wumpus changed the topic of #bitcoin-wizards to: This channel is is for discussing theoretical ideas with regard to cryptocurrencies, not about short-term Bitcoin development | http://bitcoin.ninja/ | This channel is logged. | For logs and more information, visit http://bitcoin.ninja
Quanttek has quit [Ping timeout: 250 seconds]
ahab has quit [Ping timeout: 250 seconds]
dansmith_btc has joined #bitcoin-wizards
nwilcox has joined #bitcoin-wizards
zooko has joined #bitcoin-wizards
nwilcox has quit [Ping timeout: 240 seconds]
AaronvanW has quit [Ping timeout: 246 seconds]
Ylbam has quit [Quit: Connection closed for inactivity]
PRab has quit [Read error: Connection reset by peer]
c0rw1n is now known as c0rw|zZz
PRab has joined #bitcoin-wizards
jonasschnelli has quit [Ping timeout: 246 seconds]
Krellan has quit [Ping timeout: 246 seconds]
cryptowest_ has quit [Ping timeout: 246 seconds]
GreenIsMyPepper_ has quit [Ping timeout: 246 seconds]
Luke-Jr has quit [Ping timeout: 246 seconds]
midnightmagic has quit [Ping timeout: 246 seconds]
nsh has quit [Ping timeout: 246 seconds]
Luke-Jr has joined #bitcoin-wizards
Krellan has joined #bitcoin-wizards
midnightmagic has joined #bitcoin-wizards
cryptowest has joined #bitcoin-wizards
nsh has joined #bitcoin-wizards
GreenIsMyPepper has joined #bitcoin-wizards
jonasschnelli has joined #bitcoin-wizards
frankenmint has joined #bitcoin-wizards
petertod1 is now known as petertodd
nullbyte has quit [Ping timeout: 246 seconds]
metamarc has joined #bitcoin-wizards
agorist000 has quit [Ping timeout: 252 seconds]
rusty has joined #bitcoin-wizards
King_Rex has quit [Remote host closed the connection]
<rusty>
petertodd: I created a revision of the version bits BIP. Care to comment? https://gist.github.com/rustyrussell/47eb08093373f71f87de. TL;DR: tally only on retarget periods, year-based timeout (eg. 2018), 1 retarget period delay between 95% and actual activation.
sausage_factory has joined #bitcoin-wizards
<petertodd>
rusty: I'm not sure tallying on retarget periods is actually a good idea btw, as it makes normal blocks different than retarget blocks
<rusty>
petertodd: ? Well, it saves some work and reduces the inaccuracy of continuous summation.
<petertodd>
rusty: inaccuracy re: probability overestimation?
blackwraith has quit [Ping timeout: 268 seconds]
<rusty>
petertodd: yeah, which isn't so bad at 95% (but I haven't done the numbers myself)
<petertodd>
rusty: saving some work isn't necessarily a good thing if it means we end up with two types of blocks to consider, normal and retarget blocks
<petertodd>
rusty: at 95% it's not bad
<petertodd>
rusty: I actually did a simulator for this, and don't remember finding anything interesting
<rusty>
petertodd: Yeah, wolfram alpha says probability of 90% of hashpower getting 1916 / 2016 blocks is 9.89 * 10^-17.
<petertodd>
rusty: yup!
<petertodd>
rusty: other issues are much more likely to cause issues, e.g. DoS attacks
<rusty>
petertodd: so doesn't really matter whether you measure every block, or every period. But if that threshold were ever to be revisited, I have a mild preference for retarget boundaries.
<rusty>
(Plus, the code is trivial0
<smk>
and syntax apparently
<petertodd>
rusty: sure it's trivial, but it's slightly more code than just doing it at every block :)
<rusty>
petertodd: unless you decide walking back 2016 blocks every block is too much work and write caching code :)
<petertodd>
rusty: well... the headers are in memory, so that's a very fast calculation
<petertodd>
rusty: we already do that for the soft majority mechanism anyway
zooko has quit [Ping timeout: 272 seconds]
<alpalp>
Checking every block gives a higher likliehood of early trigger, though not likely enough to make a big difference
<rusty>
petertodd: OK, well I can't marshal any immediate arguments against. I do, however, think it's neater to align changes at organised points.
