sipa changed the topic of #bitcoin-wizards to: This channel 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
mrpocoyo has left #bitcoin-wizards [#bitcoin-wizards]
Church- has joined #bitcoin-wizards
veleiro has joined #bitcoin-wizards
nabu has quit [Ping timeout: 264 seconds]
wumpus has quit [Ping timeout: 250 seconds]
wumpus has joined #bitcoin-wizards
go1111111 has joined #bitcoin-wizards
Chris_Stewart_5 has quit [Ping timeout: 240 seconds]
MaxSan_1 has joined #bitcoin-wizards
ruby32 has joined #bitcoin-wizards
MaxSan_ has quit [Ping timeout: 252 seconds]
frankenmint has joined #bitcoin-wizards
CrazyTruthYakDDS has joined #bitcoin-wizards
dEBRUYNE has quit [Quit: Leaving]
frankenmint has quit [Ping timeout: 244 seconds]
Ylbam has quit [Quit: Connection closed for inactivity]
nets1n has joined #bitcoin-wizards
Giszmo has quit [Ping timeout: 250 seconds]
MaxSan_1 has quit [Ping timeout: 258 seconds]
MaxSan_ has joined #bitcoin-wizards
ruby32d has joined #bitcoin-wizards
ruby32 has quit [Ping timeout: 250 seconds]
nets1n has quit [Remote host closed the connection]
jiggalator has joined #bitcoin-wizards
ruby32d has quit [Read error: Connection reset by peer]
akoko has joined #bitcoin-wizards
giel__ has quit [Ping timeout: 240 seconds]
jiggalator has quit [Client Quit]
Church- has left #bitcoin-wizards ["WeeChat 1.5"]
Nightwolf has quit [Read error: Connection reset by peer]
Nightwolf has joined #bitcoin-wizards
MaxSan_1 has joined #bitcoin-wizards
MaxSan_ has quit [Ping timeout: 272 seconds]
thrasher` has joined #bitcoin-wizards
adnn has joined #bitcoin-wizards
digitalmagus has joined #bitcoin-wizards
bramc has quit [Quit: Page closed]
MaxSan_ has joined #bitcoin-wizards
frankenmint has joined #bitcoin-wizards
MaxSan_1 has quit [Ping timeout: 258 seconds]
c0rw1n_ has quit [Ping timeout: 244 seconds]
Cory has quit [Ping timeout: 260 seconds]
PRab has joined #bitcoin-wizards
Pasha has joined #bitcoin-wizards
ruby32 has joined #bitcoin-wizards
frankenmint has quit [Ping timeout: 264 seconds]
Pasha is now known as Cory
rusty2 has joined #bitcoin-wizards
N0S4A2 has quit [Quit: WeeChat 1.5]
phantomcircuit_ is now known as phantomcircuit
mdavid613 has quit [Quit: Leaving.]
rusty2 has quit [Ping timeout: 250 seconds]
rusty2 has joined #bitcoin-wizards
JHistone has joined #bitcoin-wizards
justanotheruser has quit [Ping timeout: 246 seconds]
justanotheruser has joined #bitcoin-wizards
frankenmint has joined #bitcoin-wizards
chjj_ has quit [Ping timeout: 276 seconds]
adnn_ has joined #bitcoin-wizards
adnn has quit [Ping timeout: 264 seconds]
frankenmint has quit [Ping timeout: 244 seconds]
ThomasV has joined #bitcoin-wizards
<bsm117532>
I'm adding a networking delay to my braids project. I'm thinking to use a fat-tailed Cauchy distribution. Does anyone have a better suggestion for modeling Bitcoin's block propagation?
<bsm117532>
Also, for the record, the Cauchy distribution is also known as the Lorentz distribution and Breit-Wigner distribution, and is how we discover new particles in physics!
dnaleor has quit [Ping timeout: 250 seconds]
dnaleor has joined #bitcoin-wizards
<bsm117532>
Numerics I've run so far indicate a 4.8s median propagation time (measured from the orphan rate).
