#tlsnotary-chat

very interesting thought experiment: what if real gold suddenly became bitcoin-like: it could be sent over the internet. would people decide it was a horrible kind of money because mining was too centralized?

to be fair, a lot of people already do think gold mining is quite evil.

so not necessarily being inconsistent.

if "people" really cared about centralization, then Bitcoin would have already been hugely successful

btw waxwing do you start using american english spellings?

oakpacific: it's natural, as a british person who spends a lot of time on the internet

and little or no time in england :)

i just switch it around, i'm not comfortable with either anymore

like satoshi :) i think many brits, australians, canadians, irish are the same.

right you mention it, that's funny, if he was not satoshi, nobody would have thought his spellings would have meant anything

like it rightfully doesn't

yes, but it probably does indicate a non-american

ockham/occam's razor and all that

re: ockham/occam omg spelling is so annoying

waxwing: please check the draft in the mailbox to help me proofreading the email

like that US general :)

waxwing: they require the message to remian privileged so i can't post it in a public place

i was joking, it's the best way.

it looks fine but it does seem like you relegated self-test to an after-thought rather than the main thing.

waxwing: the problem is i don't know what form of demo he really has in ming

or what does he really mean by demo

i mean maybe he just wants to watch a video

replace bank-profile/teacher qualification with "bank page or other ID proof page". it looks a bit bizarre referring to teacher out of context.

which we already have

well, the video's a bit crappy so i wouldn't emphasize it. but it's there if they want it.

waxwing: can you make changes on the fly?

yeah i could. let me do that.

but the english looks fine

well, although there are typos

oakpacific: please save your original version, i am going to propose a different one.

i saved an update oakpacific

it's significantly different, so if you prefer your version, i hope you saved it :)

oakpacific: i'm afk for a little while, please go ahead and send whichever version you feel comfortable with.

waxwing: sorry was away

oakpacific: sorry was also away :)

sent the email after adding a few modifications to your version

btw, but the login was from cyrpus, are you sure it's all right?

hmmm...never mind

yes that was me. no need to say more :)

mkarrer, are you still excited about nubits? Last time I checked, it hinged on the assumption that the stake holders will do the right thing in order to maintain the 1:1 ration. Is it still the case?

I have a hard time putting my faith in the arm of flesh.

yeah, although cautious to dismiss it, i also have that problem with it. i think the system described by robert sams makes more sense.

waxwing, what's the robert sams paper link?

so i read through the dark mail specification, and i still can't figure out whether they use ECDH or not for e2e key exchange

dansmith_btc: look up seignorage shares

i linked it here yesterday, it's somewhere...

ok, I see it in logs

ah yeah you got it

dansmith_btc: one dev of them had a presentation yesterday at the meetup. they are working on a kind of subscription system, so voters who dont have enough info/time can delegate their vote to a trusted member of the community.

i think it is still a lot in progress and an experiment, but until yet they could hold the peg.

waxwing, while I appreciate you pointed to the paper, all I had to do is read the last sentence "But all of these suggestions for tackling problem 1 are speculative and I have no convictions here whatsoever. This is a hard problem"

dansmith_btc: he's talking about the problem of getting data trustlessly from outside the network. so yeah, if that's what you're focusing on, the paper isn't so interesting.

what i found interesting was the market mechanism for adjusting price using the auction.

the possible solution for the "hard" problem involving schelling points is interesting too, though.

vitalik buterin did a nice write up of it on his blog.

i didn't mean to suggest it as a perfect system, but i think it's more interesting than solutions involving something like democratic consensus.

ok, I agree that's an interesting scheme of adjusting, however I always go for the hard problem first - if it cannot be solved, I'm not as eager to research further.

right, well that's a fair enough perspective.

which leads me to the next Q: has anyone here looked into truthcoin/augur aka decentralized prediction market. If this thing succeeds, it will obsolete stuff like nubits.

I think mkarrer, if you're looking for a project to further the decentralization ideas, I'd suggest looking into truthcoin and help it mature.

dansmith_btc: thanks. still fighting to find a way how we can continue with bitsquare

dansmith_btc: does truthcoin avoid the issue of trust in getting data from outside the blockchain?

waxwing, truthcoin idea is that the human oracles will be financially incentivize to provide the correct feed.

Just like bitcoin miners are incentivized to adhere to the protocol or risk being left out.

right, it's still using that kind of schelling idea. vitalik pointed out that bitcoin is really schelling consensus for transaction ordering.

I'm not saying that truthcoin is end-all-be-all. Moreover, I really dont like truthcoin having to release its own coin.

Yes, exactly it's the same idea in truthcoin.

in all the proposals we're describing it's like that: there is no crypto guarantee of correct data, just that being outside consensus is financially disincentivised.

yes, exactly. however, there is also a risk of centralization - gravitating to a single initially-trusted feed and being afraid of deviating from that feed.

Then the feed owner can go rogue and everyone will agree to the rogue data because of fear of not going with the consensus.