<petertodd>
rusty: I think it's neater for every block to be processed exactly the same way ;)
sausage_factory has quit [Ping timeout: 240 seconds]
<rusty>
petertodd: that debate was already lost, with difficulty adjustment. I would agree if we were introducing some new concept. But it's really bikeshedding. If gmaxwell or sipa want to give an opinion...
nwilcox has joined #bitcoin-wizards
<petertodd>
rusty: well... difficulty adjustment is possibly a special case due to how it prevents attacks, and also, it *is* cached!
<rusty>
petertodd: but it's also an event in the miners' calendar. Aligning soft fork changes with it makes perfect sense. So it's slightly more work from a code POV, but simpler from a miner POV.
tripleslash has joined #bitcoin-wizards
<petertodd>
rusty: true! that's a valid argument, as it gives them an easier time window
<petertodd>
rusty: if you really wanted to be nice, you'd setup a weekly softfork time window and tie it to the median time calculation :) have fun coming to consensus on what timezone!
<aj>
rusty: i bet with the chinese miners, you could work out a weighting that turns out to be UTC+9.5!
<rusty>
petertodd: Ah, good thinking! Clearly, soft fork enforcement should be scheduled on the next leap second, so that everything breaks at once...
<petertodd>
rusty: lol
Emcy has quit [Ping timeout: 256 seconds]
<petertodd>
aj: lets add a meta-consensus layer to this, that lets miners vote on the soft-fork timing, to better fit their sleep schedules!
<petertodd>
(I'll suggest this next April)
nwilcox has quit [Ping timeout: 264 seconds]
<rusty>
petertodd: OK, I might post this to -dev for review, see if any other bikeshedding required. sipa said he'd create some code, and CodeShark_ is doing some soft fork cleanups too...
<petertodd>
rusty: oh, sipa did finally? good
<petertodd>
rusty: also, related: did you see my proposal for a non-expiration version of nVersion bits, that doesn't require state?
<rusty>
petertodd: I didn't see that... let me search logs...
<rusty>
petertodd: OK thanks. Unf. gtg, but you can see why I provided a table of years for timeouts; similar thinking. As for stateless, there's very little state (which can actually be calculated on the fly if you want to be completely stateless).
rusty has left #bitcoin-wizards [#bitcoin-wizards]
justanotheruser has joined #bitcoin-wizards
csggggg has joined #bitcoin-wizards
sparetire_ has quit [Quit: sparetire_]
<kanzure>
not sure if i am supposed to yap about bonded deployment stuff when versionbits is being described, or soft-fork timing is talked about
jtimon has quit [Ping timeout: 260 seconds]
agorist000 has joined #bitcoin-wizards
metamarc has quit [Ping timeout: 265 seconds]
cypher__ has joined #bitcoin-wizards
CodeShark_ has quit [Remote host closed the connection]
<kanzure>
"Founding cryptography on oblivious transfer - efficiently" http://mmp.cs.illinois.edu/pub/mpc-ot.pdf "transforming secure honest-majority MPC protocols into guaranteed security in MPC protocols with no honest majority"
<kanzure>
"Enigma: Decentralized computation platform with guaranteed privacy" http://enigma.media.mit.edu/enigma_full.pdf (seems to be a p2p secure multi-party computation ..thing)
luke-jr_ has joined #bitcoin-wizards
Luke-Jr has quit [Ping timeout: 244 seconds]
<kanzure>
page 7 section 5.3 doesn't make sense.. what about sybil attacks?
<kanzure>
"Using Bitcoin security deposits for punishing malicious nodes in MPC has been investigated by several scholars recently [22, 23]. We use a similar model, and extend it to penalize other malicious behaviors such as breaking correctness, which is validated by the SPDZ protocol (see section 5.1.2)"
<kanzure>
"The basic idea of Chain-Fibers is unchanged from a year ago; split the state-space up into strata and have separate transaction collators specialising in one or a number of state sub-spaces. Transactions requiring interactions from many a subspace would be accordingly more expensive (since collators would have to maintain presence on multiple chains) and take longer to execute (since there is a lesser chance that any given block would ...
<kanzure>
... contain a superset of the transaction’s subspaces). Validity of a transaction is verifiable in isolation through the provision of comprehensive Merkle proofs to its inputs alongside it in the block in which it is included."