ruby32 has quit [Remote host closed the connection]
ruby32 has joined #bitcoin-wizards
Noldorin has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
[d__d] has quit [Ping timeout: 252 seconds]
JHistone has quit [Ping timeout: 264 seconds]
adnn has joined #bitcoin-wizards
ruby32 has quit [Ping timeout: 272 seconds]
adnn_ has quit [Ping timeout: 250 seconds]
pro has quit [Quit: Leaving]
frankenmint has joined #bitcoin-wizards
Church- has joined #bitcoin-wizards
frankenmint has quit [Ping timeout: 276 seconds]
skang404 has joined #bitcoin-wizards
raedah2 has quit [Remote host closed the connection]
raedah2 has joined #bitcoin-wizards
adnn_ has joined #bitcoin-wizards
adnn has quit [Ping timeout: 240 seconds]
[d__d] has joined #bitcoin-wizards
adnn_ has quit []
[d__d] has quit [Ping timeout: 264 seconds]
[d__d] has joined #bitcoin-wizards
TheSeven has quit [Disconnected by services]
[7] has joined #bitcoin-wizards
warren_ is now known as warren
ThomasV has quit [Ping timeout: 272 seconds]
veleiro has quit [Quit: Leaving.]
rusty2 has left #bitcoin-wizards [#bitcoin-wizards]
[d__d] has quit [Remote host closed the connection]
[d__d] has joined #bitcoin-wizards
<maaku>
bsm117532: you need more documentation about your braids project
<gmaxwell>
bsm117532: in bitcoin? unfortunately many orphans are dark
<gmaxwell>
mining equipment has rather high latency, often a block is orphaned by the time it is returned to its bitcoind
<gmaxwell>
those orphans never relay
<gmaxwell>
other ones might pass the first bitcoind but not make it far enough for you to observe them.
veleiro has joined #bitcoin-wizards
ThomasV has joined #bitcoin-wizards
jaromil has joined #bitcoin-wizards
so has joined #bitcoin-wizards
henrytill_ is now known as henrytill
henrytill has quit [Changing host]
henrytill has joined #bitcoin-wizards
Giszmo has joined #bitcoin-wizards
veleiro has quit [Quit: Leaving.]
so has quit [*.net *.split]
jaromil has quit [*.net *.split]
Ylbam has joined #bitcoin-wizards
jaromil has joined #bitcoin-wizards
jaromil has quit [Changing host]
jaromil has joined #bitcoin-wizards
so has joined #bitcoin-wizards
Aranjedeath has quit [Quit: Three sheets to the wind]
frankenmint has joined #bitcoin-wizards
ThomasV has quit [Ping timeout: 244 seconds]
frankenmint has quit [Ping timeout: 272 seconds]
nabu has joined #bitcoin-wizards
akoko has quit [Ping timeout: 260 seconds]
akoko has joined #bitcoin-wizards
ThomasV has joined #bitcoin-wizards
frankenmint has joined #bitcoin-wizards
frankenmint has quit [Ping timeout: 258 seconds]
Giszmo has quit [Ping timeout: 260 seconds]
stqism has quit [Quit: Like 3 fire emojis lit rn
coup_de_shitlord has joined #bitcoin-wizards
ThomasV has quit [Ping timeout: 246 seconds]
tripleslash has joined #bitcoin-wizards
dEBRUYNE has joined #bitcoin-wizards
gsdgdfs has quit [Ping timeout: 276 seconds]
frankenmint has joined #bitcoin-wizards
frankenmint has quit [Ping timeout: 244 seconds]
dEBRUYNE has quit [Quit: Leaving]
Guyver2 has joined #bitcoin-wizards
dEBRUYNE has joined #bitcoin-wizards
frankenmint has joined #bitcoin-wizards
tripleslash has quit [Ping timeout: 244 seconds]
CubicEarth has joined #bitcoin-wizards
Jeremy_Rand_2 has joined #bitcoin-wizards
nuke1989 has quit [Read error: Connection reset by peer]
nuke1989 has joined #bitcoin-wizards
Alanius_ is now known as Alanius
Anduck_ is now known as Anduck
MaxSan_ has quit [Ping timeout: 250 seconds]
CrazyTruthYakDDS has quit [Quit: Connection closed for inactivity]
<PeterR>
Lognormal fits the empirical data fairly well (although the tails were fatter than even lognormal).
sdaftuar has left #bitcoin-wizards [#bitcoin-wizards]
<PeterR>
Note that we were measuring only a single "hop."
mdavid613 has quit [Quit: Leaving.]
mdavid613 has joined #bitcoin-wizards
PeterR has quit [Ping timeout: 250 seconds]
<bsm1175321>
PeterR thanks, lognormal is no surprise, I'll add it as an option. Also it's the one-hop distribution I need, and I will sample it on each hop.
