Timestamps are in GMT/BST.
[0:00] <toad_> aha
[0:01] <Jase> heureka
[0:01] <Jase> ?
[0:04] * Caco_Patane (n=caco@) Quit ("Will the REAL Slim BitchX please stand up?")
[0:10] * Iceman_B^Ltop (n=ice@) Quit ("Swan is hotter than Jasmine ^_^ -=SysReset 2.53=-")
[0:11] <toad_> hmmm
[0:12] * Caco_Patane (n=caco@) has joined #freenet
[0:12] <toad_> methinks there's a firewall between me and sbc which blocks fragmented packets
[0:12] <toad_> my node receives auth packets phase 0 and 1 from sbc's node
[0:12] <toad_> and it sends phase 2's
[0:12] <toad_> which are approximately 1466 bytes long
[0:13] <toad_> sbc's node only ever receives phase 0's from my node
[0:14] <toad_> but it might also get 1's, and the log is just from a point during which 0's were being sent from my node
[0:14] <toad_> anyway
[0:15] <toad_> 2's and 3's are big
[0:15] <toad_> BIG
[0:15] <toad_> over the MTU for some connections
[0:15] <toad_> some bad firewalls block fragmented packets ...
[0:15] * nextgens doesn't see the point
[0:16] <nextgens> of doing so
[0:16] <toad_> nextgens: don't they?
[0:16] <toad_> i thought some did
[0:16] <toad_> certainly they block ICMP
[0:16] <nextgens> that can lead to issues
[0:16] <toad_> indeed it does
[0:16] <nextgens> if MTU is much smaller behind
[0:16] <nextgens> and someone sends an icmp fragmentation needed ... wich won't get through
[0:16] <toad_> well, the point about blocking fragmented packets is you can't do packet inspection if the packet is fragmented, unless you defrag it yourself
[0:16] <nextgens> it will get dropped
[0:16] <toad_> which requires lots of memory/cpu
[0:17] <nextgens> but that's not on purpose
[0:17] <nextgens> yes, that's something else
[0:17] <toad_> okay so do some firewalls block fragmented packets?
[0:17] <toad_> i thought i'd seen it somewhere
[0:18] <toad_> either that, or a firewall just blocks ALL packets over a certain size
[0:18] <toad_> 1466 + 28 = 1494 bytes
[0:18] <toad_> which is over the limit for PPPoE
[0:19] <Caco_Patane> date
[0:20] <Caco_Patane> (sorry)
[0:20] * electrosy (n=steo@) has left #freenet
[0:20] <nextgens> Wed Nov 29 00:20:40 UTC 2006
[0:21] <toad_> nextgens: hmm?
[0:21] <toad_> nextgens: is it a credible scenario anyway?
[0:21] <toad_> if I was designing a crappy router i'd probably block all fragmented packets
[0:21] <nextgens> well, we can't do much about it
[0:21] <toad_> nextgens: well, if we know what it is, that would be nice
[0:21] <nextgens> we exchange compressed noderefs
[0:21] <toad_> yes but there are options
[0:22] <nextgens> and allowing fragmentation of our authentication phases could probably lead to a flowed design
[0:22] <nextgens> allowing DoS attacks
[0:22] <toad_> we can shave off a few bytes when we next change the setup protocol
[0:22] <xxfooxx> frags in and of themselves are not a problem
[0:22] <toad_> it's very heavyweight at the moment
[0:22] <nextgens> toad_> agreed
[0:22] <xxfooxx> the problems happen when you're fragmenting, and sending two packets in the place of one
[0:22] <nextgens> toad_> btw, that would be a good opportunity to do proper sts :p
[0:22] <toad_> we could exchange references using a dedicated message, after having completed auth
[0:22] <xxfooxx> one fully loaded, one with like 3 bytes..
[0:23] * DebolazX (n=DebolazX@) has joined #freenet
[0:23] <xxfooxx> sucks for performance
[0:23] <nextgens> toad_> hmm, don't we use things from the reference for unauthenticated dh ?
[0:23] <toad_> well ...
[0:23] <nextgens> like dsa groups ?
[0:23] <toad_> we do use the version, yeah :|
[0:23] * phrosty (n=phrosty@) Quit ("baseball is wrong: man with four balls cannot walk.")
[0:23] <toad_> the dsa groups we are expected to already know
[0:23] <nextgens> of couse we could send a "cleaned up" ref
[0:23] <nextgens> without anything but version and dsa parameters
[0:24] <toad_> we could drop the fields we don't actually use ...
[0:24] <nextgens> and then have a DMT message for references
[0:24] <nextgens> iirc we have a method on node to output a compressed fileset
[0:24] <toad_> we could drop the fields we don't use _in a backwards compatible way_ even
[0:24] <nextgens> and it's used only during the authentication phase
[0:25] <toad_> the point is, we already HAVE a reference
[0:25] <toad_> before we do the setup
[0:25] <nextgens> hmm that's true
[0:25] <nextgens> but our current scheme allowed to do opennet as well ;)
[0:25] <nextgens> if we leave outter encryption appart
[0:25] * phrosty (n=phrosty@) has joined #freenet
[0:25] <nextgens> or we knew the id string
[0:26] <nextgens> no ?
[0:26] <toad_> nextgens: no, opennet will need a different setup
[0:26] <toad_> the current scheme only works because both sides know the other's identity in advance
[0:26] <Zothar_Work> having a DMT for the reference could be good because then that DMT could be sent on node name change (for example); in theory, that could just be a N2NM
[0:27] * nextgens thinks that we should release some fixed build soon
[0:28] <nextgens> maybe we can drop node names and so on from the compressed ref. we send if it helps
[0:28] <nextgens> but we shouldn't hold on releasing because we are doing^wplanning "longer term" changes
[0:29] <toad_> okay...
[0:29] <toad_> nextgens: indeed
[0:29] <toad_> here's the deal
[0:29] <toad_> version, location are essential
[0:29] <toad_> lastGoodVersion is pretty much essential
[0:29] <nextgens> iirc we send location before
[0:29] <nextgens> after
[0:29] <nextgens> using a DMT message
[0:29] <toad_> ah that's true
[0:30] <nextgens> as well as a DetectedIP
[0:30] <toad_> but we'd need to not send packets until later
[0:30] <nextgens> wich we don't ever claim ;)
[0:30] <toad_> name at the moment is checked for
[0:30] <toad_> if there is no name, it fails
[0:30] <nextgens> iirc the DetectedIP one stay on the unclaimedFIFO
[0:31] <toad_> now, we could introduce a FNPNewName message, which IMHO would be a good idea
[0:31] <nextgens> +s
[0:31] <toad_> nextgens: it shouldn't...
[0:31] <toad_> nextgens: that's a bug...
[0:31] <toad_> if we introduce a FNPNewName, we can get rid of the name
[0:31] <toad_> also on the N2NM's
[0:31] <nextgens> I've noticed it but haven't got around starting eclipse to contribute to freenet yet
[0:31] <toad_> the physical.udp is important; it can be deferred but only if there's a good reason
[0:32] <nextgens> btw, that might be the problem
[0:32] <toad_> the lastGoodVersion is important
[0:32] <nextgens> it can get long with ipv6 addresses
[0:32] <toad_> nextgens: file a bug ...
[0:32] <toad_> now, the big ones
[0:32] <toad_> the peer crypto group and the peer pubkey are only checked if we don't have one, in the current code
[0:32] * mozillaman (n=chatzill@) has joined #freenet
[0:32] <toad_> they cannot be changed on a connect
[0:32] <toad_> and I expect this to remain the case for some considerable time
[0:33] <nextgens> that was a backward compatibility hack iirc
[0:33] <toad_> so
[0:33] <toad_> we can just drop the peer crypto group and the peer pubkey
[0:33] <toad_> instantly solve the problem
[0:33] <nextgens> because references lacked it
[0:33] <nextgens> and remove the hack, indeed
[0:33] <toad_> no, we only look for it if we don't already have one
[0:33] <toad_> okay
[0:33] <nextgens> why on earth shouldn't we have one ?
[0:33] * toad_ files relevant bugs ...