CodeShark has joined #bitcoin-wizards
<kanzure>
"If a collation is challenged, then a larger randomly selected set of 2 ∗ m validators must attest to the validity of the collation in order to “redeem” it. Challengers can optionally pay a higher cost k ∗ cc in order to make the required redemption set even larger, 2 ∗ k ∗ m, right up to a maximum of the entire set of nv validators."
<kanzure>
i am not so convinced about that
<kanzure>
"An attacker can always create an unavailable collation, wait for challengers to challenge, and then publish the data and start the redemption process, costing the challengers money. The attacker may do this again and again, draining challengers’ resources, until eventually challengers no longer find it economically rational to challenge bad collations, giving the attacker free rein to push through invalid collations; we call this ...
<kanzure>
... the crying wolf attack."
Giszmo has quit [Quit: Leaving.]
<Taek>
the crying wolf attack seems like a pretty big problem to me
<Taek>
The real problem is when you have multiple fibers start to interact with eachother
<Taek>
you can only validate the chain then if you've validated each of the multiple fibers
<kanzure>
the next few paragraphs suggests using bonds that can be taken by someone proving you had invalid data or something
<kanzure>
"multiple fibers start to interact" i think the proposal is "let those take a longer time, who cares" which might work dunno
heretolearn has joined #bitcoin-wizards
roxtrongo has joined #bitcoin-wizards
<Taek>
sounds like a proposal with weak fundamentals to me. "take a longer time" -> if you are on a fiber that's suddenly interacting with other fibers, do you have to start validating those other fibers too?
snthsnth has joined #bitcoin-wizards
<Taek>
or do you just accept weaker security in the form of "nobody has proven it invalid, and I think some people are both honest and fully validating"
<Taek>
it doesn't seem that different from the security model of SPV sidechains
<kanzure>
"However, the result is important as it shows that, fundamentally, the level of overhead required in order to have a blockchain of size L is roughly m ∗ log(L), a very slowly-growing value. With clever use of zk-SNARKs, perhaps even a completely constant-overhead approach for blockchains of arbitrary size can be achieved."
dEBRUYNE_ has joined #bitcoin-wizards
<Taek>
Once you start introducing SNARKs the game changes quite a bit. If it is similar to CoinWitness you might be able to restore strong security properties
<Taek>
'“Fishermen” is a term given to freelance checkers. Since block validation and availability are both important, and since it is possible that sets of validators may be contractually bribed, it is important to have a mechanism to involve additional rational individuals in acting as “whistle-blowers” to avoid bogging the other validators needlessly checking all blocks.'
<Taek>
Any time you start relying on whistle blowers raises flags in my head
<Taek>
it just doesn't strike me as a good security model
<kanzure>
do you have the same opinion of using fraud proofs?
<Taek>
largely yes
<kanzure>
well at least you're consistent
<Taek>
With Bitcoin you get the strength of certaintly that every transaction in your history is legal. Once you start depending on fraud proofs, you acknoledge that things are happening outside of what you are witnessing
<Taek>
*things relevant to security
<Taek>
Especially if you are using fraud proofs as a method of scaling, it's a lot of things
dEBRUYNE_ has quit [Ping timeout: 265 seconds]
<Taek>
and given that even most miners do SPV mining, it's not really reassuring to assume that people are going to be actively validating and constructing fraud proofs on every part of the chain
<Taek>
fraud proofs are nice if you are looking for something in between SPV nodes and full nodes
<kanzure>
is there a denial-of-service attack against fraud proofs where you can increase the size of proofs of whatever fraud you have committed to impractical sizes
priidu has joined #bitcoin-wizards
<Taek>
I'm not 100% certain in my answer, but I think that's an unsolved problem
<Taek>
We don't know of a way to prevent such attacks, but we also don't know that such attacks can't be mitigated through improved methods
moa has quit [Quit: Leaving.]
<Taek>
". As such it is likely that validators and other large stakeholder would act as Fishermen to protect their investment" -> I think that this statement underestimates the dangers of 'diffusion of responsibility'
priidu has quit [Ping timeout: 240 seconds]
<CodeShark>
[21:39] (Taek) Any time you start relying on whistle blowers raises flags in my head <--- it all comes down to incentives
<CodeShark>
If whistleblowing is actually profitable it can be a fine security model
* gmaxwell
sets skepticism dial to 1
<CodeShark>
out of 11?