chjj has joined #bitcoin-wizards
ThomasV has quit [Ping timeout: 244 seconds]
ratbanebo has quit []
<katu_>
what is this shit about GFW, do they really drop bitcoin connections?[1~
<katu_>
to me the problem with china seems like their state telecoms dont peer with outside world[1;5C[1;5C
raedah has quit [Remote host closed the connection]
<katu_>
PeterR: the mechanism is crap and expensive HK bandwidth
<katu_>
because asia
go1111111 has joined #bitcoin-wizards
<bavly>
i have a wallet on bitcoin core . and although i am sure that some one sent to my addres bitcoins they dont appear in balance or even pending
<PeterR>
What I thought was interesting was that the GFC slowed down Xthin blocks by a smaller factor than it slowed down standard blocks.
<fluffypony>
bavly: ask in #bitcoin, this is for discussing theoretical cryptocurrency ideas
raedah has joined #bitcoin-wizards
gmaxwell has left #bitcoin-wizards [#bitcoin-wizards]
<katu_>
PeterR: there are workarounds of course, like partitioning the blocks among many peers and let them bypass the narrow gullet
<PeterR>
katu_: agreed. @thezerg is working on some similar ideas right now.
<katu_>
PeterR: very good sample data is torrents with high chinese peer presence. each individual peer can push like 5kb/s internationally (even if it gets fttx speeds on mainland). but bittorrent overcomes it easily, it simply talks to 200 peers.
<katu_>
anyhow, i'm not a big fan of hacks like that as they depend on highly diversified p2p network, which bitcoin sadly is not (compared to bittorrent).
JHistone has joined #bitcoin-wizards
<katu_>
i wonder what happened to blockheader-first and set reconcilation proposals
JHistone has quit [Client Quit]
<PeterR>
Did you see the Xthin results I just linked? Xthin blocks made it through the GFC about 10 times faster than standard blocks.
JHistone has joined #bitcoin-wizards
<katu_>
PeterR: i cant read the log graph very well, whats the mean avg?
<katu_>
my mean average latency from europe to most of china is 350ms, so it should be around that number.
<PeterR>
17.4 seconds (std. blocks) versus 1.8 seconds (Xthin)
<PeterR>
It's what's shown in the first figure in the article.
<PeterR>
The *median* time for Xthin through the GFC was 0.66 sec.
<PeterR>
The mean is a lot longer than the median because of the ~lognormal distribution.
<katu_>
thats pretty decent for a federated network
<PeterR>
Thanks. We were quite pleased with the results.
<katu_>
still, true bittorrent-like partitioning can get near true network latency provided you get sufficient amount of nodes on both sides
<PeterR>
Agreed.
<katu_>
(and i kinda like the idea of federated transport as an intermediary solution). though i'd organize your PR a bit differently.
<katu_>
ie directly ask pools for small cut according to measured orphan rate you reduced
blackwraith has joined #bitcoin-wizards
<PeterR>
Yes!
<PeterR>
I was brainstorming ideas like this yesterday.
<PeterR>
I think it could be made to work somehow...
<katu_>
anyhow, gmaxwell apparently hates your guts
<katu_>
so theres a chance core will be fixed to do something smarter than it currently does
<katu_>
destroying your business model in the process
<katu_>
its a risky enterprise :)
<katu_>
(but ultimately with good outcome for bitcoin either way, so kudoz to you :)
priidu has quit [Ping timeout: 272 seconds]
<PeterR>
Haha the only pay off I'm looking for is the smile on my face when that first 1.1 MB block gets mined.
<katu_>
wait, is this endeavor part of core/classic drama?
* katu_
had the impression both had the same slow blockprop
<PeterR>
Xthin has been part of Bitcoin Unlimited since March and is now part of XT too.
<PeterR>
There is an experimental branch of Classic that supports it, but the main Classic client does not (at the moment)
<katu_>
oh, you do primitive set recon
<katu_>
i thought you were merely a relay network
<katu_>
PeterR: any plans for true IBF epstein set recon?
raedah has quit [Remote host closed the connection]
blackwraith has quit [Ping timeout: 244 seconds]
<PeterR>
Personally, I think Xthin + subchains + optimistic mining (headers first) will give us all the scaling we need.