[0:34] <toad_> nextgens: we should now
[0:34] <nextgens> what about fixing them on the fly ? :)
[0:34] <toad_> so you're right
[0:34] <nextgens> I suggest we remove dead code
[0:34] <toad_> yeah
[0:34] <nextgens> same goes for ARKs
[0:34] <nextgens> we do send crypto parameters there for the same reason
[0:34] <nextgens> and we shouldn't anymore imo
[0:34] <mozillaman> Anything new to report on freenetwatch? :)
[0:35] <nextgens> unless we plan to use ARKs for opennet
[0:35] <toad_> mozillaman: a new build fixing lots of bugs very soon
[0:35] <toad_> nextgens: why not have ARKs on opennet?
[0:35] <nextgens> toad_> I meant unless we use ARKs as a swapping medium
[0:36] <nextgens> toad_> crypto parameters shouldn't be updated from arks
[0:36] <toad_> right
[0:36] <toad_> they can't be
[0:36] <toad_> you change your pubkey, you have to get new darknet connections
[0:36] <toad_> for now anyway
[0:36] <nextgens> but the backward compatibility hack does it if nothing is set iirc
[0:36] * nextgens wrote it
[0:37] <nextgens> so if you wipe out the "on-connect" part, I suggest you have a look to arks as well ;)
[0:37] <toad_> hmmm?
[0:37] <toad_> if nothing is set, we'll accept a new pubkey yeah
[0:37] <toad_> same with arks i think...
[0:37] <nextgens> yep
[0:38] * nextgens suggests you remove both things alltogether
[0:38] <toad_> okay, so the ARK should not be included in the reference which is exchanged on connection either?
[0:38] <nextgens> on the basis that everyone has "up to current standard" node references
[0:38] <mozillaman> toad_: Could you describe or give list of these "things" please? :)
[0:39] <nextgens> that might be a good idea to allow it to be updatable
[0:39] <toad_> hmmm
[0:39] <toad_> the current code allows us to overwrite an old ARK
[0:39] <toad_> is that a good thing?
[0:39] * nextgens thinks so
[0:39] * mozillaman just loves the work ARK
[0:39] <nextgens> there is no security issue involved there
[0:39] <mozillaman> *word
[0:39] <toad_> mozillaman: 1003 will fix the bandwidth bug, the USK insert bug, and the N2NTM flood bug
[0:39] <toad_> mozillaman: that big enough for you?
[0:40] <toad_> nextgens: so we DO want the ARK in the ref then?
[0:40] <nextgens> it might 3nl@rge your p3n1s as well
[0:40] <nextgens> toad_> I think we do
[0:40] <mozillaman> toad_: YIPPIE!
[0:40] * mozillaman begins writing the article
[0:40] <OctobersDark> so i can sell my big black truck..good.
[0:41] <mozillaman> toad_: This will be a mandatory then?
[0:41] <mozillaman> the release
[0:41] <nextgens> as they are mtu changes, it ought to
[0:41] <mozillaman> no gential enlargement...
[0:41] <nextgens> at least it ought to be "time bombed"
[0:41] <CIA-14> toad * r11099 /trunk/freenet/src/freenet/node/FNPPacketMangler.java: Logging
[0:42] <toad_> mozillaman: no, well not _immediately_
[0:42] <toad_> but yes it probably ought to be time-bombed, since 998 was
[0:42] <toad_> which has those bugs
[0:42] * nextgens plans to head to bed
[0:42] <toad_> ok, /me deals with this side-issue first
[0:42] <nextgens> I've got to wakeup early tomorrow
[0:44] <CIA-14> toad * r11100 /trunk/freenet/src/freenet/node/PeerNode.java:
[0:44] <CIA-14> Don't pick up DSA group/pubkey on connect any more. It should be in all references.
[0:44] <CIA-14> Make DSA group/pubkey final.
[0:46] <mozillaman> toad_: Two more questions
[0:46] <mozillaman> 1) when will it be out?
[0:46] <nextgens> when it's ready
[0:46] <mozillaman> 2) will it be a mandatory, if so when will it time-up?
[0:46] <mozillaman> For the article :P
[0:46] <CIA-14> toad * r11101 /trunk/freenet/src/freenet/node/PeerNode.java:
[0:46] <CIA-14> Require a DSA group and DSA pubkey in *any* reference passed into PeerNode.
[0:46] <CIA-14> Remove old back compat hack to deal with builds < 970 changing the signature calculation.
[0:46] <Zothar_Work> ..which is a standard opensource development answer, as I'm sure you know :)
[0:47] <mozillaman> Actually
[0:47] <nextgens> toad_> may you have a look to the ARK part too while you're at it ?
[0:47] <mozillaman> 2 hours after the last person who asks.
[0:47] <mozillaman> Is the standard answer.
[0:47] <Zothar_Work> heh :)
[0:48] <mozillaman> Nextgen's reveals "[it will be ready] when it's ready".
[0:48] <CIA-14> toad * r11102 /trunk/freenet/src/freenet/node/PeerNode.java: Make name an optional field in post-connection-setup ref exchange.
[0:48] <CIA-14> toad * r11103 /trunk/freenet/src/freenet/node/PeerNode.java: Spacing
[0:49] * xxfooxx (n=foo@) Quit ("Leaving")
[0:49] <Zothar_Work> ...fast and furious commits are also standard opensource development practice... :)
[0:49] <toad_> mozillaman: soon
[0:49] <toad_> mozillaman: it will be out soon
[0:50] <toad_> mozillaman: when will it be mandatory? some time in december ...
[0:50] <mozillaman> :)
[0:50] <toad_> nextgens: what do you want me to do with the ARKs? i thought you said it should stay?
[0:50] <toad_> nextgens: currently we allow ARKs to be changed when we connect, but not at any other time
[0:52] <nextgens> toad_> I think that allowing the ARK address to be updated is a good thing security wise
[0:52] <Zothar_Work> well, that's the same as node names actually, so a noderef DMT would be needed I think
[0:52] <nextgens> toad_> as opennet/whatever might reveal your ARK's pubkey at some point ...
[0:52] <Zothar_Work> (or, again, a N2NM)
[0:52] <nextgens> toad_> and you might want to get rid of some "old" peers
[0:52] <nextgens> toad_> and to rekey your ARK
[0:53] <Zothar_Work> (which can handle a SimpleFieldSet easily)
[0:53] <toad_> nextgens: ok, so i don't need to change anything
[0:53] <nextgens> that's my views at least
[0:53] <toad_> Zothar_Work: an N2NM could handle a _large_ SFS, which makes it very appealing
[0:53] <nextgens> and I don't see any strong point suggesting there is something to do there
[0:54] <nextgens> toad_> but what's the point of sending physical.udp in the compressed reference ?
[0:54] <nextgens> that could be huge too
[0:54] <nextgens> and is useless imo
[0:55] <toad_> nextgens: it can be large
[0:55] <toad_> nextgens: but not as large as the pubkey/dsa group
[0:55] <toad_> nextgens: why not send it in the reference?
[0:55] <nextgens> because it's known
[0:55] <toad_> part of it is known
[0:56] <toad_> we want our peer to know the entirety of it
[0:56] <toad_> if we are multi-homed, for example, we want them to know the alternate routes to us
[0:56] * nextgens thinks we should send it in DetectedIP
[0:56] <toad_> DetectedIP only sends our view of the OTHER node's IP address
[0:56] <nextgens> or an other DMT message
[0:56] <toad_> which is different from our view of _our_ IP address
[0:56] <nextgens> hmm, true
[0:57] <nextgens> ok, keep it then
[0:57] <toad_> well, why send anything? :)
[0:57] * toad_ thinks we should send a reasonable amount of (small) stuff, but we can drop a lot
[0:57] <nextgens> but still : that might matter as its lengh isn't stable
[0:57] <toad_> e.g. the pubkey/dsa group
[0:57] * toad_ notes that when we implement STS, we will want the identity to be dropped
[0:57] <nextgens> those strings are bigger than dsa parameters imho
[0:58] <nextgens> no as we said we want to keep outter encryption
[0:58] <nextgens> don't we ?
[0:58] <nextgens> hmm
[0:58] <nextgens> in fact we have to know it
[0:58] <nextgens> so yes, dropping it makes sense
[0:59] <nextgens> toad_> dsa parameters are 256bits long at most, right ?
