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
gill3s has quit [Read error: Connection reset by peer]
alferz has joined #bitcoin-wizards
shesek has joined #bitcoin-wizards
alferz has quit [Ping timeout: 244 seconds]
ielo has quit [Ping timeout: 264 seconds]
NewLiberty has quit [Ping timeout: 244 seconds]
alferz has joined #bitcoin-wizards
shesek has quit [Ping timeout: 240 seconds]
shesek has joined #bitcoin-wizards
NewLiberty has joined #bitcoin-wizards
shesek has quit [Ping timeout: 255 seconds]
sparetire_ has quit [Quit: sparetire_]
Tiraspol has quit [Remote host closed the connection]
Tiraspol has joined #bitcoin-wizards
jaekwon has quit [Remote host closed the connection]
priidu has quit [Read error: Connection reset by peer]
AaronvanW has quit [Ping timeout: 246 seconds]
shesek has joined #bitcoin-wizards
zooko has joined #bitcoin-wizards
zooko has quit [Remote host closed the connection]
miswism is now known as mistake
Tiraspol has quit [Read error: Connection reset by peer]
Dr-G has quit [Disconnected by services]
Dr-G2 has joined #bitcoin-wizards
Tiraspol has joined #bitcoin-wizards
Tiraspol has quit [Read error: Connection reset by peer]
Tiraspol has joined #bitcoin-wizards
zebedee has joined #bitcoin-wizards
c0rw1n is now known as c0rw|zZz
justanotheruser has quit [Ping timeout: 244 seconds]
kyuupichan has quit [Ping timeout: 244 seconds]
zebedee is now known as kyuupichan
<gmaxwell>
Wow. bitcoin-development has reached a new low.
<gmaxwell>
Recently Peter R posted a very long highly technical paper (with nice illustrations) which argued that there is an implicit economic block size limit which would result in the creation of a fee market as a result of orphaning.
<gmaxwell>
Peter R's had previously made similar arguments to a non-technical audience using an overly technical presentation (which was not fitting for the material or the audience)-- I felt his activities were intentionally obfscuating. And I responded correcting his technical errors and included a line calling him out on it. He responded at lenght complaining that I was trying to suprress his ability to
<gmaxwell>
communicate, and of course not addressing the technical remarks.
<gmaxwell>
A month or so later he published this paper. Learning from the poor expirence last time I, instead, contacted him in private. Hoping that he would revise or retract the incorrect aspects of his work, and that we could have a more productive interaction privately.
<gmaxwell>
His work failed completely, in my view, due to to unrelated mistaken assumptions on his part. I explained each in detail, including providing fairly concrete constructions for situations under which they would be violated.
<gmaxwell>
His response (to my read) suggested that he understood my position, and agreed _at least_ that they presented significant limitations on his work.
<gmaxwell>
He said that he would take a number of corrective actions.
<gmaxwell>
In the time since, he has posted publically about his work, promoting it, and making no corrective action; which I understand can just be the result of revisions taking time.
<gmaxwell>
A third party begain extension work, consulting with him privately-- and he failed to disclose my communicaton to this person. Their extension work fails due to the same mistaken assumptions in Peter R's work.
<gmaxwell>
I wrote Peter R again and complained that his lack of action and failure to communicate this to the other author has wasted their time.
<gmaxwell>
I then sent, without comment, the correspondance with Peter R to this other author.
<gmaxwell>
The other author is angry with Peter due to his misconduct, and wrote him complaining. After which Peter R just turned around and blasted me on the list as if I'd done something wrong.
<kanzure>
yeah peterr's email fails to mention that the disagreement was about the assumptions or the lack of idnetification of those assumptions, and instead thinks your original emails were about something else apparently
<kanzure>
he also seems to be confused about why you would be concerned about wasting people's time
<gmaxwell>
Af if the situation wasn't bad enough where people who benefited from PeterR's mistaken arguments, including ones whos own work contributed to their invalidation, have failed to speak up... I'm getting attacked for sharing a completely uninteresting techincal discussion with another author, whos work was undermined by Peter R's conduct, when he'd already written me that claiming that he intended
<gmaxwell>
to publish the substance of our discussion.
<gmaxwell>
So what did he expect me to do, go waste my time in another long 1:1 discussion with the other author who was mislead by his work?
