timed out))
[3:30] * sleon|tuX (i=test@e180060224.adsl.alicedsl.de) Quit ("Ex-Chat")
[3:49] * sanity (n=ian@81-178-94-218.dsl.pipex.com) has joined #freenet
[4:09] * MrNaughty (n=mrnaught@S0106001125372935.ed.shawcable.net) Quit ("\(^_^)/'")
[4:21] * MrNaughty (n=mrnaught@S0106001125372935.ed.shawcable.net) has joined #freenet
[4:59] * TheSeeker (i=Fridlekh@67-40-225-97.phnx.qwest.net) has joined #freenet
[5:53] * unknown_ (n=NullAcht@dslb-082-083-229-243.pools.arcor-ip.net) Quit (Remote closed the connection)
[7:15] * TheSeeker (i=Fridlekh@67-40-225-97.phnx.qwest.net) Quit (Read error: 110 (Connection timed out))
[7:53] <linyos> lalalaal
[8:54] * TheSeeker (i=Fridlekh@67-40-225-97.phnx.qwest.net) has joined #freenet
[9:14] * rara (n=zeboul@pha75-2-81-57-114-186.fbx.proxad.net) has joined #freenet
[9:14] <rara> hello
[9:38] * rara (n=zeboul@pha75-2-81-57-114-186.fbx.proxad.net) Quit ("Leaving")
[10:18] * nextgens_ is now known as nextgens
[10:54] * sbc (n=sbc@83.72.201.52.ip.tele2adsl.dk) has joined #freenet
[12:12] * JoshParadroid (n=josh@paradroid.com) Quit (Remote closed the connection)
[12:12] * JoshParadroid (n=josh@paradroid.com) has joined #freenet
[12:41] * Ash-Fox (i=UNKNOWN@eco138.neoplus.adsl.tpnet.pl) Quit ()
[12:46] * sleon|tuX (n=sleon|tu@e180060224.adsl.alicedsl.de) has joined #freenet
[12:46] * rara (n=zeboul@pha75-2-81-57-114-186.fbx.proxad.net) has joined #freenet
[12:46] <rara> hello
[12:48] * Ash-Fox (i=UNKNOWN@eco138.neoplus.adsl.tpnet.pl) has joined #FreeNET
[12:58] * Ash-Fox (i=UNKNOWN@eco138.neoplus.adsl.tpnet.pl) Quit ()
[13:00] * verl (i=verl@h56n2fls31o877.telia.com) Quit (Read error: 104 (Connection reset by peer))
[13:34] * Ash-Fox (i=UNKNOWN@eco138.neoplus.adsl.tpnet.pl) has joined #FreeNET
[14:18] * rara (n=zeboul@pha75-2-81-57-114-186.fbx.proxad.net) Quit ("Leaving")
[14:59] * NullAcht15 (n=NullAcht@dslb-082-083-221-026.pools.arcor-ip.net) has joined #freenet
[15:13] <toad_> hrllo
[15:14] <hobx_> g
[15:14] * louisb (n=lou@CPE0013f70131cb-CM0012c91006bc.cpe.net.cable.rogers.com) has joined #freenet
[15:21] <toad_> hi louisb
[15:21] <toad_> |
[15:21] <toad_> yay
[15:21] <toad_> not too bad
[15:22] <hobx_> tongues?
[15:23] <toad_> hmm?
[15:26] <hobx_> was wondering if you were speaking in tongues
[16:07] <toad_> hobx_: eh?
[16:09] <hobx_> never mind
[16:19] * NullAcht15 (n=NullAcht@dslb-082-083-221-026.pools.arcor-ip.net) Quit (Read error: 113 (No route to host))
[16:22] * NullAcht15 (n=NullAcht@dslb-082-083-221-026.pools.arcor-ip.net) has joined #freenet
[16:24] * rara (n=zeboul@pha75-2-81-57-114-186.fbx.proxad.net) has joined #freenet
[16:25] <rara> re
[16:25] <rara> hello
[16:33] * louisb (n=lou@CPE0013f70131cb-CM0012c91006bc.cpe.net.cable.rogers.com) Quit (Read error: 110 (Connection timed out))
[16:53] * Rom|Bed (n=romster@tor/session/x-9a4aa44954b5fda8) has joined #freenet
[16:53] * Romster (n=romster@tor/session/x-cd09f48e17d1e838) Quit (Nick collision from services.)