[1:00] <nextgens> toad_> physical.udp can be much longer than that
[1:00] <nextgens> each char. beeing a byte
[1:00] <toad_> nextgens: no
[1:00] <toad_> nextgens: look at your noderef
[1:00] * nextgens hasn't got any here
[1:00] <nextgens> I'm on the laptop
[1:01] * nextgens goes on bulix
[1:01] <nextgens> hmm
[1:01] <nextgens> ok, they are huge
[1:03] <toad_> we use big keys
[1:03] <CIA-14> toad * r11104 /trunk/freenet/src/freenet/node/PeerNode.java: Identity is optional.
[1:04] * toad_ wonders if we need an FNP message to tell the node when our other IPs change
[1:04] <nextgens> well, if you drop dsa parameters, we are almost 100% sure to win enough space for the auth packet to be below the mtu
[1:04] <toad_> yep
[1:04] * nextgens thinks that's enough for releasing the build
[1:05] <toad_> yep
[1:05] * toad_ working on it
[1:05] <nextgens> and that we should leave improvements for "when we rewrite low level stuffs"
[1:05] <nextgens> btw, any ETA on that yet ?
[1:06] <toad_> ooh, we don't check the testnet flag on connect
[1:06] <toad_> need to fix that!
[1:07] * nextgens goes to bed
[1:07] <toad_> nextgens: cya
[1:07] <nextgens> toad_> we don't ... but we check it when adding references
[1:07] <toad_> nextgens: right, but if it changes between restarts, we could end up with a bridge
[1:07] <toad_> at least until the node restarts
[1:08] <nextgens> we can't change it on the fly :)
[1:08] <nextgens> as it involves rekeying
[1:08] <nextgens> but adding the check doesn't hurt, agreed
[1:09] <nextgens> bbl
[1:20] <CIA-14> toad * r11105 /trunk/freenet/src/freenet/node/ (FNPPacketMangler.java Node.java NodeARKInserter.java):
[1:20] <CIA-14> Don't include unnecessary, invariant fields in the noderefs exchanged at the end of the connection setup process.
[1:20] <CIA-14> More work will eventually be done on this, but this should shrink the phase 2/3 auth packets enough to prevent most MTU problems.
[1:20] <toad_> testing would be appreciated!
[1:21] * Zothar_Work fires up his testing engine
[1:28] <CIA-14> toad * r11106 /trunk/freenet/src/freenet/node/NodeDispatcher.java: If NodeDispatcher dispatches an FNPIPAddressDetected, it should say so, not leak it to the _unclaimed set.
[1:28] <Zothar_Work> r11105 gives no problems to report
[1:29] <toad_> okay, will prepare a changelog ...
[1:33] <toad_> hmmm i think the changelog poster is broken
[1:36] <Zothar_Work> toad_: BTW, have you found the unclaimedFIFO Message Counts infobox on the /stats/ page informative?
[1:36] <toad_> Zothar_Work: yes, thanks for adding them
[1:36] <toad_> see #11105 for example
[1:36] * phrosty (n=phrosty@) Quit ("baseball is wrong: man with four balls cannot walk.")
[1:37] <Zothar_Work> sometimes I make tweaks/adds with the intention that they mean something to some other dev, like you :)
[1:37] <Zothar_Work> (I figure r11105 might have been related, which is what made me think to ask)
[1:37] <Zothar_Work> +d
[1:39] <toad_> :)
[1:40] <toad_> brb
[1:44] * fridim (n=fridim@) has joined #freenet
[1:45] <toad_> Zothar_Work: are you subbed to the CVS list?
[1:45] <toad_> Zothar_Work: have you got any messages since 11092?
[1:45] <Zothar_Work> yes
[1:45] <Zothar_Work> on CVS list
[1:45] * Zothar_Work checks
[1:45] <toad_> you've had messages since 11092?
[1:45] <Zothar_Work> I've got all the way to 11106
[1:46] <toad_> hmmm so it's something on my ned
[1:46] <toad_> end
[1:47] <mozillaman> I suddenly can't get peers past my NAT
[1:47] <mozillaman> It was working yesterday
[1:47] <toad_> mozillaman: result of the new build?
[1:47] * mozillaman wonders if he got an auto-update
[1:47] <mozillaman> I suspect
[1:48] <mozillaman> I can usually get 8 open connections at ny one time
[1:48] <mozillaman> Freenet 0.7 Build #1001 r11082:11083M
[1:48] <mozillaman> Suggestions?
[1:48] * mozillaman can't get the new freenetwatch updated :P
[1:49] <toad_> mozillaman: hmmm so it's not the result of 11106?
[1:49] <toad_> mozillaman: suggest you try 11106 :)
[1:49] <mozillaman> How can I do so without using update.cmd?
[1:49] <mozillaman> Is there a URL for the new .jar? :)
[1:50] <toad_> update testing
[1:50] <toad_> or shut down, download it manually over the jar
[1:50] <toad_> then restart
[1:50] <mozillaman> k, URL?
[1:50] <toad_> hmmm
[1:50] <toad_> downloads.freenetproject.org/alpha/
[1:50] <toad_> somebody send me a test email please?
[1:53] <toad_> Nov 29 01:52:48 emu postfix/smtp[15969]: 3C2729BCDF: to=<toad@amphibian.dyndns.o
[1:53] <toad_> rg>, orig_to=<matthew@freenetproject.org>, relay=none, delay=0, status=bounced (
[1:53] <toad_> mail for amphibian.dyndns.org loops back to myself)
[1:53] <toad_> hmmm
[1:53] <toad_> any ideas?
[1:54] <mozillaman> misconfiguration in aliases or domains
[1:56] <toad_> http://www.topology.org/linux/postfixloop.html
[1:56] <mozillaman> This node has not been able to connect to any other nodes so far; it will not be able to function normally. Hopefully some of your peers will connect soon; if not, try to get some more peers. You need at least 3 peers at any time, and should aim for 5-10.
[1:56] <toad_> need to add it tto mydestination
[1:56] <mozillaman> :s
[1:56] <toad_> mozillaman: do you have any peers on the darknet page? has your ip address been detected? has it changed?
[1:56] <toad_> has your port number changed?
[1:56] <mozillaman> toad_: Yes, no, yes, no
[1:56] <mozillaman> I set the IP hint though
[1:56] <mozillaman> didn't help
[1:57] <Zothar_Work> test email sent
[1:57] <toad_> mozillaman: your ip changed?
[1:57] <toad_> mozillaman: to a new valid ip address?
[1:57] <mozillaman> DynamicDNS
[1:57] <mozillaman> Should be
[1:57] <toad_> all your peers have you with your dyndns address?
[1:57] <mozillaman> It's a ADSL line
[1:57] <mozillaman> I doubt
[1:57] <mozillaman> I never set it
[1:58] * mozillaman thought the new detection stuff didn't need it
[1:58] <toad_> ok, so they don't
[1:58] <toad_> and let me guess, all your peers are behind NATs?
[1:59] <Zothar_Work> toad_: bounced: <matthew@freenetproject.org> (expanded from <toad@amphibian.dyndns.org>): possible alias database loop for matthew
[1:59] <mozillaman> More than likely :P
[1:59] <mozillaman> Can I just set IP address override?
[1:59] <mozillaman> Then get a new peer
[1:59] <toad_> it won't help
[1:59] <toad_> they need to know your new address
[1:59] <toad_> getting any new peer will fix it via ARKs
[1:59] <mozillaman> They won't auto-find it?
[2:00] <toad_> they can't find it from an ARK if you have zero connections
[2:00] <mozillaman> after I'm on freenet again
[2:00] <mozillaman> That's what I meant :P
[2:00] <mozillaman> OK, I have a plan
[2:00] <mozillaman> Can I enter
[2:00] <mozillaman> my.dyndns.host.com
[2:00] <mozillaman> into the override field?
[2:00] <Zothar_Work> yep, that's what should be there if you use it
[2:00] <mozillaman> Or, does it need special formatting?
[2:00] <mozillaman> yup
[2:01] <Zothar_Work> straight FQDN hostname
[2:01] <Zothar_Work> (sp)
[2:01] <Zothar_Work> ?
[2:01] <Zothar_Work> eh
[2:01] <mozillaman> OK, now, I need a sucker....er....volunteer's ref :P
[2:01] * mozillaman gets to -ref's
[2:03] * mozillaman has 1 connectio
[2:03] * mozillaman rubs hands with glee
[2:04] * mozillaman wonders how long the ARKs will take :s
[2:04] <Zothar_Work> could be interesting
[2:04] * mozillaman waits....
[2:04] <mozillaman> and waits...