<kanzure>
"trying to suppress his ability to communicate" deserves the typical lecture about how talking about vulnerabilities/problems is not suppression, etc.
<gmaxwell>
I was already considering just posting the discussion in public, due to his unresponsiveness in correction and his continued statements which seem to be in conflict with his apparent (to me?) agreement that I have a point.
<kanzure>
were his assumptions clearly stated and the obvious violations enumerated? if the answer's no then i think that should be clarified; there's no good reason to have wrong assumptions or even probably-wrong-assumptions silently failing to make an appearance in a research doc.
<kanzure>
looking at his bitcoin-dev email it looks like he is claiming that "simplifying assumptions are good and important" and not at all addressing your actual concerns "that assumptions should be explicitly stated especially if they might be wrong or simplified too much in certain directions that might cause problems in the resulting analysis". i think there's a difference.
Tiraspol has quit [Read error: Connection reset by peer]
<kanzure>
"Just because the proof makes simplifying assumptions, doesn't mean it's not useful in helping us to understand the dynamics of the transaction fee market." it helps to understand dynamics under those assumptions, but assumptions should be stated clearly. especially if they are possibly contentious anyway.... i think he's just conflating your claims with some other paranoid stuff.
Tiraspol has joined #bitcoin-wizards
<gmaxwell>
It's especially obnoxious that his last communication to me was "I'm going to take these corrective actions" but today, after yelling at him again in private his response was to say " I look forward to a white paper demonstrating otherwise!".
<gmaxwell>
kanzure: the simplifying assumption he makes that isn't disclosed in the paper is that miner centeralization isn't a concern or a consideration.
kyuupichan has quit [Read error: Connection reset by peer]
<kanzure>
he already agrees that his assumptions are simplified, so i don't see why he wants a whitepaper in reply. maybe just dump your email to pdf and format with latex, who cares.
kyuupichan has joined #bitcoin-wizards
Tiraspol has quit [Client Quit]
<gmaxwell>
The incorrect belief he has about relality is that block propagation time is necessairly linearly proportional to block size. He waxes on for two pages of equations and mumblb mumble shanon-hartley theorem. I responded with a simple protocol suggestion (which we've talked about here in the past) that makes it O(1) and not iblt-like O(1) but O(1) under much weaker assumptions.
<gmaxwell>
because I should be forced to waste hundreds of hours is a pretext battle with an anonymous party who has already demonstrated a remarkable lack of intellctual integrity on several occasions?
<kanzure>
this all seems pretty simple to fix by email in reply to him; if he prefers to think of himself as a victim after that then whatever nobody can stop him.
Tiraspol has joined #bitcoin-wizards
snthsnth has joined #bitcoin-wizards
<gmaxwell>
Not sure how. He's currently attempting to exploit that people are not going to spend their time reading the discussion and notice that at the end of he was agreeing to revise it his work, only to weeks later demand a paper from me... from something the anyone else needs only a two line correction for.
<gmaxwell>
fucking toxic enviroment.
<kanzure>
the agreement he gave you doesn't matter, what's more important is the original details (the assumptions that you wanted to point out so that others might not waste as much time). everything else is just peterr trying to reframe what you said.
jtimon has joined #bitcoin-wizards
PaulCape_ has quit [Quit: .]
PaulCapestany has joined #bitcoin-wizards
Tiraspol has quit [Read error: Connection reset by peer]
everyBloc has joined #bitcoin-wizards
Tiraspol has joined #bitcoin-wizards
<fkhan>
fwiw, after reading the actual discussion the context of peter's email seemed exaggeratted
<fkhan>
and, the miner block propogation scheme was something i hadn't heard proposed before
<fkhan>
i mean, the weaker assumption version
jtimon has quit [Ping timeout: 240 seconds]
<gmaxwell>
It's been discussed a couple times in the past in varrious guises, e.g. as a way to get faster soft-confirmation. I didn't expect PeterR to be aware of it (though he should have made clear at least that his proprtionality could be arbritarily small), thus the discussion.
<kanzure>
because it reads as "you made these comments to me"
<kanzure>
"reasonable to demand" -> "reasonable to insist", only you or i would use the word demand there :P
<kanzure>
otherwise seems fine
<kanzure>
whoops too late
<moa>
gmaxwell: "level centralisation" should be "level of centralisation" ... i think
<gmaxwell>
yea, sorry, I need to be more patient.