However this is solvable. It's just a theoretical attack akin to the bitcoin's 51% attack.

right. that's what most worries me about these systems. there's incentive to cooperate, but in unusual circumstances you could have a "phase transition" to a new equilibrium and then the whole thing gets busted. still it's a very interesting technological concept(s).

dansmith_btc: i dont think 51% attack is soluble. it's intrinsic to the system. good thing with bitcoin is, it doesn't destroy the whole system at once.

waxwing, would you agree that the big picture here is that if sth like truthcoin succeeds, we can have a colored USD coin pegged to the truthcoin's USD feed. This way we will have high-frequency btc<->usd exchanges on the blockchain.

right, something along those lines.

what i'm keen to see though is that such a system is robust rather than brittle.

i.e. it may fail a bit, but there's negative rather than positive feedback if that happens.

i mean, there's so many of these things it's hard to keep up. i tried to read up on bitusd, but it didn't look right to me. it seemed to be essentially a trust based system, even though it clearly would work under normal circumstances.

actually i retract "clearly". i can barely remember how it works now.

yes, I agree, it is hard to keep up here.

Intellectual honesty is important in this new space. People are running to and fro trying to allign themselves to the next 2.0 project. They choose a project, then become emotionally, financially attached to it and lose critical thinking.

i become so emotionally attached to TLSNotary, that I started to favor RSA over ECDH, and TLS 1.1 over all other vers

speaking of which oakpacific, what if we step up the publicity campaign and pressure bitstamp and the likes to look into tlsnotary.

or at least solicit a response from them on the matter.

well, the course of action we are following can be seen as first steps toward that goal

especially at a crisis like this when they scramble to save face.

if we can convince tether.to ,then it's like bitfinex is doing it

at that time, bitstamp would have little excuse

i was always lukewarm about the idea because a big company is not going to give *all* its users the chance to read its fiat account, so if it ends up being trusted auditors anyway, tlsnotary doesn't add that much advantage. but i think, introducing it into the conversation doesn't hurt.

the small problem may be that bitstamp's cop-out would be that we are not an established company so they cant have business with us.

I was thinking along the lines of starting our own company TLSNotary Inc which will be the auditor for all those exchanges.

yeah, this is a variant on the "tlsnotary as centralized web service" idea.

that's a bit dangerous, it's like we get out of the shadow, and march against the banks

it doesn't have to be seen as "against banks", especially in the audit function. rather than the trade function.

okay

once you get into the big corporate level view of it, though, banks do have functions that allow third party verification. they're not used much, but they do exist.

oakpacific, bitstamp may gets a special permission from their bank to be allowed to be audited in such a way.

but most likely govt regulators will see us as any other auditing firm and we just can't get all the accreditations

well, there are so many business-friendly places to register a firm in.

i suspect there's some middle ground, but it's hard to pinpoint it.

like the isle of man, maybe

well, we wouldn't say our authentication results have any legal power anyway

something like the ardeva model or bitreserve model , where they want new, innovative methods outside the norm, but at the same time they're a centralised service because ordinary users find this stuff intimidating.

those guys want it packaged for them, and they want exclusivity if they can get it, too.

problem is we started from a very p2p view and built it in that way. not a centralized, closed package.

wrt tlsn as a webservice - we could start small and provide a free service where folks could notarize their pages with our service's digital signature on it.

waxwing: all i want from these corps are reps so i don't see where is the problem

it's like the p2p goal is doomed if some centralized services use TLSNotary

oakpacific: bootstrapping right

if we have some orgs we can point to and say "look they trust tlsnotary so why shouldn't you", things would be much easier

yeah quite. i think we're pretty much all in agreement on this.

*it's not like the p2p goal is doomed

ah that makes more sense. lol.

also, since people are gonna use these services anyway, helping them is still a worthwhile....albeit not so enjoyable activity

dansmith_btc: good idea

oakpacific: remember your funny acronym? the NSA got there first: http://www.spiegel.de/media/media-35670.pdf

or gchq rather

waxwing: we are being watched

gchq should be calling it BADARSE really...

lol

so they are off watch recently and haven't got waxwing's memo to use queen's english exclusively

so they're using mobile ads to track people? i'm shocked, shocked !

from http://www.spiegel.de/international/world/new-s...

seems like a new set of docs, i think

i am sorry but this one is boring, despite the name

holy shizzle

SPARTY is a backdoor in the ssh daemon for *nix based on openssh prtable. it allows a public key to be embedded in the sshd binary and will then always grant a root login shell if presented with the proper key pair for that key. In other words, it behaves as if the given key is in ~/.ssh/authorized_keys

it goes on from there, but img so can't copy

wtf are the last lines?

the BCS is dead?

yeah

i would tell you, but then i'd have to kill you

i loved the fact that it had letters of appreciation

waxwing: what if i told you, it means the Bardeen-Cooper-Schrieffer theory has been superseded

waxwing: well i think i have a clue why this doc looks like this

i guess they have been recruiting exceptionally brilliant system intruders, who would usually not write their reports in a very conformative way

so openssh is basically an open door?

huh they have a "DSD wiki"?

yeah why not, i saw them refer to wikis in other places.

i don't think it means that

i remember applebaum got a lot of flack for saying 'ssh is broken' at CCC. Maybe this is what he meant.

i don't think it's backdoored that way

the dude talks about a lot of exploits