[2:05] <mozillaman> networkSizeEstimateSession: 7 nodes
[2:05] <mozillaman> BACKED OFF
[2:05] * mozillaman cries
[2:07] <mozillaman> 4.25min later, still no more peers
[2:07] <Zothar_Work> it might take a bit as this _is_ worst case...
[2:07] <mozillaman> It gets worse
[2:07] <Zothar_Work> the dyndns will help a lot in the future though
[2:07] <mozillaman> I have to leave in 10min :P
[2:08] <mozillaman> Thanks :)
[2:08] <Zothar_Work> well, the peers should pick up the dyndns in your ARK if it got inserted successfully
[2:08] * mozillaman should make a note about this in FreeNetWatch
[2:08] <mozillaman> # Requests: 3
[2:08] <mozillaman> # ARK Fetch Requests: 15
[2:08] <mozillaman> # Total Output: 313 KiB (438 Bps)
[2:09] <mozillaman> Well, I'll see everyone tomorrow!
[2:09] <mozillaman> #1003 day ;)
[2:09] <Zothar_Work> laters
[2:09] * oierw (n=oierw@) Quit (Read error: 145 (Connection timed out))
[2:15] <toad_> relay_domains = amphibian.dyndns.org
[2:15] <toad_> yet it still gets "mail for amphibian.dyndns.org loops back to myself"
[2:15] <toad_> any postfix wizards here?
[2:17] * mozillaman (n=chatzill@) Quit ("Chatzilla 0.9.77 [Firefox 2.0/2006101023]")
[2:20] <toad_> okay, it's a dns problem...
[2:25] <toad_> emu.freenetproject.org is listed as my only MX
[2:25] <toad_> this is due to the dyndns interface not really being able to deal with the concept that you might have two MX's...
[2:25] <toad_> despite having a "backup MX" checkbox
[2:26] <toad_> the wierd thing is, it _was_ working
[2:27] <Zothar_Work> perhaps trace the alias path; perhaps X matches alias Y which matches alias Z which points back to X or something
[2:27] <Zothar_Work> gotta
[2:27] <Zothar_Work> gotta go
[2:28] <Zothar_Work> be back tomorrow morning GMT-6
[2:28] * Zothar_Work (n=chatzill@) Quit ("Chatzilla 0.9.76 [Firefox 1.5.0.8/2006102516]")
[2:53] <CIA-14> toad * r11107 /trunk/freenet/src/freenet/node/Version.java:
[2:53] <CIA-14> 1003 changelog:
[2:53] <CIA-14> - 1002 Mandatory on December 13.
[2:53] <CIA-14> - Fix the USK inserts bug. (Broke jSite inserts except for the first time).
[2:53] <CIA-14> - Probably fix the node-to-node messages flood bug.
[2:53] <CIA-14> - Significantly reduce the packet size of the final stages of connection setup, hopefully fixes some connection issues where the user is behind a stupid firewall.
[2:53] <CIA-14> - Require a DSA group and public key in any noderef.
[2:57] <toad_> good night
[2:57] * toad_ (n=toad@) Quit (Remote closed the connection)
[3:00] * Ralith (n=ralith@) has joined #freenet
[3:17] * fridim (n=fridim@) Quit (Read error: 110 (Connection timed out))
[3:20] * fridim (n=fridim@) has joined #freenet
[3:58] * phrosty (n=phrosty@) has joined #freenet
[4:26] * fridim (n=fridim@) Quit (Read error: 110 (Connection timed out))
[4:27] * anonymouse (n=a@) Quit (Read error: 54 (Connection reset by peer))
[4:27] * anonymouse (n=a@) has joined #freenet
[4:28] * fridim (n=fridim@) has joined #freenet
[4:37] * ShipHead (i=ShipHead@) Quit (Remote closed the connection)
[4:39] * ShipHead (i=ShipHead@) has joined #freenet
[5:00] * Aranjedeath (n=aranjede@) has joined #freenet
[5:15] * Electrosys (n=steo@) has joined #freenet
[5:15] * Electrosys is now known as electrosy
[5:16] <electrosy> hello.
[5:16] <electrosy> How do you browse the old edition of the sites?
[5:17] <TheSeeker|Gone> USK/site/55/ == SSK/site-55/
[5:17] <TheSeeker|Gone> use SSK and input the edition you want
[5:22] * phrosty (n=phrosty@) Quit ("baseball is wrong: man with four balls cannot walk.")
[5:38] * electrosy (n=steo@) Quit (Read error: 110 (Connection timed out))
[5:51] * phrosty (n=phrosty@) has joined #freenet
[6:12] * timmy2chk (n=violent@) Quit ("so what?")
[6:55] * sanity (n=ian@) has joined #freenet
[6:55] * ChanServ sets mode +o sanity
[7:04] * mazzanet_ (n=mazzanet@) has joined #freenet
[7:10] * mazzanet_ (n=mazzanet@) Quit ("Magic mushrooms")
[7:10] * mazzanet (n=mazzanet@) Quit (Read error: 110 (Connection timed out))
[7:16] * Aranjedeath (n=aranjede@) Quit ()
[7:54] * sanity (n=ian@) Quit ()
[7:56] * HellViper (n=hellvipe@) has joined #freenet
[8:04] * lat (n=gerhard@) Quit ("Leaving")
[8:06] * lokadin_ (n=loki@) has joined #freenet
[8:26] * mazzanet (n=mazzanet@) has joined #freenet
[8:27] * mazzanet is now known as notmazzanet
[8:34] * notmazzanet (n=mazzanet@) Quit ("Leaving")
[8:35] * notmazzanet (n=mazzanet@) has joined #freenet
[8:36] * notmazzanet is now known as mazzanet
[8:40] * whiterabbit (n=Whiterab@) has joined #freenet
[8:47] * HellViper (n=hellvipe@) has left #freenet
[8:53] * mazzanet (n=mazzanet@) Quit ("Off to toy with PPPoE")
[8:59] * traderef (i=YC4You@) has joined #freenet
[9:02] <traderef> anyone have good freenet links to test ?
[9:08] * Goopdoop (n=myfriend@) has joined #freenet
[9:21] * whiterabbit (n=Whiterab@) Quit ("KVIrc 3.2.5 Anomalies http://www.kvirc.net/")
[9:47] * Dynam (n=TakeThis@) Quit (Read error: 145 (Connection timed out))
[9:47] * whiterabbit (n=Whiterab@) has joined #freenet
[9:52] * Goopdoop (n=myfriend@) Quit ()
[10:21] <sleon> traderef: ummm, i would not allow you to test my links :)
[10:21] <sleon> traderef: cause that way danger my anonymity
[10:24] <nextgens> sleon> update :)
[10:29] * Hadraniel (i=andy@) has joined #freenet
[10:31] <traderef> so how can I process ?
[10:53] * railk (n=railk@) has joined #freenet
[10:54] * railk (n=railk@) Quit (Client Quit)
[11:14] * ShipHead (i=ShipHead@) Quit (Remote closed the connection)
[11:18] * MineHaunter (n=van85@) Quit (Read error: 104 (Connection reset by peer))
[11:31] * ShipHead (i=ShipHead@) has joined #freenet
[11:32] * anonick (i=anonymou@) has joined #freenet
[11:35] * sandos (n=sandos@) Quit (Read error: 148 (No route to host))
[11:44] * MineHaunter (n=van85@) has joined #freenet
[11:54] <Hadraniel> hi! when people try to add my node they get the message "The integrity of the reference has been compromized!" ... what's wrong?
[11:56] * ShipHead (i=ShipHead@) Quit (Client Quit)
[12:03] * mazzanet (n=mazzanet@) has joined #freenet
[12:13] * toad_ (n=toad@) has joined #freenet
[12:13] * ChanServ sets mode +o toad_
[12:22] <toad_> how are your nodes behaving?
[12:22] * MineHaunter (n=van85@) Quit (Read error: 131 (Connection reset by peer))
[12:22] <toad_> hmmm
[12:23] <toad_> sbc's node is still not connected, i was sure the auth packet size changes would fix it ...
[12:24] <toad_> same with sandos's
[12:24] * toad_ needs to talk to them, get logs, fix bug
[12:35] <toad_> [OT] Vista dominance is inevitable: http://news.bbc.co.uk/1/hi/business/6191812.stm
[12:35] <toad_> [OT] this will result in the death of online freedom as we know it ... oh well, it was a nice internet...