<gmaxwell>
My bad pratices are made worse by the recent norm of things showing up sensationalized on reddit before I can get a response out.
<moa>
take it easy
<moa>
that guy is making my blood boil with his demands for an equally vacuous 'white paper' to refute his 'conclusions' drawn from some seriously questionable assumptions
<moa>
it's like a new layer of high priests of economics is attempting to install themselves as 'system experts' or some such ....
<kanzure>
i don't think blood is supposed to boil
dEBRUYNE has joined #bitcoin-wizards
<moa>
I think the kids might say ... "dude, do you even code?"
<gmaxwell>
Well you can see Peter Todd blew up on the original post.
<gmaxwell>
It's infuriating when someone makes an intellectually weak argument, but slateers it in so much (very nicely constructed) pretext and trappings that people who aren't interested or lack the context to evaluate it on its merits are too busy being mesmorized with the leather binding. :)
<gmaxwell>
or at least it's infuriating if its done intentionally.
adam3us has joined #bitcoin-wizards
<dgenr8>
Dumb question. If actual block propagation requires 0 or almost 0 information xfer, why it there any need to limit the notional size of the block at all?
<moa>
gmaxwell: I was willing to give him the benefit of the doubt but I called him out on some of the obvious flaws and he only refutes what he can or ignores
<moa>
so it looks like it's a "made for public consumption" argument from leather-bound authority designed to sow doubt ... or he just doesn't want to confront big porblems
<gmaxwell>
dgenr8: Fine distinction; information must be transfered at _some time_, but not with the block. You could imagine instead that there was no block limit but was instead a global transaction limit. That would be equally sufficient, --- it's just that the having a global, consistent, transaction limit already implies a block limit.
<dgenr8>
do we need a global transaction limit?
hazirafel has quit [Remote host closed the connection]
NLNico has joined #bitcoin-wizards
<gmaxwell>
dgenr8: PeterR's argument hingest critically that block propagation is delayed from the moment the solution is found proportional to the information transmitted. He then argues that the existance of a "healthy fee market" (term of art, defined in his paper) unescapable; because miners must limit their blocks size to prevent those delays from costing them subsidy.
<gmaxwell>
His argument fails on several points (1) the more cenetalized miners are (under his model) the more fees they can collect, and at the limit of a single miner, his proposed mechenism results in an infinite sized blocks and fees of 0. He failed to consider miners could respond to high propagation costs by centeralizing.
<moa>
he makes zero attempt to show how "healthy fee market" --> security ... and ignores any questioning along that line, just for reference
jtimon has joined #bitcoin-wizards
<gmaxwell>
(2) the whole argument depends on relay times being proportional to block size (they're already not with matt's relay network protocol, but there is some weak relation: so I gave another example system which much more strongly eliminates the relationship).
<gmaxwell>
(3) He depends on the subsidy, his model becomes much less effective without it. And in general even without assuming the above to points, as moa notes the "market" his model produces doesn't neceessarily do anything useful. E.g. may be purely theoretical.
<gmaxwell>
(and would be increasingly so as subsidy falls away).
<gmaxwell>
So no biggie. I mean, I argue things less well advised than PeterR's all the time. Fortunately, in private discussion he agreed with me about most of the substance of my complaints.
<gmaxwell>
(He seems to not quite buy my argument for absolutely 0 proportionality, but he agrees that it can be very small and so not of pratical importance)
everyBlo_ has joined #bitcoin-wizards
everyBloc has quit [Ping timeout: 246 seconds]
<aj>
gmaxwell: fwiw, his analysis assumes the subsidy exists by assuming that a 0-size block is profitable when producing the "neutral profit curve" (M_supply(Q)) in eqn 6
<aj>
gmaxwell: at least, i think that's where it was last time i looked...
<gmaxwell>
aj: my impression when I read it initially was that getting rid of subsidy didn't break it, but it did make the concrete numbers so high as to be obviously unrealistic, even leaving the rest alone.
<aj>
gmaxwell: from the paper -- "We conclude by noting that the analysis presented in this paper breaks down when the block reward falls to zero. It suggests that the cost of block space is zero; ..."
<gmaxwell>
oh well indeed then!