[16:53] * Rom|Bed is now known as Romster
[17:21] * Hory (n=Miranda@82.78.27.85) has joined #FreeNet
[17:28] * Hory (n=Miranda@82.78.27.85) Quit (Read error: 54 (Connection reset by peer))
[17:32] * Hory (n=Miranda@82.78.27.85) has joined #FreeNet
[17:38] * rara (n=zeboul@pha75-2-81-57-114-186.fbx.proxad.net) Quit ("Leaving")
[17:41] * Hory (n=Miranda@82.78.27.85) Quit ("http://CyberLore.net - Recommendations on the best games, freeware and websites.")
[18:08] <toad_> hi folks
[18:08] <Ash-Fox> Hey
[18:14] <JoshParadroid> so, what's this i hear about URIs?
[18:14] <toad_> hmmm?
[18:15] <JoshParadroid> USKs even!
[18:15] <toad_> ahhh
[18:15] <toad_> that's my proposed new backend-independant updatable keytype
[18:16] <toad_> it could be implemented using TUKs, DBRs or just plain editions, but the mechanism is controlled by the node, not the user
[18:16] <toad_> so the user doesn't need to worry (much) about it
[18:16] <JoshParadroid> good idea
[18:16] <toad_> well, ian said not before 0.7; i will take it up with him again after i have the download manager done, as i think there's room for some major integration there
[18:16] <toad_> i'm not entirely sure how much work it is
[18:17] <toad_> it's likely to be more work than DBRs
[18:17] <JoshParadroid> is there any kind of good primer on TUKs around?
[18:17] <toad_> but it'd be nice to not have editions and DBRs on 0.7.0's release
[18:17] <toad_> JoshParadroid: mostly it's on the mailing list
[18:18] <toad_> basically... editions are SSK@key,entropy/sitename/number//<site>
[18:18] <toad_> DBRs are SK
[18:18] <toad_> DBRs are SSK@key,entropy/sitename-<timestamp>//
[18:18] <toad_> (internally)
[18:18] <toad_> TUKs are a sort of versioned SSK
[18:18] <JoshParadroid> ah
[18:18] <toad_> TUKs make all the editions go through the same key
[18:18] <toad_> which is bad from some points of view
[18:19] <toad_> TUKs are pretty easy to implement, but we don't know for sure that they will work, and we know there will be some issues with it
[18:19] <toad_> so with all this complexity, i figured we ought to centralize control over updating into the node
[18:19] <toad_> hence
[18:19] <toad_> USKs
[18:19] <hobx_> TUK TUK
[18:19] <hobx_> what is a TUK?
[18:19] <hobx_> wtf! :-)
[18:20] <toad_> USK@key,entropy/sitename/number//
[18:20] <toad_> which means "fetch the site, edition at least <number>, but more recent if you have it"
[18:20] <toad_> if you want a specific edition you do SSK@key,entropy/sitename/number//
[18:21] <toad_> oh and it sends a HTTP permanent-redirect if it gives you a more recent one
[18:21] <toad_> so your URL changes to reflect the new link
[18:21] <toad_> so if you give it to somebody else, they have a more recent version
[18:21] <toad_> and obviously if it is implemented by editions it would have to do background fetching of later versions
[18:21] <toad_> hobx_: TUK = versioned SSK
[18:21] <toad_> multiple versioned SSKs can exist with the same key
[18:22] <toad_> you can fetch by exact version, you can fetch the latest, or you can fetch all of them
[18:22] <toad_> they are LRU'd as individual files
[18:22] <hobx_> good luck with that.
[18:22] <toad_> there are some issues with load/caching/selective request propagation
[18:22] <Ash-Fox> Interesting.
[18:23] <toad_> <hobx_> good luck with that. --- is that good or bad? sarcasm across regional frontiers often doesn't parse well...
[18:23] <toad_> even bristol to london!
[18:23] <toad_> let alone to sweden
[18:24] * toad_ used to think "yeah, right" meant "no" ... :)
[18:24] <hobx_> bad. the last thing you need is more complexity.
[18:24] <hobx_> yeah, right can mean no.
[18:24] <toad_> yes but some londoners use it to mean "yeah, right"
[18:24] <hobx_> it can mean that too.
[18:24] <hobx_> Sarcasm is pointless if it becomes formulaic.