[12:35] <toad_> [OT] :|
[12:41] <toad_> [OT] so Microsoft and IBM, the two big proponents of TCPA, and the two big software patent holders, will carve up the world between them ... everyone runs a trusted OS, development can only happen with megacorp patronage, free software is enslaved or killed, and ultimately concerns about piracy result in remote deletion of both dangerous programs ("viruses and trojans and hacking/filesharing tools") and illegal files
[12:42] <toad_> [OT] obviously that would include freenet ... linux has failed ...
[12:42] <toad_> [OT] oh well, if it comes to pass, I suspect I can find an alternative career ... :|
[12:44] <Hadraniel> toad_: looks as if my ref paste was buggy ... works now. thanks.
[12:45] <toad_> [OT] hrrm, the EU could ride to the rescue ... http://news.bbc.co.uk/1/hi/technology/6058512.stm (second last paragraph)
[12:46] <toad_> [OT] but TCPA isn't a regular monopolies issue, it's insidious ...
[12:48] <toad_> [OT] /me fully expects a signature from a central and supposedly impartial authority to be required for any security software run on your PC under Vista ... either that, or you have full control, but it taints your TCPA signature and therefore you can't access trusted download services, send email to your friends, ultimately even connect to the internet
[12:48] <toad_> Hadraniel: hmmm?
[12:50] * Caco_Patane (n=caco@) Quit ("No windows for this server")
[12:52] <_ph00> [ot] in theory, a possible solution would be building some kind of 'free internet'
[12:52] <_ph00> another completely separated network
[12:52] <_ph00> (I know... too difficult)
[12:52] <toad_> _ph00: yeah, assuming you can get critical mass, and assuming it's even legal
[12:53] <_ph00> yeah
[12:53] <_ph00> and those are both the first two most obvious problems, I'm pretty sure as soon as you start seriousy thinking about it, you end up with a big pile of problems
[12:54] <toad_> yeah you get more
[12:54] <toad_> but ultimately that's what we're looking at
[12:54] <toad_> illegal, home-made hardware and illegal, home-made software
[12:54] <toad_> and illegal, home-made networks
[12:54] <_ph00> if freenet could stay up without help by official providers...
[12:54] <_ph00> hem..
[12:55] <_ph00> how?
[12:55] <_ph00> dostributed wifi?
[12:55] <_ph00> i*
[12:55] <nextgens> hi
[12:55] <toad_> _ph00: there are options, but i'm worried about manufacturing
[12:55] * nextgens might have found an internship opportunity :))
[12:55] <_ph00> yup
[12:56] <nextgens> working on a cool subject :
[12:56] <nextgens> "Implementation of 802.11s (WiFi-based) Mesh Networking Functionalities "
[12:56] <toad_> _ph00: the chinese will almost certainly use TCPA to secure their own internet, so we can't rely on them producing bootleg hardware
[12:56] <_ph00> manufacturing *is* an issue
[12:56] <_ph00> maybe dissident chinese?
[12:56] <toad_> this is all assuming it does eventually reach the de juris mandatory phase
[12:56] <toad_> which there are many indications it may
[12:56] <_ph00> can steal some, or something
[12:57] <toad_> if it doesn't, you only have to build an entire alternative internet
[12:57] <_ph00> thou shalt not steal, but what the hell...
[12:57] <_ph00> :P
[12:57] <toad_> _ph00: :)
[12:57] <toad_> _ph00: thou shalt not deal with folk who usually sell cocaine and AK-47's either ...
[12:57] <_ph00> so we're 1:1 so far
[12:57] <_ph00> we should meet again in hell
[12:58] <_ph00> (maybe we already did)
[12:58] <_ph00> that is
[12:58] <toad_> also, if shit like the IPRED2 passes, it may be illegal merely to run free software without megacorp patronage - EVEN WITHOUT MANDATORY TCPA!
[12:58] <toad_> and when i say illegal i mean a criminal offence
[12:58] <toad_> because of software patents
[12:59] <_ph00> so we go underground, meet the chinese, the irani, the cuban etc dissident and hit back
[12:59] <_ph00> *somehow*
[12:59] <toad_> :)
[12:59] <_ph00> (hm... do we need a plan?)
[12:59] <nextgens> that's worst than that
[12:59] <_ph00> is it?
[12:59] <nextgens> see how laws go ...
[12:59] <_ph00> yeah
[12:59] <nextgens> for instance in France ...
[13:00] <toad_> I dunno, it seems very insidious and abstract ... we're talking about cyberspace civil war here ... creating a new War on Drugs perhaps
[13:00] <_ph00> I'm awarre of that. I see it like it's not much as 0we' should go underground, more like "they" pushing us
[13:00] <nextgens> insteed of targetting brazillions of "copyright infringer" they target the few dozen people writting tools enabling "dumb" users to do it more easily
[13:00] <toad_> it's possible to explain this stuff to regular people
[13:00] <toad_> nextgens: exactly
[13:00] <toad_> nextgens: initially, they prosecute them, using DADVSI and IPRED2
[13:01] <_ph00> cyberspace civil war? In court? How about "cyberspace is not real*?"
[13:01] <toad_> nextgens: but with mandatory TCPA (or maybe even with de factor mandatory TCPA), they can remote delete "malicious software"
[13:01] * nextgens isn't gonna buy any TCPA compliant hardware
[13:01] <toad_> nextgens: defined as software which threatens the interests of the copyright holders, as well as software which threatens the interests of the _user_
[13:01] <toad_> nextgens: you think?
[13:01] <nextgens> but I'm worried by poeple who don't care
[13:02] <toad_> nextgens: I will ... but I won't install any TCPA compliant software
[13:02] <nextgens> like people buying mactels ;)
[13:02] <toad_> it won't be possible to buy hardware which doesn't have TCPA ...
[13:02] <_ph00> don't understand = don't care
[13:02] <nextgens> that won't help
[13:02] <toad_> oh?
[13:02] <nextgens> the hardware will require TCPA enabled OS ;)
[13:02] <toad_> not yet
[13:02] <nextgens> you want an example ? :)
[13:02] <_ph00> it would, only too difficult
[13:02] <toad_> not for some considerable time
[13:02] <nextgens> see AMD's PIC
[13:02] <toad_> you need to make it universal before you can make it mandatory
[13:02] <nextgens> toad_> you're wrong :)
[13:03] <toad_> what will initially be deployed is strictly optional
[13:03] <toad_> unfortunately it can easily become de facto mandatory because of network effects
[13:03] <nextgens> toad_> http://www.amdboard.com/pic.html
[13:03] <toad_> and once it's de facto mandatory it can easily be made de jure mandatory
[13:03] <_ph00> someone should start the "cyberspace is not real" movement <= that ?
[13:03] <nextgens> http://www.gensw.com/pages/stories/amdpicst.htm
[13:04] <nextgens> "# Boot Security (Cryptographic handshake between firmware and OS/Application) assures the service provider of system integrity Links to bootsec landing page "
[13:04] * LucisFerens (i=user@) has joined #freenet
[13:04] <toad_> nextgens: ooh, nice
[13:04] <toad_> nextgens: One Laptop Per Child with TCPA !
[13:04] <_ph00> :P
[13:04] <nextgens> I'm a member of the "robotic squad" of my school
[13:04] <toad_> One Personal Non-Computing Device Per Child
[13:04] <nextgens> and we were planning to use such devices ...
[13:05] <nextgens> but we saw how restricted their use it
[13:05] <toad_> I assume you get to burn the key into ROM?
[13:05] <nextgens> nah
[13:05] <toad_> or do you need to pay AMD $1,000,000 to sign your software?