<gmaxwell>
Though thats undone by a single altruistic participant that always pays a 1 satoshi fee transaction regardless of whats going on in every block. :)
jtimon has quit [Ping timeout: 244 seconds]
smellymoo has joined #bitcoin-wizards
<smellymoo>
fluffypony?
jtimon has joined #bitcoin-wizards
<phantomcircuit>
gmaxwell, i really want to respond to his last post on the mailing list with "let me explain to you what packets are"
dEBRUYNE has quit [Ping timeout: 265 seconds]
kmels has joined #bitcoin-wizards
<gmaxwell>
Should I respond further? I wrote a response... pointing out that basically every block on the network today is mined with a protocol that sends constant data when a block is found.
<gmaxwell>
it's ... offensive that he's trying to argue that what I'm suggesting is theoretical and not physically possible.
<phantomcircuit>
gmaxwell, im not sure there's any point; he's clearly got some kind of agenda and is refusing to admit the truth
superobserver has joined #bitcoin-wizards
<smellymoo>
tin foil hat
<gmaxwell>
phantomcircuit: thats the "bitcoin spinoffs" guy.
adam3us has quit [Quit: Leaving.]
<kanzure>
phantomcircuit: fwiw i would be interested in the "what are packets" answer :-)
<kanzure>
always good to cover the basics
<smellymoo>
lol
smellymoo has quit [Quit: see you.]
<gmaxwell>
phantomcircuit: pfft, don't make the SNR levels worse
<gmaxwell>
phantomcircuit: if you feel like causing trouble is important for your well being, send the guy an email about why he suddenly became so concerned about publishing private email threads now when he wasn't a week ago? :)
<gmaxwell>
man, at least I only sent to to a single person the guy was actively screwing over. :-/
<phantomcircuit>
gmaxwell, heh
CodeShark_ has quit [Ping timeout: 255 seconds]
NewLiberty has quit [Ping timeout: 246 seconds]
Giszmo has quit [Quit: Leaving.]
davec has quit [Read error: Connection reset by peer]
davec has joined #bitcoin-wizards
AlphaTech is now known as TechnoLithium
TechnoLithium is now known as AlphaTech
AlphaTech is now known as HornyTech
HornyTech is now known as AlphaTech
AlphaTech is now known as pre-AlphaTech
pre-AlphaTech has quit [Disconnected by services]
AlphaTech has joined #bitcoin-wizards
Cory has quit [Ping timeout: 260 seconds]
Pasha has joined #bitcoin-wizards
Pasha is now known as Cory
AlphaTech is now known as Agent_Zach
Agent_Zach is now known as AlphaTech
<ryan-c>
will a transaction propagate through a node if that node has not seen the output the transaction spends?
<phantomcircuit>
ryan-c, no
justanotheruser has joined #bitcoin-wizards
priidu has joined #bitcoin-wizards
arubi_ has joined #bitcoin-wizards
snthsnth has quit [Ping timeout: 256 seconds]
ThomasV has joined #bitcoin-wizards
Huxy- has joined #bitcoin-wizards
ebfull has quit [Remote host closed the connection]
joesmoe- has quit [Ping timeout: 244 seconds]
Huxy has quit [Ping timeout: 244 seconds]
s1w has quit [Ping timeout: 244 seconds]
berndj has quit [Ping timeout: 244 seconds]
Madars has quit [Ping timeout: 244 seconds]
Madars has joined #bitcoin-wizards
joesmoe has joined #bitcoin-wizards
jtimon_ has joined #bitcoin-wizards
s1w has joined #bitcoin-wizards
berndj has joined #bitcoin-wizards
kmels has quit [Ping timeout: 252 seconds]
s1w is now known as Guest57542
jtimon has quit [Ping timeout: 250 seconds]
Guest57542 has quit [Changing host]
Guest57542 has joined #bitcoin-wizards
Guest57542 is now known as s1w
AaronvanW has joined #bitcoin-wizards
c-cex-yuriy has joined #bitcoin-wizards
frankenmint has joined #bitcoin-wizards
[1]psztorc has joined #bitcoin-wizards
psztorc has quit [Ping timeout: 260 seconds]
[1]psztorc is now known as psztorc
psztorc is now known as Guest55765
AaronvanW has quit [Ping timeout: 246 seconds]
hazirafel has joined #bitcoin-wizards
jtimon_ has quit [Ping timeout: 255 seconds]
hazirafel has quit [Ping timeout: 272 seconds]
AaronvanW has joined #bitcoin-wizards
blackwraith has joined #bitcoin-wizards
priidu has quit [Ping timeout: 250 seconds]
Guest55765 has quit [Ping timeout: 260 seconds]
SDCDev has quit [Ping timeout: 256 seconds]
NLNico has quit [Quit: Leaving]
SDCDev has joined #bitcoin-wizards
sparetire_ has joined #bitcoin-wizards
mistake is now known as prosodyVeContext
dEBRUYNE has joined #bitcoin-wizards
bedeho has quit [Ping timeout: 265 seconds]
ThomasV has quit [Quit: Quitte]
Quanttek has joined #bitcoin-wizards
polyclef has quit [Quit: polyclef]
Casper- has joined #bitcoin-wizards
Mably has joined #bitcoin-wizards
polyclef has joined #bitcoin-wizards
polyclef has quit [Client Quit]
LeMiner2 has joined #bitcoin-wizards
LeMiner has quit [Ping timeout: 246 seconds]
bildramer has quit [Ping timeout: 272 seconds]
bildramer has joined #bitcoin-wizards
ThomasV has joined #bitcoin-wizards
moa has quit [Quit: Leaving.]