<gmaxwell>
CodeShark: so say you have some fraud proof thing and it's profitable to report fraud (and somehow you've avoided making it profitable for participants to conspire to create fake fraud). OK. So because the system works, it's never used. So it's not actually profitable in practice, though maybe it would be very profitable if it were ever used.
<gmaxwell>
But monitoring has an ongoing cost, regardless of if there is fraud.
<gmaxwell>
So this doesn't sound sufficiently stable for a security system.
<CodeShark>
right, agreed - outsourcing vigilance might still be profitable, though
<CodeShark>
You can still earn even if you don't detect fraud
<CodeShark>
but then you need to prove lack of fraud
<gmaxwell>
if you can efficiently prove a lack of fraud you just require the blocks to always carry one!
<CodeShark>
yes, indeed...Or...you could do something along the lines of your dual block challenge idea
<CodeShark>
Or....nvm
<gmaxwell>
sometimes wish I hand't mentioned that! :)
CodeShark has quit [Remote host closed the connection]
CodeShark has joined #bitcoin-wizards
<CodeShark>
yeah, it's hard to find a good incentives model for fraud proofs
Burrito has quit [Ping timeout: 244 seconds]
<CodeShark>
might be better to stick with partial checks and probabilistic verification
Dr-G2 has quit [Ping timeout: 244 seconds]
<CodeShark>
Is there any way to revive the fraud proof concept? My skepticism meter just went to 0.5
<gmaxwell>
I think what Taek said was right.
<gmaxwell>
They're useful in certian contexts.
<gmaxwell>
They can make reduced security modes better; they can make systems more efficient on average.
<gmaxwell>
Like, I think a bitcoin that was still scaled within the realm of single hobbiests but expensive enough to run that few would is okay with a fraud proof mostly world.
<gmaxwell>
Except for the partition and censor problems which I've never come up with a strong solution for.
<CodeShark>
I don't think that's what we're after, though
<gmaxwell>
Well I think thats all you can juice out of that scheme. In general I the fraudproofy stuff doesn't look so hot from a worst case analysis.
CodeShark_ has joined #bitcoin-wizards
snthsnth has quit [Ping timeout: 244 seconds]
Dr-G has joined #bitcoin-wizards
ThomasV has quit [Ping timeout: 264 seconds]
Dr-G has left #bitcoin-wizards ["Leaving"]
CodeShark is now known as CodeShark__
CodeShark_ has quit []
CodeShark has joined #bitcoin-wizards
Burrito has joined #bitcoin-wizards
ThomasV has joined #bitcoin-wizards
Ylbam has joined #bitcoin-wizards
melvster has quit [Ping timeout: 240 seconds]
antgreen` has quit [Remote host closed the connection]
melvster has joined #bitcoin-wizards
dEBRUYNE_ has joined #bitcoin-wizards
paveljanik has joined #bitcoin-wizards
JackH has joined #bitcoin-wizards
paleh0rse has quit []
roxtrongo has quit [Remote host closed the connection]
damethos has joined #bitcoin-wizards
frankenmint has quit [Remote host closed the connection]
priidu has joined #bitcoin-wizards
dEBRUYNE_ has quit [Ping timeout: 246 seconds]
gill3s has joined #bitcoin-wizards
ThomasV has quit [Ping timeout: 240 seconds]
csggggg has quit [Ping timeout: 255 seconds]
csggggg has joined #bitcoin-wizards
Logicwax has quit [Ping timeout: 244 seconds]
Tiraspol has quit [Ping timeout: 250 seconds]
Tiraspol has joined #bitcoin-wizards
DougieBot5000 has quit [Quit: Leaving]
Logicwax has joined #bitcoin-wizards
c0rw|zZz has quit [Ping timeout: 256 seconds]
c0rw|zZz has joined #bitcoin-wizards
roxtrongo has joined #bitcoin-wizards
roxtrongo has quit [Ping timeout: 272 seconds]
bedeho has quit [Ping timeout: 250 seconds]
ThomasV has joined #bitcoin-wizards
Logicwax has quit [Ping timeout: 244 seconds]
agorist000 has joined #bitcoin-wizards
sparetire_ has joined #bitcoin-wizards
Logicwax has joined #bitcoin-wizards
cypher__ has quit [Ping timeout: 255 seconds]
cypher__ has joined #bitcoin-wizards
agorist000 has quit [Ping timeout: 244 seconds]
cypher__ has quit [Ping timeout: 255 seconds]
metamarc has joined #bitcoin-wizards
CodeShark has quit [Ping timeout: 260 seconds]
CodeShark__ has quit [Read error: Connection reset by peer]
<psztorc>
Major conclusion is that centralization is tied to the cost of running a full node.
belcher has joined #bitcoin-wizards
<psztorc>
In the process, I also go on a bit of an intense tour of mining and hard/soft forks.