[13:05] <nextgens> it's protected flash memory
[13:05] <nextgens> you need to pay :)
[13:05] <toad_> :|
[13:06] <nextgens> initialy they planned not to restrict the use of a specific OS
[13:06] <nextgens> linux was meant to be the OS used
[13:06] <toad_> ahh well, MIT got there first ... time to market is everything, we can reasonably hope that the PICs will be less deployed than the OLPCs at least in the short run
[13:06] <toad_> nextgens: with GPL 2 they could still do that
[13:06] <toad_> nextgens: linus is an idiot
[13:06] <nextgens> and one of M$' staff saw the project ... said that's cool ... and incitated AMD to bundle winCE and only that
[13:06] <nextgens> </rants>
[13:07] <toad_> well, linus is a great co-ordinator/programmer/etc, who built a great kernel
[13:07] <toad_> unfortunately his political sense is either malicious or stupid
[13:07] <_ph00> assume it's malicious
[13:07] <toad_> I dunno
[13:08] <toad_> for small values of "malicious"
[13:08] * nextgens thinks he doesn't care :)
[13:08] <toad_> right, it doesn't matter that much
[13:08] <nextgens> afaik he choosed GPL by default
[13:08] <nextgens> but doesn't like the current policy of the FSF
[13:09] <toad_> well, just like us, he put COPYING in and forgot about it
[13:09] <toad_> many files didn't have headers explicitly specifying 2+
[13:09] <nextgens> so he acted to put bars in their weels
[13:09] <toad_> so, he then decided he doesn't like the GPL 3, so he'll just make it all GPL 2 only
[13:09] <toad_> this is why I had to contact everyone to get their permission to put in GPL2+ headers everywhere
[13:09] <toad_> which still isn't completed because a lot of people are seemingly impossible to contact
[13:10] * toad_ can probably exclude some of them though by tracing the code
[13:10] <toad_> it's the FSF's fault really, the default should be "2 or later" :)
[13:11] <nextgens> toad_> are your DNS issues solved now ?
[13:11] <toad_> yes
[13:11] <toad_> it's working
[13:11] <nextgens> ok, cool
[13:12] * nextgens has got two interresting project proposals
[13:12] <nextgens> one on mesh networking ... the other one on scallability issues, exchanging video on p2p networks
[13:13] <toad_> oh, if anyone has an email address/phone number/address for giannij (Gianni Johannssen), mjr (Mark J Reynolds?), gthyni (Goran Thyni), iakin, hirvox, pascal666, blured75, Jogy, kodist, nacktschneck, odonata, or bobsan, tell me!
[13:14] <nextgens> toad_> when we have transport plugins, we NEED to implement a DCCP based one
[13:14] <toad_> actually I bet I can pick up Jogy from the bug tracker
[13:14] <toad_> nextgens: DCCP?
[13:14] <toad_> what's DCCP?
[13:14] * mozillaman (n=chatzill@) has joined #freenet
[13:15] <nextgens> a layer 4 protocol on the OSI model
[13:15] <nextgens> Datagram Congestion Control Protocol
[13:15] <toad_> ahh, Jogy missed it because I wasn't able to send him email, okay, i'll ask him
[13:15] <toad_> nextgens: what's it used for?
[13:16] <nextgens> toad_> that's trivial ... most have accounts on mantis
[13:16] <nextgens> and we have their mails
[13:16] <toad_> yeah, i just realised that
[13:16] <toad_> sadly many of the older ones don't
[13:16] <nextgens> btw, a grep on /etc/spamassassin/white.lst might give you some as well
[13:16] <toad_> giannij's completely disappeared off the face of the internet for example (witness relocation, perhaps? :) )
[13:17] <toad_> good idea
[13:18] <nextgens> you can try to ask NickServ too
[13:19] <toad_> yeah
[13:19] <toad_> I did try using sourceforge nyms, and looking up their real names from there and searching for them
[13:19] <toad_> that worked for some
[13:20] * krts (i=krts@) has joined #freenet
[13:21] <toad_> brb
[13:21] * krts (i=krts@) has left #freenet
[13:29] * LucisFerens (i=user@) Quit (Remote closed the connection)
[13:29] * greycat (i=rfc1413@) has joined #freenet
[13:31] * LucisFerens (i=user@) has joined #freenet
[13:35] <nextgens> http://emu.freenetproject.org/~nextgens/stages/802.11 Mesh Network Implementation.doc
[13:35] <nextgens> http://emu.freenetproject.org/~nextgens/stages/P2P Video Distribution.doc
[13:35] <nextgens> http://emu.freenetproject.org/~nextgens/stages/DCCP Simulation.doc
[13:36] <nextgens> all of them seems to be cool
[13:36] <nextgens> and interesting
[13:36] <toad_> nextgens: you are working on repeating everything a bazillion times, or using directionals, or something I haven't heard of?
[13:36] <nextgens> no, why ? :)
[13:37] <nextgens> I just got up early today
[13:37] <nextgens> and think writing down things
[13:37] <mozillaman> New freenetwatch out: http://127.0.0.1:8888/USK@jotJldLVFPDEnvRqfhBWsnXPQpOS%7eQrawxFjgsLZcFQ%2cxnNqE4Z%7ezMHmIUmqrA0oziUFSXNOAC7OhOOH4yhcBq4%2cAQABAAE/freenetwatch/35/
[13:38] <nextgens> mozillaman> I've never managed to get -34 here
[13:38] <mozillaman> strange
[13:38] <toad_> nextgens: "mesh networking" usually means one aerial per device, every node repeats everything, bandwidth is very low
[13:38] <mozillaman> Mabey not enough network latency?
[13:38] <nextgens> and 35 is dnfing too
[13:38] <mozillaman> :'(
[13:38] <mozillaman> Keep trying
[13:38] * mozillaman 's understanding is nodes will keep looking for the content
[13:39] <nextgens> toad_> as far as I understood : it's not reapeating everything
[13:39] <nextgens> but using intelligent "routing"
[13:39] <toad_> nextgens: hmmm ok
[13:39] <nextgens> at a lower level
[13:39] <toad_> mozillaman: works for me
[13:39] <nextgens> like "I know I can reach A,B,C" .... but that's all
[13:39] <mozillaman> toad_: :)
[13:40] <mozillaman> nextgens: Get more peers! :P
[13:40] <toad_> nextgens: i also heard all the good mesh routing schemes broadcast info on every node
[13:40] * nextgens has toad in his peers
[13:40] <toad_> nextgens: so they don't scale beyond a couple thousand nodes
[13:40] * sandos (n=sandos@) has joined #freenet
[13:40] <nextgens> well, that doesn't matter imo
[13:40] <sandos> toad_, I dont have your ref ;)
[13:40] <nextgens> as WiFi is designed to be LAN anyway
[13:41] <toad_> a good, efficient, scalable mesh network algorithm would be _really_ cool
[13:41] <nextgens> mozillaman> got -35 too now
[13:41] <mozillaman> nextgens: Great :)
[13:41] <toad_> nextgens: we use it for WAN effectively
[13:41] <toad_> nextgens: so do many businesses
[13:41] <nextgens> but it's not designed for that purpose :)
[13:41] <toad_> maybe so, but it works
[13:41] <nextgens> and in businesses case, I don't see how mesh would help
[13:42] <nextgens> they don't need it
[13:42] <toad_> WiMax is designed for WAN, but only one network can exist in a single area and frequence
[13:42] <toad_> y
[13:42] <toad_> so I expect it to be very expensive licensing
[13:42] <mozillaman> What about peering?
[13:42] * mozillaman finds that link....
[13:43] <toad_> nextgens: street-level access is really expensive, you need loads of APs
[13:43] <mozillaman> http://www.meraki.net/products.html
[13:43] <mozillaman> That
[13:43] <mozillaman> Uses peer-to-peer access points
[13:43] <mozillaman> it's been proven
[13:43] <mozillaman> It can have multiple net inputs at various peers
[13:43] <toad_> nextgens: if you can do it a la freifunk, with just everyone having their own omni, you could make a community network rather easily
[13:43] <mozillaman> They've been using it to bring the net to whole communities
[13:44] <toad_> mozillaman: BW uses a hierarchical structure
[13:44] <toad_> mozillaman: "mesh networks" are usually extremely inefficient on bandwidth
[13:44] * mozillaman just knows that one work, nothing more
[13:44] <toad_> at least, that's been what our research has told us
[13:44] <mozillaman> http://roofnet.net
[13:46] <toad_> Networks containing dozens of Minis can provide seamless indoor coverage across entire apartment complexes and neighborhoods.