dEBRUYNE has quit [Ping timeout: 246 seconds]
Dr-G2 has quit [Read error: Connection reset by peer]
optimator has quit [Read error: Connection reset by peer]
Dr-G has joined #bitcoin-wizards
Dr-G has joined #bitcoin-wizards
optimator has joined #bitcoin-wizards
[Derek] has quit [Ping timeout: 240 seconds]
[Derek]afk has joined #bitcoin-wizards
[Derek]afk is now known as [Derek]
[Derek] has quit [Changing host]
[Derek] has joined #bitcoin-wizards
ThomasV has quit [Ping timeout: 256 seconds]
blackwraith has quit [Read error: Connection reset by peer]
jtrag has joined #bitcoin-wizards
jtrag has left #bitcoin-wizards [#bitcoin-wizards]
ThomasV has joined #bitcoin-wizards
priidu has joined #bitcoin-wizards
SDCDev has quit [Read error: Connection reset by peer]
SDCDev has joined #bitcoin-wizards
SDCDev has quit [Changing host]
SDCDev has joined #bitcoin-wizards
xabbix has joined #bitcoin-wizards
xabbix has joined #bitcoin-wizards
ThomasV has quit [Ping timeout: 265 seconds]
[Derek] has quit [Remote host closed the connection]
[Derek] has joined #bitcoin-wizards
[Derek] is now known as Guest65439
Guest65439 is now known as [Derek]
[Derek] has quit [Changing host]
[Derek] has joined #bitcoin-wizards
mkarrer has joined #bitcoin-wizards
belcher has joined #bitcoin-wizards
spinza has quit [Excess Flood]
c0rw|zZz is now known as c0rw1n
spinza has joined #bitcoin-wizards
roxtrongo has joined #bitcoin-wizards
paveljanik has joined #bitcoin-wizards
Guyver2 has joined #bitcoin-wizards
gill3s has joined #bitcoin-wizards
Casper- has quit [Read error: Connection reset by peer]
mkarrer has quit [Remote host closed the connection]
c-cex-yuriy has quit [Quit: Connection closed for inactivity]
azariah has quit [Remote host closed the connection]
azariah has joined #bitcoin-wizards
Guest77360 has joined #bitcoin-wizards
dEBRUYNE_ has joined #bitcoin-wizards
dEBRUYNE has quit [Ping timeout: 246 seconds]
dEBRUYNE_ has quit [Read error: Connection reset by peer]
dEBRUYNE__ has joined #bitcoin-wizards
mjerr has joined #bitcoin-wizards
dc17523be3 has quit [Ping timeout: 264 seconds]
dc17523be3 has joined #bitcoin-wizards
ThomasV has quit [Ping timeout: 255 seconds]
dc17523be3 has quit [Ping timeout: 246 seconds]
dc17523be3 has joined #bitcoin-wizards
jaekwon has joined #bitcoin-wizards
jaekwon has quit [Remote host closed the connection]
jaekwon has joined #bitcoin-wizards
frankenmint has quit []
belcher has quit [Ping timeout: 250 seconds]
<kanzure>
er, why don't we have various web-of-trust reputation stuff? was there a good reason for not having this yet?