<kanzure>
lots of bolded text, not sure why
AnoAnon has joined #bitcoin-wizards
AnoAnon has quit [Read error: Connection reset by peer]
DougieBot5000 has joined #bitcoin-wizards
roxtrongo has joined #bitcoin-wizards
AnoAnon has joined #bitcoin-wizards
AnoAnon has quit [Read error: Connection reset by peer]
ttttemp has quit [Remote host closed the connection]
Starduster has quit [Ping timeout: 244 seconds]
rusty2 has left #bitcoin-wizards [#bitcoin-wizards]
jtimon has quit [Ping timeout: 246 seconds]
ttttemp has joined #bitcoin-wizards
zooko has joined #bitcoin-wizards
paveljanik has quit [Quit: Leaving]
guruvan- has joined #bitcoin-wizards
guruvan- is now known as Guest12813
Guest12813 has left #bitcoin-wizards [#bitcoin-wizards]
Starduster has joined #bitcoin-wizards
gill3s has joined #bitcoin-wizards
zwick has quit [Quit: WeeChat 1.3]
ThomasV has joined #bitcoin-wizards
frankenmint has joined #bitcoin-wizards
CodeShark has joined #bitcoin-wizards
ASTP001 has joined #bitcoin-wizards
nwilcox has joined #bitcoin-wizards
nullbyte has quit [Read error: Connection reset by peer]
JackH has quit [Ping timeout: 244 seconds]
nullbyte has joined #bitcoin-wizards
StephenM347 has joined #bitcoin-wizards
ASTP001 has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
Quanttek has quit [Ping timeout: 256 seconds]
KFl6WxEUI1Zbp2 has joined #bitcoin-wizards
<smooth>
psztorc: your diagram and satoshi quote are kind of busted, since gnutella has supernodes
<waxwing>
argh .. that 'decentralized' vs 'distributed' picture again .. makes no sense :(
adam3us has joined #bitcoin-wizards
<waxwing>
"It is an Engineering Requirement that Bitcoin be “Above the Law”" +1 psztorc
<waxwing>
i like that way of putting it. people tend to think you're some angsty teen anarchist if you only say the second part :)
ThomasV has quit [Ping timeout: 240 seconds]
<maaku>
that's a very nice tweet-sized expression of decentralization, thanks psztorc
Quanttek has joined #bitcoin-wizards
memymo_ has joined #bitcoin-wizards
nullbyte has quit [Ping timeout: 252 seconds]
nullbyte has joined #bitcoin-wizards
<waxwing>
psztorc: typo "you need to be join the club, and stay in"
Yoghur114 has joined #bitcoin-wizards
KFl6WxEUI1Zbp2 has quit [Quit: ZZZzzz…]
damethos has quit [Quit: Bye]
memymo_ has quit [Ping timeout: 252 seconds]
gill3s has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
roxtrongo has quit [Remote host closed the connection]
Newyorkadam has joined #bitcoin-wizards
<psztorc>
thanks everyone
<nullbyte>
its good, psztorc
<psztorc>
smooth: Do you think the gnutella supernodes harms the line of reasoning? I do wish Satoshi had just said "Bittorrent".
<belcher>
the name bitcoin alludes strongly to bittorrent
<psztorc>
Yeah and even the concept of a tracker kind of parallels the genesis block hash
<psztorc>
I did not consider the gnutella supernodes, but I think the ideas survive that incongruity..Satoshi clearly *does* mean "headless", no single point of failure, hard to attack, 100% decentralized.
Starduster_ has joined #bitcoin-wizards
<smooth>
psztorc: i dont know enough about how gnutella actually works. i just thought it was a poor fit with the message you were trying to convey with the picture (and following text)
<psztorc>
The other confusing thing is that it kind of overlaps with the Usenet analogy later, hard to know to put all of that in a good order.