[13:46] <toad_> so it doesn't scale
[13:47] * nextgens thinks that the best use of mesh networks is in ad-hoc mode
[13:47] <nextgens> ie : you "plug & play" everywhere
[13:47] <nextgens> without any infrastructure
[13:48] <toad_> which means you can connect to the APs in range, and that's it. no routing, because that won't scale anyway
[13:48] <nextgens> you need such things to provide street-level access ... agreed
[13:49] <nextgens> but I'm not convinced it matters that it doesn't scalle
[13:49] <nextgens> -l
[13:49] <toad_> well, it matters to BW, who would like to use mesh networks for last-mile, but can't
[13:49] <nextgens> as most people will connect "to the internet"
[13:49] <nextgens> and won't route in the mesh
[13:49] <nextgens> they will try to access an AP though the mesh because they are to far
[13:49] <nextgens> or something like that
[13:51] <toad_> yeah maybe
[13:51] <toad_> if the mesh cannot reliably deliver internet access then everyone with an AP will also have a wired broadband connection :)
[13:51] <toad_> but as roofnet shows, you can do it for really small networks
[13:52] <mozillaman> It's been done for 75 nodes I think
[13:52] <mozillaman> with MIT tests
[13:52] * sanity (n=ian@) has joined #freenet
[13:52] * ChanServ sets mode +o sanity
[13:53] * mozillaman feels more sane already
[13:55] <toad_> hmmm
[13:55] <toad_> ExOR looks interesting
[13:55] * toad_ wonders if it's patented ... I mean, it's MIT, how could it NOT be patented? ;)
[13:55] <nextgens> people on frost love me
[13:56] <nextgens> :D
[13:57] <nextgens> http://ocw.mit.edu/NR/rdonlyres/Engineering-Systems-Division/ESD-342Spring-2006/BD610EFA-BB5F-44FA-8B79-BDC69566C96B/0/final_roofnet.pdf ?
[13:59] * sleon sets mode +v anonick
[13:59] * Temps_sideral (n=jag@) has joined #freenet
[14:01] <anonick> sleon, tnx
[14:10] <toad_> aha
[14:10] <toad_> ExOR only works for large files
[14:10] <toad_> that figures ...
[14:10] <toad_> large being over 100kB
[14:11] * Zothar_Work (n=chatzill@) has joined #freenet
[14:16] * fabfac (i=laloggia@) has joined #freenet
[14:20] * nextgens logs on mantis
[14:26] <toad_> nextgens: :)
[14:28] * toad_ posted about ExOR, roofnet, meraki to bw-tech ...
[14:28] <toad_> bbiab
[14:49] * kasvio (n=kasvio@) has joined #freenet
[14:55] * sanity (n=ian@) Quit (Read error: 110 (Connection timed out))
[14:56] * tmecklem (n=mecklem_@) has joined #freenet
[14:59] <anonick> nextgens, i'm unable to join on #freenet-refs. Do you can help me or i need to register my gpg key to freenode?
[15:02] * LucisFerens (i=user@) Quit (Remote closed the connection)
[15:04] * anonick (i=anonymou@) Quit ("Sto andando via")
[15:08] <CIA-14> toad * r11108 /trunk/freenet/src/freenet/crypt/Util.java: Delete unused function written by giannij, who we can't seem to get hold of.
[15:09] * xxfooxx (n=foo@) has joined #freenet
[15:09] <xxfooxx> any advice re ref bots?
[15:10] <Zothar_Work> what kind of advice?
[15:11] <xxfooxx> sorry i'll clarify
[15:11] <xxfooxx> is it a good or bad idea to exchange refs with ref bots?
[15:11] <Zothar_Work> it's the same as exchanging with people, but more automated
[15:12] <Zothar_Work> I can give you details for your own if you'd like
[15:13] <xxfooxx> interesting.. I've heard people suggest they're more likely to be used to eavesdrop
[15:14] <Zothar_Work> there was a bot on the refs channel awhile back that was thought to be logging the channel that was banned I believe, though a bot is not required to log the channel; refbot.py does not log the channel unless someone is changing/adding stuff
[15:14] <xxfooxx> ok
[15:15] <CIA-14> toad * r11109 /trunk/freenet/src/freenet/ (22 files in 5 dirs):
[15:15] <CIA-14> Add some more GPL headers. Also delete some dead comments.
[15:15] <CIA-14> Still waiting for the following copyright holders:
[15:15] <CIA-14> iakin, jogy, giannij, mjr, gthyni, hirvox, pascal666, kodist, nacktschneck, bobsan, odonata, devrandom.
[15:24] <CIA-14> toad * r11110 /trunk/freenet/AUTHORS: Add an AUTHORS file.
[15:27] <xxfooxx> would more freenet users make it faster?
[15:27] <CIA-14> toad * r11111 /trunk/freenet/AUTHORS: Put lower list in alphabetical order too.
[15:27] <mozillaman> Of course it would!
[15:27] <mozillaman> More users! More users!
[15:27] <xxfooxx> wouldnt they just add to the load?
[15:27] <toad_> we don't really know
[15:28] <toad_> but generally there are costs to scaling up
[15:28] <xxfooxx> im not math genious, but i'd imagine more users = more load, but i suppose some percentage will be idle..
[15:28] <kasvio> but more users = more bw right?
[15:28] <xxfooxx> only one way to find out.. =)
[15:28] <toad_> indeed, there are benefits too
[15:28] <toad_> one of the benefits is more users = more true darknet connections
[15:29] <toad_> that makes the network far more robust
[15:29] <toad_> as well as improving topology
[15:29] <toad_> not so much the latter once opennet is done though
[15:30] * newbie (n=dude@) has joined #freenet
[15:30] <newbie> hi
[15:30] <newbie> what does "Unknown external address" mean?
[15:31] <newbie> i get this message on the fproxy page
[15:32] <kasvio> just that your node can't find your external ip I guess
[15:32] <newbie> is that a problem?
[15:32] <newbie> how do i fix it?
[15:33] * anonick (i=anonymou@) has joined #freenet
[15:33] <kasvio> it might make connecting to others harder (for your node), not sure how to fix it
[15:36] <Zothar_Work> if your external IP address is in your ref at http://127.0.0.1:8888/darknet/myref.fref then you can probably safely ignore the message; I believe there's a bug filed for that
[15:40] <newbie> ok
[15:40] <newbie> thx
[15:40] * anonick (i=anonymou@) Quit (Client Quit)
[15:43] * tmecklem (n=mecklem_@) has left #freenet
[15:43] * LucisFerens (i=user@) has joined #freenet
[15:51] <newbie> http://dark-code.bulix.org/s02e5w-24577
[15:52] <kasvio> you should post that in #freenet-refs :)
[15:52] <toad_> newbie: #freenet-refs
[15:53] <newbie> sorry
[15:53] <newbie> wrong channel
[15:53] <newbie> :)
[15:53] <toad_> it does have your real IP in, which is good...
[15:53] <toad_> bbiab
[16:19] * sbc (n=sbc@) has joined #freenet
[16:20] * traderef (i=YC4You@) Quit ()
[16:22] <sbc> hi toad. I just need to get some food, but I should be here all night/evening, if you need some logs, just tell me.
[16:29] * railk (n=railk@) has joined #freenet
[16:38] * mozillaman_ (n=chatzill@) has joined #freenet
[16:42] * mozillaman (n=chatzill@) Quit (Connection timed out)
[16:42] * mozillaman_ is now known as mozillaman
[16:44] * mozillaman (n=chatzill@) Quit (Read error: 104 (Connection reset by peer))
[16:56] * railk (n=railk@) Quit ("Cya, wouldn't want ta be ya!")
[16:57] * paveq_ (n=paveq@) has joined #freenet
[17:09] <sandos> toad_, if you have added me, I dont have you added. Need your ref.
[17:09] * paveq (i=paveq@) Quit (Read error: 110 (Connection timed out))
[17:30] * oCc (n=nodeOne@) has joined #freenet
[17:30] <nextgens> sbc> toad said earlier he would need some
[17:32] <nextgens> toad_> iakin is a .5 contributer iirc
[17:32] <sbc> nextgens: Do you know if he needed some special logging settings? I could just start logging now, and send him everything at the end of the evening?
[17:33] <nextgens> toad_> isn't ordonata == Omega666 == Robert Guerra ?
[17:33] <sbc> or should I just wait around for him to finish up chatting #freenet-chat and come in here and work? :P
[17:33] <xxfooxx> say... what happens if/when my dynamic ip changes?
[17:34] <xxfooxx> do i have to get new refs?
[17:34] <nextgens> sbc> he wanted to debug connection issues iirc
[17:34] <nextgens> :)
[17:34] <nextgens> toad_> hirvox is a .5 contributor too iirc
[17:34] <nextgens> toad_> devrandom is on i2p
[17:35] <nextgens> xxfooxx> no
[17:35] * Baughn (n=svein@) has joined #freenet
[17:36] <xxfooxx> phew
[17:36] <kasvio> xxfooxx: if both you and your peer have gotten a new ip you can't connect (assuming neither use ddns).