<jcorgan>
can u be more specific?
arubi_ has quit [Quit: Leaving]
<kanzure>
you know, a drop-in reputation library based on web-of-trust consumption
<jcorgan>
i myself am very skeptical of the whole concept; i don't think trust is transitive
bedeho has joined #bitcoin-wizards
jaekwon has quit [Remote host closed the connection]
belcher has joined #bitcoin-wizards
roxtrongo has joined #bitcoin-wizards
ThomasV has joined #bitcoin-wizards
Guest77360 is now known as amiller
amiller has quit [Changing host]
amiller has joined #bitcoin-wizards
<amiller>
kanzure, i have been thinking about that a lot, its tricky though
<kanzure>
is it missing a good use case? is that why?
<amiller>
i think trust is transitive but not all versions of trust are the same, and any application needs to be precise about them
<amiller>
just to give an example of how crude it is for now... i like showing people the Bitcoin-otc web of trust of one of these working OK in practice
<amiller>
ok but the meaning of the edges, represented by a number 1-10, is pretty hard to understand and unlikely to be unifrom
<null_radix>
so its basically like the trustDavis scheme?
King_Rex has joined #bitcoin-wizards
<amiller>
ah trustDavis is fascinating and is a bunch more complicated!
<amiller>
they have some similar underlying principles like representing trust as edges on a graph and getting an 'inference' as some flow computation on it
<null_radix>
ah right
<amiller>
okay so if you look at the guide for bitcoin otc it has some fuzzy descritions, like you sohuld give someone a 1 if you have one good trade, and 10 if they are your mother or osmething
<amiller>
obviously people differ in how they decide things like that
<amiller>
nanotube was working on a more detailed description
<amiller>
that was quantified
<amiller>
like how much % of national gdp wouldyou trust them with
<amiller>
but still there's a big diference between possible meanings there, like trust them to repay it on demand? trust htme to repay it eventually but they may take their time? trust them to never repay it, but invest it wisely for themselves?
<null_radix>
nanotube?
<kanzure>
nanotube the bitcoin-otc and lesswrong person
<null_radix>
ok; google is showing lots of stuff on carbon nanatubes :P
<kanzure>
amiller: i have no good answers for that. there's an ontology problem and also a grounding problem there...
<amiller>
so, you don't need to assign a meaning like that if there is some consequence
<kanzure>
iirc the reason i brought it up today was for some reason like "what about a storage network where people can volunteer their storage, automatically sign up to be part of the collective (no proof-of-storage involved), and their ability to join is judged based on reputation regarding their other reputational decisions in other similar networks that they have joined and provided services to".
<kanzure>
and there's probably payment-related reasons i guess.. but a lot of payment stuff can be done without web-of-trust so there's much less motivation from that.
metamarc has quit [Ping timeout: 246 seconds]
ThomasV has quit [Ping timeout: 240 seconds]
King_Rex has quit [Remote host closed the connection]
kang_ has joined #bitcoin-wizards
afk11 has joined #bitcoin-wizards
<kang_>
From yesterday, "gmaxwell - Information must be transfered at _some time_, but not with the block." in response to "If actual block propagation requires 0 or almost 0 information xfer, why it there any need to limit the notional size of the block at all?". So how is correctness of an SPV proof verified?
jtimon has joined #bitcoin-wizards
<jcorgan>
kanzure: my skepticism about web-of-trust models is that, while I might judge you completely trustworthy for X, i might also think you are clueless about how much you trust others about X. so a strong edge from me to you and a strong edge from you to someone else does not mean i trust that other person in the slightest.
ThomasV has joined #bitcoin-wizards
frankenmint has joined #bitcoin-wizards
<kanzure>
jcorgan: sure; i wouldn't be able to trust my own previous trust lines anyway because who knows what prior scheme i was using or whether i was accurately running that scheme at all? etc.
luny has quit [Read error: Connection reset by peer]
luny has joined #bitcoin-wizards
<jcorgan>
a more accurate distillation of "reputation" might be, "how much do people i think are good judges of character rate this person"
<jcorgan>
and that web is only 1 deep
<jcorgan>
and even then it becomes only a sort of bayesian prior to set initial expectations
luny has quit [Read error: Connection reset by peer]
luny has joined #bitcoin-wizards
frankenmint has quit [Remote host closed the connection]
Quanttek has quit [Ping timeout: 250 seconds]
nwilcox has joined #bitcoin-wizards
<amiller>
jcorgan, kanzure, i think that a reputation system should have some transaction output, not just informatio
<amiller>
that way it may be possible to make an incentive argument that shows accurate trust reporting is incenitve compatible
<amiller>
that is a key principle of trustdavis
<amiller>
when you tell the system you trust someone, you are actually putting up some money as insurance collateral for them
<amiller>
if you trust the wrong people, you can lose the colatearl
<kanzure>
how do you lose the collateral exactly?