Starduster has quit [Ping timeout: 244 seconds]
StephenM347 has quit []
antgreen has quit [Ping timeout: 268 seconds]
<smooth>
the usenet analogy is an interesting one, but perhaps flawed because spam on usenet made it intolerable to use, not just expensive
<smooth>
so much of the use value was destroyed by spam in a way that wouldn't necessarily be the case with bitcoin (I dont actually to personally sort through other people's junk transactions)
<smooth>
*dont actually need to
<smooth>
nor are people creating bitcoin spam to sell me something
frankenmint has quit [Remote host closed the connection]
<zooko>
smooth: sounds exactly right.
<zooko>
Usenet was *not* killed by spam soaking up its bulk data capacity.
<zooko>
It was killed by spam soaking up its human-navigation capacity.
<smooth>
zooko: the volume de facto centralized it though, which was part of psztorc's point i think
<smooth>
back in the day you just needed a modem and some peer willing to feed you. later it became almost entirely buying access to a large server from some corp
<psztorc>
I intended to express the concept that, while the data was itself decentralized, the actual *data layer* (the cables and computers) weren't private and were therefore subordinate to a central legislature.
<zooko>
I haven't seen psztorc's write-up yet, sorry.
<zooko>
I don't know if I agree with "volume centralized Usenet" and I don't think that even if it did that was relevant to Usenet's death.
<zooko>
But maybe psztorc knows more about the history of Usenet than I do. :-)
<smooth>
i agree death was more human based
<smooth>
human capacity
<zooko>
psztorc: that sounds wrong to me. Usenet was never successfully legislated/policed.
<zooko>
It was unnecessary for it to be, since it died due to spam overloading the human navigation systems long before legislators or police knew what it was. :-)
<zooko>
I believe it is still in use for bulk illegal data today.
<psztorc>
Funny you should mention that, I bring it up.
<smooth>
yeah i agree. charles schumer never went on tv and called for banning usenet
<zooko>
Okay, sorry, I'll shut up until I read your writings. :-)
<smooth>
although if he knew what was on there he certainly woudl have
<psztorc>
But I don't know a lot about Usenet...I did some background reading and multiple sources agreed that law enforcement put pressure on ISPs to deprecate the service.
<smooth>
i think that is largely true
<smooth>
maybe not even law enforcement per se, but their own lawyers
<psztorc>
And basically 0% of the current internet users even know what it is (which deserves some explanation).
<psztorc>
But I really do NOT know a lot about Usenet, and would welcome any corrections.
<psztorc>
My Usenet point is that, if the government *really* cared (big if), they could close Usenet down.
<zooko>
psztorc: okay, I'm 85% sure that what you say is 92% correct. Gotta run. :-)
<psztorc>
Ok thanks : )
jtimon has joined #bitcoin-wizards
kmels has quit [Ping timeout: 255 seconds]
DougieBot5000 has quit [Quit: Leaving]
zooko has quit [Ping timeout: 260 seconds]
jtimon has quit [Remote host closed the connection]
moa has joined #bitcoin-wizards
snthsnth has joined #bitcoin-wizards
<kanzure>
"todd trees"
maraoz has quit [Ping timeout: 264 seconds]
nullbyte has quit [Ping timeout: 264 seconds]
Newyorkadam has quit [Quit: Newyorkadam]
nullbyte has joined #bitcoin-wizards
Quanttek has quit [Ping timeout: 264 seconds]
Quanttek has joined #bitcoin-wizards
Quanttek has quit [Ping timeout: 250 seconds]
c0rw1n is now known as CactusPerson
CactusPerson has quit [Disconnected by services]
c0rw1n has joined #bitcoin-wizards
mountaingoat has quit [Ping timeout: 250 seconds]
Rynomster has joined #bitcoin-wizards
SDCDev has quit [Read error: Connection reset by peer]
DougieBot5000 has joined #bitcoin-wizards
blackwraith has joined #bitcoin-wizards
SDCDev has joined #bitcoin-wizards
Londe has quit [Quit: Londe]
Londe has joined #bitcoin-wizards
priidu has quit [Ping timeout: 256 seconds]
Rynomster has quit [Ping timeout: 256 seconds]
Starduster_ is now known as Starduster
jgarzik has quit [Read error: Connection reset by peer]