[17:37] <nextgens> kasvio> that's a false statement if at least one of your peer is using a static address and the node manages to connect to it
[17:39] <Baughn> Say, is bin/1run.sh supposed to give me "This script isn't meant to be used more than once" on my /first/ run?
[17:39] <greycat> I'd guess not.
[17:39] <nextgens> no
[17:39] <nextgens> :)
[17:39] <nextgens> I'll fix it
[17:39] * sanity (n=ian@) has joined #freenet
[17:39] * ChanServ sets mode +o sanity
[17:40] <Baughn> Should I be using a different tarball than the one on http://freenetproject.org//download.html ?
[17:40] <nextgens> Baughn> no, that's the one
[17:40] <nextgens> Baughn> I suggest you delete freenet-stable-latest.jar and freenet-ext.jar and retry 1run.sh
[17:41] <nextgens> the new tarball has been uploaded
[17:43] <Baughn> java.lang.NoClassDefFoundError: freenet/node/NodeStarter
[17:43] <Baughn> Sound familiar?
[17:44] <Baughn> ..also, curl doesn't like the ilink
[17:44] <Baughn> *link
[17:45] <Baughn> ..needs -L, fine
[17:46] * gruffy (n=daniel@) has joined #freenet
[17:49] <Baughn> nextgens: Okay, now it works; I needed to run update.sh
[17:56] * Iceman_B^Ltop (n=ice@) has joined #freenet
[18:00] * sanity (n=ian@) Quit ()
[18:01] * rguerra (n=rguerra@) has joined #freenet
[18:02] <toad_> nextgens: iakin contributed to 0.5, we used some of the 0.5 code in 0.7
[18:02] <toad_> nextgens: i sent a mail to his address from devl ...
[18:03] <toad_> and i found a graphic designer of the same name who probably isn't him, but is also a swede
[18:03] <toad_> nextgens: odonata = edt iirc
[18:04] * rguerra (n=rguerra@) Quit (Client Quit)
[18:07] <toad_> nextgens: if both sides are NATed then _ONE_ side changing IP can break it
[18:07] <toad_> if all your peers are NATed then your IP changing can result in no connections
[18:10] <nextgens> yes, if ALL your peers are natted
[18:12] <toad_> which is the case for a lot of ppl
[18:13] * makomk (n=aidan@) has joined #freenet
[18:22] * Hadraniel (i=andy@) has left #freenet
[18:26] <toad_> anyone got a working freemail?
[18:27] <toad_> can i send anyone a message?
[18:27] <sbc> toad_: I run freemail. If it works is another question. Feel free to mail me though.
[18:27] <toad_> address?
[18:30] <sbc> anything@sbc.freemail
[18:33] <toad_> okay, sent a message in reply to your previous one
[18:34] * electrosy (n=steo@) has joined #freenet
[18:34] <electrosy> Hello.
[18:35] <toad_> rehi electrosy
[18:35] <toad_> everything working?
[18:36] <electrosy> Think so. I haven't injected much since yesterday, Things seem fine.
[18:36] <electrosy> One thing though.
[18:36] <electrosy> I keep receiving the same NTNM over and over from the same node.
[18:36] <electrosy> like five times a day or something. we both have build 1003 now.
[18:37] <electrosy> brb.
[18:38] <toad_> electrosy: still? aaaaaaargh!
[18:38] <toad_> electrosy: it always has the same number?
[18:38] <toad_> electrosy: can you get into contact with the owner of that node, and get hold of his peer-data directory for that node?
[18:38] <sbc> toad_: Got your freemail. Have just sent reply.
[18:38] * newbie (n=dude@) Quit (Read error: 110 (Connection timed out))
[18:39] <electrosy> Damn, i dunno., i deleted them all. , And im not sure if i can get peer-data directory for that node. Would would getting peer-data entail?
[18:39] <electrosy> brb
[18:39] <toad_> electrosy: it's just a directory on disk, but you need to get him to give you it ...
[18:40] <toad_> in a zip file or something
[18:41] * kasvio (n=kasvio@) Quit ("Leaving")
[18:42] <nextgens> toad_> maybe his peer isn't up to date
[18:43] <nextgens> toad_> weren't you willing to debug the connection bug with sbc ?
[18:44] <sbc> nextgens: I think we are (or toad is trying to use smart phishing tricks to lure my webbank detail out of me in private chat...)
[18:45] <nextgens> :)
[18:45] <nextgens> ok
[18:46] <electrosy> toad_: I will see what i can do.
[18:46] <CIA-14> toad * r11112 /trunk/website/pages/news.php: SoC summary. Needs some trimming.
[18:46] * tuna (n=cccschne@) has joined #freenet
[18:46] <toad_> nextgens: huh?
[18:47] <toad_> nextgens: sbc is up to date yes
[18:47] <nextgens> toad_> I meant electrosy and his N2NM flooding problem
[18:47] <toad_> nextgens: he said he was running 1003
[18:47] <toad_> nextgens: does the website auto-update?
[18:48] <nextgens> it ought to
[18:48] <nextgens> FreenetLogBot> c'mon
[18:48] <nextgens> toad_> you just have to ask him ;)
[18:50] <electrosy> toad_: What directory do I need him to zip up and give me?
[18:50] <nextgens> toad_> may I tweak it a bit ?
[18:50] <electrosy> toad_: I am sending him a message now asking for his rev# of his build and a zip of the directory.
[18:50] <CIA-14> toad * r11113 /trunk/website/pages/news.php: Take out number that was wrong anyway, slight explanation, link to google SoC site.
[18:51] <toad_> electrosy: extra-peer-data-<port number>
[18:52] <CIA-14> toad * r11114 /trunk/website/pages/news.php: Full names
[18:53] <toad_> (full names and also thanks to google)
[18:54] <CIA-14> toad * r11115 /trunk/website/pages/news.php: Formatting fix, Dave Baker
[18:54] <nextgens> toad_> maybe you should use the parenthesis
[18:54] <nextgens> like :
[18:55] <nextgens> (<i>dbkr</i> - David Baker )
[18:55] <toad_> hmm maybe
[18:55] * toad_ doesn't propose to do any more work on it right now though
[18:55] <toad_> if somebody wants to trim them a bit that would probably be a good idea
[18:56] * tuna (n=cccschne@) has left #freenet
[18:58] * tuna (n=cccschne@) has joined #freenet
[18:59] <toad_> [OT] hrrm, i thought aircraft were radioactive anyway because they use depleted uranium as ballast ...
[19:01] <toad_> [OT] (CNN reports radioactive traces found on aircraft .. presumably the assholes who brought in the pollonium ... )
[19:02] <toad_> nextgens: would it be trivially possible to make the logger not record [OT] lines? or does it already not record them? it's standard practice apparently for some groups...
[19:03] <nextgens> toad_> it does record them
[19:03] <nextgens> and I'm against selective logging
[19:03] <toad_> why?
[19:04] <nextgens> but technically speaking, that's feasible
[19:04] <nextgens> because when a chan. is logged we assume everything is logged
[19:05] <toad_> well, it's just a matter of reducing the noise to signal ratio for those reading the logs
[19:05] <nextgens> otherwise some pleople might want/start using an @offthebook prefix on each of their lines
[19:05] * mozillaman (n=chatzill@) has joined #freenet
[19:05] <toad_> nextgens: you think they'll discuss illegal content on the theory that because they include [OT] it won't be logged ?
[19:06] <nextgens> I dunno whether that will be illegal or not
[19:06] <nextgens> the point is some people might be against logging
[19:06] <nextgens> and might want to prefix all their posts using it
[19:06] <toad_> ahhh
[19:06] <toad_> well then they're stupid
[19:06] <toad_> this channel is logged even if we don't log it
[19:06] <toad_> they have to assume that
[19:06] <tuna> hi nextgen i can not connect my node is not working but added other refs
[19:07] <toad_> it's logged by most of the IRC clients here, and probably by some TLAs and XTLAs too
[19:07] <toad_> if you're strongly opposed then we can leave it
[19:07] <sbc> what's a XTLA?
[19:09] * mozillaman reminds everyone that another new freenetwatch was just published
[19:09] <mozillaman>