<kanzure>
a bunch of people collude to take it...?
trippysalmon has joined #bitcoin-wizards
dEBRUYNE__ has quit [Ping timeout: 255 seconds]
<jcorgan>
it seems a little hand-wavey but i agree that opinions become more believable when something is at stake
<jcorgan>
it still wouldn't address the non-transivity of that opinion
ThomasV has quit [Ping timeout: 255 seconds]
bigreddmachine has joined #bitcoin-wizards
CodeShark has joined #bitcoin-wizards
bigreddmachine has quit [Client Quit]
kang_ has quit [Quit: Page closed]
jaekwon has joined #bitcoin-wizards
jaekwon has quit [Remote host closed the connection]
everyBlo_ has quit [Remote host closed the connection]
everyBloc has joined #bitcoin-wizards
hazirafel has joined #bitcoin-wizards
hazirafel has quit [Remote host closed the connection]
arubi has quit [Quit: Leaving]
snthsnth has joined #bitcoin-wizards
arubi has joined #bitcoin-wizards
jtimon has quit [Ping timeout: 240 seconds]
Quanttek has joined #bitcoin-wizards
dEBRUYNE__ has joined #bitcoin-wizards
nwilcox has joined #bitcoin-wizards
DougieBot5000_ has joined #bitcoin-wizards
mkarrer has joined #bitcoin-wizards
DougieBot5000 has quit [Ping timeout: 256 seconds]
smk has joined #bitcoin-wizards
dEBRUYNE__ is now known as dEBRUYNE
everyBloc has quit []
chmod755 has joined #bitcoin-wizards
afk11 has quit [Ping timeout: 260 seconds]
DougieBot5000_ has quit [Ping timeout: 264 seconds]
paveljanik has joined #bitcoin-wizards
tripleslash has joined #bitcoin-wizards
tripleslash is now known as [\\\]
mjerr has quit [Ping timeout: 260 seconds]
Dizzle has joined #bitcoin-wizards
<polyclef>
web of trust is particularly prone to poisoning via the Dunning Kruger effect. non expert users don't know enough to rate others properly
<polyclef>
and expert users might reasonably trust no one
DougieBot5000 has joined #bitcoin-wizards
mkarrer__ has joined #bitcoin-wizards
mkarrer has quit [Ping timeout: 240 seconds]
ThomasV has joined #bitcoin-wizards
Dizzle has quit [Remote host closed the connection]
bigreddmachine has joined #bitcoin-wizards
<polyclef>
possible test data: look at who lost money in gox and see what percentage have high trust ratings from others.. not perfect, because for most there's a difference in risk tolerance with ones own money vs what they might recommend to others, but it is likely instructive
<nanotube>
so you could trust someone a lot for say, food/movie recommendations, but not much for financial transactions. you can trust someone a lot personally, but not give credence to their assignments of trust to others, etc.
<nanotube>
one problem with any public trust network like that though, is the likely complete loss of anonymity. a person can be pretty well specified by the people they trust and are trusted by.
<kanzure>
yes yes but then you get into ontology sync issues like "the other person calls food something entirely different"
<kanzure>
"oops your word for high-security-thing is what i consider to be low-security-thing" and other catastrophes
<nanotube>
at which point you don't trust them for food. :)
<kanzure>
re: storage service providers, web-of-trust seems like a possibly okay way to subscribe providers for a large-scale distributed storage service, where you use web-of-trust instead of formal review for the introduction of random participants with random hardware.
<nanotube>
nothing is going to be perfect. the point is to just provide some more useful decision-support info. presumably some useful information can be derived if say, 10 of my friends trust $guy for $100, then probably i'm safe to transact with $guy up to $100.
<Taek>
"nothing is going to be perfect" -> but you do need to make sure that it's not highly exploitable in some way
priidu has quit [Read error: Connection reset by peer]
<nanotube>
true enough
<kanzure>
re: trusting people with specific amounts, there's a bit of a gap that i wouldn't trust some people with, but the person who i would trust $10 with and not $10k with might be someone i trust $1B with :-)
Quanttek has quit [Ping timeout: 256 seconds]
<nanotube>
kanzure: haha well if you have highly unusual utility functions, you should make that known on your profile when you assign such trust, then. :)
<kanzure>
nanotube: one way to solve the ontology/tagging/word meaning problem is something like this: whoever introduces a new term has to also be the "central authority" by which that term will be evaluated. you can do some centralization tricks with web-of-trust like this, i think.
<nanotube>
could you give me an example of something like that?
priidu has joined #bitcoin-wizards
<nanotube>
(the non-linear monetary trust thing, that is)
<kanzure>
well it's just context stuff, it's not a raw discontinuous utility function on its own?
<kanzure>
it's not good to assume that a utility function is going to be smooth as you keep adding dimensions (other context)
<kanzure>
i don't even know what a smooth 8-curve looks like
<nanotube>
i just mean if you trust someone for $X, but wouldn't trust them for $X-positivenumber, it seems ... unusual and requires description. :)
<nanotube>
i keep thinking of things in terms of monetary values, because that's what otc is about. certainly some higher-dimensional things like food flavors or movies are not ordinal like that
King_Rex has joined #bitcoin-wizards
<polyclef>
instead of an identity heavy approach, something that uses anonymized past transactions might be better. after a successful transaction, sign an pseudonymized token that indicates rating and tx size. sibyll attacks are possible with a trivial version, but I think this has less issues than an identity focused approach. so instead of saying 10 of my friends trust X for up to $100, instead use 10
<polyclef>
people I know have all done transactions that they didn't have cause to complain about with range 10-200, avg 145
rusty has joined #bitcoin-wizards
kmels has quit [Ping timeout: 246 seconds]
andytoshi has joined #bitcoin-wizards
zooko has joined #bitcoin-wizards
mkarrer__ has quit [Remote host closed the connection]
Guyver2 has quit [Quit: :)]
spinza has quit [Excess Flood]
snthsnth has quit [Ping timeout: 250 seconds]
Dizzle has joined #bitcoin-wizards
spinza has joined #bitcoin-wizards
snthsnth has joined #bitcoin-wizards
spinza has quit [Excess Flood]
Quanttek has joined #bitcoin-wizards
spinza has joined #bitcoin-wizards
snthsnth has quit [Read error: Connection reset by peer]
ThomasV has quit [Ping timeout: 268 seconds]
rusty has quit [Ping timeout: 244 seconds]
snthsnth has joined #bitcoin-wizards
<nanotube>
polyclef: how would you avoid identifiability of individuals who participated in transactions?
<nanotube>
some kind of ring signatures scheme?
kmels has joined #bitcoin-wizards
bigreddmachine has quit [Remote host closed the connection]
moa has joined #bitcoin-wizards
Quanttek has quit [Ping timeout: 264 seconds]
chmod755 has quit [Quit: Ex-Chat]
AnoAnon has joined #bitcoin-wizards
AnoAnon has quit [Max SendQ exceeded]
rusty has joined #bitcoin-wizards
rusty has quit [Ping timeout: 246 seconds]
gill3s has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
adam3us has quit [Quit: Leaving.]
Mably has quit [Ping timeout: 250 seconds]
kmels has quit [Ping timeout: 246 seconds]
bildramer has quit [Quit: Chi mai dell'Erebo fra le caligini, sull'orme d'Ercole e di Piritoo conduce il pié?]
bildramer has joined #bitcoin-wizards
belcher has quit [Ping timeout: 240 seconds]
zooko has quit [Ping timeout: 246 seconds]
jouke has quit [Ping timeout: 264 seconds]
jouke has joined #bitcoin-wizards
belcher has joined #bitcoin-wizards
Dizzle has quit [Quit: Leaving...]
nwilcox has quit [Ping timeout: 264 seconds]
kmels has joined #bitcoin-wizards
nwilcox has joined #bitcoin-wizards
nwilcox has quit [Ping timeout: 265 seconds]
<MRL-Relay>
[tacotime] yeah, that works fine (ring signature with key image)
rusty has joined #bitcoin-wizards
roxtrongo has quit [Remote host closed the connection]
rusty has quit [Remote host closed the connection]