<katu_>
personally, i trust gmaxwell on this, optimistic mining assumes everyone behaves rationally within bitcoin closed system
<katu_>
but thats an assumption too weak
raedah has joined #bitcoin-wizards
<PeterR>
But optimisitic mining results in a small profitability advantage for a miner. How do you stop them from doing it?
<instagibbs>
katu_, I've become more pessimistic re:IBLT and the like
<instagibbs>
obviously mempool sync is nice
sausage_factory has quit [Read error: Connection reset by peer]
<instagibbs>
Matt has a new relay network up and running which does UDP+FEC for latency reduction(plus gossip for mempool). Something like that seems to be the best bet.
<instagibbs>
it's not p2p though, more work would be needed for that
<Taek>
you can solve the optimistic mining problem by allowing conflicting transactions to appear in a block (and ignoring them). This breaks SPV, but there are some latency based methods of fixing it
<instagibbs>
(you can't just relay data willy nilly, actually have to validate or get DoS banned)
<Taek>
for example, in the coinbase transaction you could include the merkle root of all valid transaction in the block 3 blocks back
<instagibbs>
Taek, I thought the DAG Days were over ;) another one
<Taek>
I'm pretty convinced that a dag construction is superior in most ways
<PeterR>
Still waiting to read Bob's paper :D
<Taek>
and I'm also convinced that, if SPV is a blocking factor, there are simple ways to bring it back
<instagibbs>
Taek, I do not believe a tree is the best thing, surely :)
<Taek>
I am curious what your doubts are
<bsm1175321>
SPV is straightforward if you can organize things into blocks from the DAG, that's what my "cohorts" are.
<instagibbs>
tree meaning the chain we have now
<Taek>
oh
<Taek>
even if you are using a chain, you can still enable safe opportunistic mining without breaking SPV
<instagibbs>
I think some sort of DAG like thing plus miners not being in charge of validation is best. But that's still, erm, unbaked
<Taek>
you just add a few blocks of latency
<Taek>
well, and it's a hardfork
<Taek>
unless you do 'segregatedtransaction' :)
<katu_>
instagibbs: simplest solution would be just to punish miners who intentionally break the network with selfish / "optimistic" mining. that way the dynamic system can be reduced to the most simple game theoretic model where everybody loses.
<katu_>
and the incentive is to simply build robust relaying.
<katu_>
sadly the punishment does not work that well in practice
<katu_>
miners still mine on spv pools etc :/
<PeterR>
katu_: I don't see optimisitic mining as a problem to be "solved" (I see it as a positive). What exactly are you worried about? That miners will stop validating *completely*?
<katu_>
PeterR: remind me what optimistic means again. mining on top of a tip you didnt verify "optimistically" ?
<PeterR>
It means you would start mining once you've received the block header and verified the PoW but before you've downloaded and verified all the transactions.
<katu_>
it simply breaks core assumptions that double spenders cant hijack hashpower for their attempt, not even temporarily. it still effectively boosts hashpower of attempted doublespend.
<PeterR>
You set a timer and STOP optimistic mining if you don't verify the complete block by the deadline
<katu_>
yes.
<katu_>
and that time period is hijacked hashpower
<katu_>
the longer the deadline, the worse it gets. i agree that short deadlinse (<100ms) wouldnt matter much
<PeterR>
Yes, that is a valid point. I'd like to see the math for this, however. I bet it comes out as insignificant for any reasonable optimistic-mining time period.
<katu_>
but past second it starts to feel uncomfortable
<katu_>
thats 1/600 of a big chinese pool for me, yay
<PeterR>
I was thinking about 6 sec (1%).
<PeterR>
But I don't have any math to back that up...just gut feel.
<PeterR>
Maybe I'll try to analyze this properly....
Sosumi has quit [Quit: Bye]
<katu_>
PeterR: granted, this concerns 1-conf transactions for the most part, but might break some more complex assumptions (ie parasitic chains, contracts)
<PeterR>
Yes.
<PeterR>
Ok got to get back to work. Thanks for the chat!
PeterR has quit [Quit: Page closed]
SyncYourDogmas has joined #bitcoin-wizards
<nicolagreco>
exit
nicolagreco has left #bitcoin-wizards [#bitcoin-wizards]
nicolag has joined #bitcoin-wizards
<nicolag>
can anyone help me understand Async Common Subset?