[18:25] <toad_> hobx_: there are bottleneck issues with it anyway; generally speaking it's a bad thing to channel a lot through a single key
[18:25] <toad_> you could implement streams with just key sequences (and passive requests)
[18:25] <hobx_> Ian didn't hit the roof about this?
[18:25] <JoshParadroid> that reminds me of the other thing to ask! how does this subscribing thing move data around?
[18:25] <toad_> it wouldn't be quite as efficient as pub/sub, but there are lots of problems it wouldn't suffer from
[18:26] <toad_> JoshParadroid: we gave up on the subscribe thing, for now
[18:26] <toad_> we probably should update the web site about it
[18:26] <JoshParadroid> ah, ok
[18:26] <toad_> subscribe gels with passive requests and TUKs, and we don't want to implement those yet
[18:26] <toad_> we'll probably end up implementing streams over passive requests, maybe with some variant on TUKs thrown in
[18:27] <JoshParadroid> what's a passive request?
[18:27] <toad_> request that persists until it is satisfied or until the node forgets about it
[18:27] <toad_> works very similarly to pub/sub, but simpler
[18:27] <toad_> anyway it's 0.8 stuff
[18:28] <JoshParadroid> ok - i think there's enough for me to learn about 0.7 first
[18:28] <toad_> hobx_: ian didn't hit the roof, but is of the view that we shouldn't do any new updating before 0.7 for time reasons
[18:29] <toad_> i think we should *probably* do USKs, but i won't pursue it until i'm convinced it wouldn't be a lot of work
[18:29] <hobx_> oki doki
[18:47] * Hory (n=Miranda@82.78.27.85) has joined #FreeNet
[18:47] * Hory (n=Miranda@82.78.27.85) Quit (Read error: 54 (Connection reset by peer))
[18:50] * Elly casts Orb of Entropy on toad_ for 3d6+24 damage and 1d6 str drain
[18:53] <toad_> ouch
[18:53] * toad_ summons Nisha
[18:54] <toad_> (which is a bear the size of an elephant with 3000 hit points and loads of damage)
[18:54] * Elly casts Nether Bolt for 11d8 damage! :D
[18:55] <toad_> :)
[18:55] <toad_> hi Elly
[18:55] <Elly> hi
[18:55] * Elly summons a Hecatonchieres (hundred-handed, fifty-headed outcasted demi-god; can Teleport Without Error at will and lots of other fun stuff)
[18:56] <Elly> whoo, 24% fail on Nether Bolt
[19:04] <toad_> Elly: game?
[19:05] * Hory (n=Miranda@82.78.27.85) has joined #FreeNet
[19:05] <toad_> hmmm
[19:05] <toad_> how do i set a property on several files at once in subversion?
[19:05] <Elly> toad_: Hengband
[19:07] * sbc (n=sbc@83.72.201.52.ip.tele2adsl.dk) Quit ("Leaving")
[19:11] * rara (n=zeboul@pha75-2-81-57-114-186.fbx.proxad.net) has joined #freenet
[19:23] * MrNaughty (n=mrnaught@S0106001125372935.ed.shawcable.net) Quit ("\(^_^)/'")
[19:25] * MrNaughty (n=mrnaught@S0106001125372935.ed.shawcable.net) has joined #freenet
[20:08] * MikeW (i=Mike@159-134-232-76.as1.cbr.castlebar.eircom.net) has joined #freenet
[20:32] * Hory` (n=Miranda@82.78.27.85) has joined #FreeNet
[20:33] * nextgens (n=nextgens@ADijon-151-1-28-147.w83-194.abo.wanadoo.fr) Quit (""ya mum quit IRC :'("")
[20:44] * Hory (n=Miranda@82.78.27.85) Quit (Connection timed out)
[21:10] * nextgens (n=nextgens@ADijon-151-1-28-147.w83-194.abo.wanadoo.fr) has joined #freenet
[21:29] <TheSeeker> 'morning
[21:37] <toad_> hi TheSeeker
[22:02] * Hory` (n=Miranda@82.78.27.85) Quit ("http://CyberLore.net - Recommendations on the best games, freeware and websites.")
[22:26] * gathers (n=gathers@c-d442e255.434-1-64736c11.cust.bredbandsbolaget.se) has joined #freenet
[22:33] * MHq (n=whocarez@81-208-106-73.ip.fastwebnet.it) has joined #freenet
[22:34] * danderson (n=dave@natulte.net) has left #freenet
[22:34] <TheSeeker> Anything new with 0.7 testing?
[22:40] <TheSeeker> On a side note, will the 0.7 datastore be mostly one large file (virtual disk) ? ... Having 'only' 6 gigs of datastore, if all the chunks are 32k each, that's 192000 files
[22:41] <TheSeeker> which is 14.5 times as many files as I have in my datastore now...
[22:41] <toad_> indeed
[22:42] <toad_> since the files are fixed size, monolithic datastore is pathetically easy
[22:42] <TheSeeker> yey
[22:42] <toad_> it won't necessarily be superencrypted though
[22:42] <toad_> well not for the initial release anyway
[22:43] <toad_> not unless somebody volunteers to code that bit before it comes out; it's not a priority for me
[22:43] <toad_> also there is more work to be done on the store before release; it needs to have a common index for the header- and data- stores
[22:43] <toad_> and it will probably have major memory usage issues when we start getting multigigabyte stores
[22:43] <TheSeeker> fine by me, I can put it on any encrypted disk... I'm more worried about freenet killing my available hdd space with all the overhead of that many files.
[22:44] * MikeW (i=Mike@159-134-232-76.as1.cbr.castlebar.eircom.net) Quit (Read error: 104 (Connection reset by peer))
[22:44] <toad_> working out all these issues will IMHO require a long beta test...
[22:45] <TheSeeker> I'm still getting the phantom keys peoblem, where the node status page shows a lot more keys in my DS than actually exist.
[22:45] * rara (n=zeboul@pha75-2-81-57-114-186.fbx.proxad.net) Quit ("Leaving")
[22:46] <MHq> having greater datastore means that pages will load faster?
[22:46] <MHq> or is that a matter of available bw
[22:46] <TheSeeker> having greater datastore mean that less data will fall off the network.
[22:46] <MHq> ok
[22:46] <toad_> having said that... we may have issues with temporary files
[22:46] <TheSeeker> there's a slightly greater chance that you will already have the data cached in your store when you ask for it too
[22:47] <toad_> i expect the node to use quite a lot of temporary files when downloading a 2GB splitfile...
[22:48] <toad_> we might make the client-cache also provide temp space buckets, but if we do that, we have to implement the client-cache
[22:48] <toad_> also it would mean you can't download anything bigger than your client-cache
[22:48] <TheSeeker> can the node not decode each chunk upon completion and insert the data into the single file beign constructed?
[22:48] <toad_> only on a segment level
[22:49] <toad_> and yes i suppose it could write the data into the file being constructed on a segment level...
[22:49] <TheSeeker> so a 'segment' can consist of several chunks?
[22:49] <toad_> it might even be able to write the temp data, and keep a map in RAM...
[22:49] <toad_> obviously, we'd have to rename the file so that it doesn't look complete...
[22:50] <TheSeeker> most p2p programs do... something like TMP-<hash>_filename
[22:50] <TheSeeker> then when it's done, the file is renamed and moved to the destination
[22:51] <MHq> yah..like eMule
[22:51] <toad_> well
[22:51] <toad_> do we want it to be moved to the destination, or do we want it to be written there in the first place, as a temporary file?
[22:51] <toad_> both options have advantages; the former is cleaner, but requires more disk space
[22:52] <toad_> (at least it does if you have to do a copy-and-delete move)
[22:52] <toad_> (i.e. if it's not on the same drive)
[22:53] <TheSeeker> A "move" in windows is a copy/delete ... the delete is just automatic... if there's an eror during the move, the file is still OK in it's original location.
[22:53] <toad_> no, there is such a thing as a fast move even in windows
[22:53] <TheSeeker> a move on the same drive just updates the index, so is fast.
[22:53] <toad_> anyway the point is, you'd need a lot of space in the temporary location
[22:54] <TheSeeker> which is also the case for most other p2p software
[22:54] <TheSeeker> you need at least as much temp space as you ahve downloads going.
[22:54] <TheSeeker> *have
[22:54] <toad_> true, and it is also the case if we don't write the data direct
[22:54] <toad_> if we use a bunch of temp files instead
[22:55] <toad_> also there are related security issues
[22:55] <toad_> i suppose those come up anyway
[22:55] <toad_> if you have downloads which persist across restarts, you have to write the list somewhere
[22:57] <TheSeeker> As long as the security issues are limited to those where someone has direct access to the machine, it's not so bad. paranoid users should be running on encrypted disks and have easy access to a quick-kill switch for all hardware :P
[22:58] <toad_> thermite on the hard disk... or perhaps just a big AC electromagnet wrapped around it :)
[22:58] <toad_> or more realistically, you have to type in a passphrase to boot, which decrypts the encrypted drives
[22:59] <TheSeeker> I was thinking power switch... if you're running everything encrypted, nobody should be able to access the data on the disks?
[23:00] * nextgens_ (n=nextgens@ADijon-151-1-82-62.w83-196.abo.wanadoo.fr) has joined #freenet
[23:01] <toad_> hi nextgens
[23:02] <TheSeeker> If I were really paranoid, I'd set up a virtual machine (vmware, bochs, something) and run freenet on that. run the virtual OS with encrypted storage, as well as encrypt the HDD file in the underlying OS...
[23:02] <nextgens_> sorry :$
[23:02] <nextgens_> my connection is really unstable :/
[23:04] <TheSeeker> I lurve my connection. I've had to restart freenet a couple times in the last week due to flashing exclamation point of doom, but my internet is solid... have sent 28.1G and recieved 29.4G since I last rebooted.
[23:06] <toad_> woah
[23:11] * nextgens (n=nextgens@ADijon-151-1-28-147.w83-194.abo.wanadoo.fr) Quit (Read error: 110 (Connection timed out))
[23:14] <sandos> Chain FORWARD (policy ACCEPT 336M packets, 321G bytes)
[23:14] <sandos> Chain OUTPUT (policy ACCEPT 116M packets, 480G bytes)
[23:14] <sandos> :)
[23:14] <sandos> uptime
[23:14] <sandos> 23:11:35 up 4 days, 14:21, 3 users, load average: 1.89, 1.79, 1.70
[23:17] <TheSeeker> what kind of connection do you have? :x
[23:19] <sandos> 10Mbit
[23:19] <TheSeeker> synchronous?
[23:20] <sandos> yeah, its FD via fibre/ethernet
[23:20] <sandos> will be moving in 25 days to ethernet half-duplex though :(
[23:20] <TheSeeker> who'd you have to kill to get that?
[23:20] <sandos> nobody ;)
[23:20] <sandos> you have to live in Sweden.. ;)
[23:21] <TheSeeker> :(
[23:21] <TheSeeker> I'm nowhere near sweden
[23:21] <sandos> I can get 100Mbit aswell ;)
[23:21] <sandos> sadly not when I move
[23:21] <sandos> and they dont plan to upgrade the net at the new place for another 2 years...
[23:22] <sandos> though that date has been pushed up before so it probably wont happen for 3 years
[23:23] <mazzanet> toad_: does scp-ing into /home/groups/f/fr/freenet/public_html/snapshots/ get 'mirrored' across to emu?
[23:23] * sleon|tuX (n=sleon|tu@e180060224.adsl.alicedsl.de) Quit (Read error: 110 (Connection timed out))
[23:23] * sleon|tuX (i=test@e180045092.adsl.alicedsl.de) has joined #freenet
[23:25] <nextgens_> mazzanet> no
[23:25] <nextgens_> mazzanet> don't do it :)
[23:25] * nextgens_ is now known as nextgens
[23:26] <mazzanet> i see
[23:26] <nextgens> what do you want to do ? :)
[23:27] <nextgens> may I help ?
[23:27] <mazzanet> doesn't look like a copy of the webinstaller lives on emu
[23:27] <mazzanet> i don't want to do anything :P
[23:27] <nextgens> erf,
[23:27] <nextgens> yes, I can download one there ...
[23:28] * nextgens isn't using windows either
[23:30] <toad_> bbl
[23:30] <nextgens> mazzanet> is it better ?
[23:30] <nextgens> mazzanet> should I add something else ?
[23:31] <toad_> back monday
[23:31] * toad_ (i=toad@pdpc/supporter/active/toad-with-underline) Quit (Remote closed the connection)
[23:32] <mazzanet> nextgens: meh. i'll be rebuilding them sometime in the next few days anyway
[23:33] <nextgens> ok, will see then ^-^
[23:35] * sleon|tuX (i=test@e180045092.adsl.alicedsl.de) Quit ("Ex-Chat")
[23:51] <linyos> no point in encrypted files
[23:51] <linyos> block device encryption!
[23:51] <linyos> device mapper!
Irc logs of #freenet : 2008 2007 2006 2005
These logs were automatically created by FreenetLogBot on chat.freenode.net using the Java IRC LogBot.