Church- is now known as Gilfoyle
edvorg has quit [Ping timeout: 260 seconds]
CrazyTruthYakDDS has joined #bitcoin-wizards
laurentmt has joined #bitcoin-wizards
raedah has quit [Remote host closed the connection]
afk11 has joined #bitcoin-wizards
raedah has joined #bitcoin-wizards
rusty2 has joined #bitcoin-wizards
frankenmint has quit [Remote host closed the connection]
frankenmint has joined #bitcoin-wizards
laurentmt has quit [Quit: laurentmt]
dEBRUYNE has quit [Quit: Leaving]
SyncYourDogmas has quit [Ping timeout: 246 seconds]
James_ has joined #bitcoin-wizards
dnaleor has quit [Quit: Leaving]
dEBRUYNE has joined #bitcoin-wizards
jannes has quit [Quit: Leaving]
zooko has joined #bitcoin-wizards
dnaleor has joined #bitcoin-wizards
<midnightmagic>
not really sure why rizun is welcome here given the fairly extreme provocation and promotion of cartoons of people being killed.
dingus has quit [Quit: later]
ThomasV has joined #bitcoin-wizards
<fluffypony>
I felt it prudent to keep quiet, katu_ was doing just fine in defense of secure system design :-P
Noldorin has joined #bitcoin-wizards
Guest10320 is now known as [Derek]
[Derek] is now known as Guest58413
Guest58413 is now known as [Derek
[Derek is now known as [Derek]
[Derek] has quit [Changing host]
[Derek] has joined #bitcoin-wizards
ThomasV has quit [Ping timeout: 276 seconds]
frankenmint has quit [Remote host closed the connection]
ThomasV has joined #bitcoin-wizards
Jeremy_Rand_2 has quit [Remote host closed the connection]
hashtag_ has joined #bitcoin-wizards
ThomasV has quit [Ping timeout: 252 seconds]
dnaleor has quit [Quit: Leaving]
dnaleor has joined #bitcoin-wizards
bavly has quit [Quit: Page closed]
frankenmint has joined #bitcoin-wizards
<dgenr8>
lognormal propagation is much simpler to work with than the alternatives i looked at. it's a bit hard to justify though - often it arises from a multiplicative sequence, and how does that model t(one network hop)?
rusty21 has joined #bitcoin-wizards
rusty2 has quit [Ping timeout: 258 seconds]
frankenmint has quit [Remote host closed the connection]
JackH has quit [Remote host closed the connection]
<dgenr8>
katu_: optimistic miners build empty blocks, so no help to double spenders
nonaTure has quit [Ping timeout: 252 seconds]
rusty21 has quit [Ping timeout: 240 seconds]
nonaTure has joined #bitcoin-wizards
frankenmint has joined #bitcoin-wizards
zooko has quit [Remote host closed the connection]
chjj has quit [Ping timeout: 244 seconds]
Davasny has quit [Remote host closed the connection]
<bsm1175321>
dgenr8: I have a network simulation for braids, latencies are calculated as distances on the surface of a sphere. I'm just inserting an additional delay drawn from a lognormal distribution to model the long-tails of real-world propagation.
<katu_>
dgenr8: the offending transaction is already embedded in the block prior on which optimistic miners blindly mine on, thus lending him more hashpower.
<katu_>
my assumption is they also blindly broadcast without waiting for the block to verify.
<katu_>
(sadly i couldnt find concise description of what xthin actually _does_, thus i'm assuming the worst)
chjj has joined #bitcoin-wizards
<katu_>
in case it witholds the block, attack surface is smaller, but still exists - winning hashes are semi-public for pools. it becomes non-technical debate whether or not individual miners would leak those, again, something im not entirely comfortable with.
MoALTz has quit [Quit: Leaving]
dnaleor has quit [Quit: Leaving]
Tiraspol has quit [Remote host closed the connection]
Tiraspol has joined #bitcoin-wizards
Tiraspol has quit [Changing host]
Tiraspol has joined #bitcoin-wizards
frankenmint has quit [Remote host closed the connection]
frankenmint has joined #bitcoin-wizards
dnaleor has joined #bitcoin-wizards
skyraider has joined #bitcoin-wizards
CrazyTruthYakDDS has quit [Quit: Connection closed for inactivity]
go111111111 has joined #bitcoin-wizards
go111111111 has quit [Remote host closed the connection]
JHistone has quit [Quit: Leaving]
murch has quit [Quit: Leaving.]
PeterR has joined #bitcoin-wizards
<PeterR>
katu_: there's a diagram and a brief description of how Xthin works in Fig. 1 here: