log☇︎
88200+ entries in 0.597s
ben_vulpes: dafuq is a favicon anyways
mircea_popescu: out of about thirty studios, twenty five respond, within a day mp has one dozen amply adequate locations for a casting.
mircea_popescu: aww! wooden inside is not from a good lyfe.
mircea_popescu: much to alf's relief, it's not a buncha baluba.
mircea_popescu: (there was a cold snap, and that place drew just about 19 amps consistently for days, i'm pretty sure)
mircea_popescu: so yes, a+b+c, pretty stacked deck.
mircea_popescu: yes, pissing self is common in a) women, especially if well used (in practice, this is often seen around childbirth) ; b) scared people, especially if young ; c) electricity mishaps, especially if high voltage.
mod6: it looked as if someone lit off a M80 in her hand.
asciilifeform: why use a medium where enemy can inject.
asciilifeform: mircea_popescu's scheme answers 'E infers key, floods A with megatonne of shit' with 'A renegotiates key with all of A's peers via magical channel that E cannot saturate with shit'
asciilifeform: and mircea_popescu's favorite rhetorical device is 'this is not relevant.' hey, it's the man's own blog, who am i to say that he has to consider my observations relevant. but this nonsense won't keep rainwater out of a tent, much less ft meade.
asciilifeform: (picture a weapon that kills the same POPULATION PERCENTAGE of the firer's army as of the enemy. it would be quite useless. any weapon requires asymmetry to work.)
asciilifeform: E can ding A and B for less cost to himself than they incur from said ding. which is the elementary particle from which ddos is made.
asciilifeform: as it stands, the cost to E of replaying any or all of the packets between A and B, or reordering, is 0.
mircea_popescu: so far this isn't a shared view.
asciilifeform: and came to conclusion that there is a massive hole in the scheme sketched out in subj link, that i can drive a maersk cargo liner through.
mod6: when you think it's a good idea, we can add a project to tb0t if it seems prudent.
mod6: its a good dialogue though.
mircea_popescu: not even just how long. it's just a sort of coupling. "these two have a solution to us army general problem"
Framedragger: " 'how long node a and b have been in communion'" - right, that i can understand, that it's not good
asciilifeform: especially not a useful tidbit such as 'how long node a and b have been in communion'
mircea_popescu: Framedragger understand what the problem is here. it's not that you give away the secret number 40084059534. it's that you give away the fact that you and X both somehow live in a world where ++ means something.
Framedragger: asciilifeform: tho am not sure if it's necessarily bad, to leak incremental nonces. i mean, i know you don't want to leak one single useful bit to da enemy; i don't know if it's a practical constraint, even if it is laudible.
a111: Logged on 2016-09-09 15:12 asciilifeform: it is unfortunately also one of the weak points in mircea_popescu's original sketch, as an enemy who turns up a 'for' binding signature can make an attribution of station ownership.
Framedragger: http://btcbase.org/log/2016-09-09#1538479 << wouldn't a simple nonce prevent against replay attacks (if you can call them that), tho? or did you refer to a case where the enemy attains the actual privkey of the station? ☝︎
mod6: yah, totally a different ball of wax.
mod6: i'll have to reread all of this stuff; i have difficulty with a mental model of this. but i do follow the detail points.
asciilifeform: a correctly-operating gossiptron must not be distinguishable from an unplugged cord , to anyone not in its peerage.
Framedragger: http://btcbase.org/log/2016-09-09#1538451 << it's good you acknowledge that, 'cause very time i want to point out "where's your $spec for $x guyz" i feel kinda shit 'cause i should do more, myself. which is a point. and yet the scandalocity remains! ☝︎
asciilifeform: enemy must NOT have the ability to determine that an arbitrary ip contains a gossip node.
asciilifeform: ('because he does not know the ip to ask' is an idiot answer. you don't even need to be at ft meade, Framedragger's scanner could easily be modified to send a challenge req to entire ipv4 space every couple of hours.)
asciilifeform: and anyone who - like mircea_popescu earlier - states that 'but all-comer will ~not~ be able to request a challenge' is stuck explaining why not.
asciilifeform: (if all-comer can get a challenge, this not only makes you ddosable tcp-style, but turns your gossip net into a ddosatron weapon for any idiot who can get spoofed packets into it)
asciilifeform: earlier in thread, mircea_popescu pointed out - correctly - that signature is not the only form of cryptoauthentication, and that there is also challenge-response, a la deedbot etc.
asciilifeform: mod6: presently it is a machine which contains a box labeled ???, which is the answer to a mathematical problem.
asciilifeform: mod6: no lord in his right mind would connect his royal key to a machine that emits signed material autonomously.
asciilifeform: that is, can infer a link between a station key and royal key.
asciilifeform: it is unfortunately also one of the weak points in mircea_popescu's original sketch, as an enemy who turns up a 'for' binding signature can make an attribution of station ownership. ☟︎
phf: (that idea is actually introduced in original gossip spec, but it's not obvious without rereading relevant bits a few times that it allows you to have secondary key
asciilifeform: and nowhere is it mandated that a station has only a sole key
asciilifeform: what i've referred to as 'station key' is simply a nonspoofable substitute for the ip address.
a111: Logged on 2016-09-09 14:51 mod6: <+asciilifeform> mod6: the correct pill is to decouple (at least from enemy's pov) gossip station keys from royal keys. << so to clairify this a bit, at least for myself, the gossipd node that I run and operate would verify transmissions sent to it with mod6-battlestation-key, but when I would actually go to send something out it would sign with mod6 "royal" or "personal" key?
a111: Logged on 2016-09-09 14:36 mod6: someday, when we have a tmsr cryptolib & rsa-o-matic, maybe we can starting building a gossipd prototype.
mod6: phf: ah, indeed then, we should rework it with what we've since discussed and rethought since 2015 or whenever, and see if a prototyping process can begin.
phf: we had a thread about it two weeks ago, where the conclusion was that gossipd as written in the only available spec has all kinds of problems and shouldn't be implemented/used
mod6: more than anything, i just want to be able to contribue a bit better to these gossipd discussions - which will be a vital part of the infrastructure.
mod6: i'd have to dig it up, but i thought there was a trilema that outlined the 'spec' back a while ago.
mircea_popescu: the two year old http://trilema.com/2015/artifexd-a-better-ircd-rfc/ ?
mircea_popescu: come to think of it, is there such a thing as a gossipd spec even ? what's the current canonical version ?
mod6: but, maybe the first step, to recognize Framedragger, is to update the spec. there has been much discussion about this, and maybe I'm having a hard time keeping all of the details in my head.
mod6: i don't want to disagree with what I just said here: <+mod6> a lot of new questions and discussions will arise once someone starts to prototype gossipd out.
mod6: <+asciilifeform> mod6: the correct pill is to decouple (at least from enemy's pov) gossip station keys from royal keys. << so to clairify this a bit, at least for myself, the gossipd node that I run and operate would verify transmissions sent to it with mod6-battlestation-key, but when I would actually go to send something out it would sign with mod6 "royal" or "personal" key? ☟︎
shinohai: I agree with trinque tho that V has a place in /usr/bin/local of the future
mod6: a lot of new questions and discussions will arise once someone starts to prototype gossipd out.
mod6: someday, when we have a tmsr cryptolib & rsa-o-matic, maybe we can starting building a gossipd prototype. ☟︎
mod6: V has a future. interesting to think about where it may go from here.
a111: Logged on 2016-09-09 10:50 mircea_popescu: incidentally asciilifeform mod6 trinque et all : what would you think about adding a p2p client to V ? it'd be like a daemon which listens for connections from peers and maintains lists of pressables.
mod6: <+asciilifeform> http://btcbase.org/log/2016-09-09#1538037 << lel what has mircea_popescu been drinking...? << It is an intersting thought; a distributed V, which rails against the complaints of "omfg another centralized piece of equipment!!11" ☝︎
asciilifeform: ~how~ to do this in a protocolic (rather than promisetronic) way, is another matter.
scriba: Logged on 2016-09-09: [00:55:24] <mod6> This hypothetical solution, even if it does work, wouldn't make it a one-button-push solution. Why? A person would need to get V + V.sig, verify, create a .wot dir, sync the patches + seals, manually or automatically, press the tree, then navigate into the pressed tree, and then `make`.
mod6: <+mircea_popescu> http://log.mkj.lt/trilema/20160909/#23 << yes mod6 but to clarify : "one button" refers to the situation where the user already has a trusted copy of V, and a .sig directory populated as per his taste. these are part of the definition of identity, and going forward can and should be assumed present. << Sounds fair. Thanks for outlining this.
mod6: Will have some more to report on in a few days. Salud!
asciilifeform: this is a healthy thing
PeterL: I feel like BingoBoingo has not posted anything in a while ...
mircea_popescu: though i suppose it should be a bunch of tiny azn dicks rather than one big one... hm...
asciilifeform: 'Lastly, it must be noted that there hasn't been any significant amount of these specimens on the market in approximately 25 years, and they are often only found in old collections. Recently I was informed that a Korean mining company has started a large commercial operation at Corocoro, specifically mining for Copper, leaving little possibility that more of these specimens will available in the future, or at least until the mining op
mircea_popescu: also, pure coper comes native, if you live in a land that doesn't suck you can just get yourself a chunk from the local cave.
Framedragger: [like, *of course* the only reason we want a spec which allows for data in initial syn packet is for shitphones to be able to load google ads quicker. use for security???! nowai]
asciilifeform: mircea_popescu: also i must sadly point out that aluminum wire is a forced mistake -- copper price.
mircea_popescu: sure, sure. a proper alloy saves you from shitty geometry, but you can't use zamac.
mircea_popescu: i never contested apple's a fucking zoo.
mircea_popescu: it's zeus, a sort of nero, the overpowerful lucky bastard.
mircea_popescu: but anyway - the greek idea of gods doesn't have them as a sort of christian deity, omnipotent, all wise etc. it's not fucking allah.
Framedragger: besides, they'd get confused themselves, what with deliberately no message authenticity; and we shall have a good time. am i stretching here?
Framedragger: mircea_popescu: how about i (an nsa employee, say) just make a filter which grabs all observed gossipd traffic (packet timing or w/e, and if it's an actually new transmission protocol, then supreme joy is me) and send it for further analysis. i shall assume that while it's not certain which messages are legit and which are not, the offending t3rr0rist group is too lazy to transmit proper false positives to provide noise,
Framedragger: yes this is lulzy and a disservice and i agree the reputation is not insanely great.., so to speak.
mircea_popescu: so you take a few points out.
mircea_popescu: for instance by rejecting outliers in a set. what you're doing then, in proper terms, is saying "since we have all this space left in the hierarchy cube, how about we use it to improve the dataset while still pretending it's just a sheet"
mircea_popescu: but, importantly, statistics works on datapoints - which are each as much a datapoint as all the others ! this is major.
mircea_popescu: it may be flattenable, and a large part of what we'll be doing is explore those graphs etc, but atm this is premature, we don't even have a prototype.
asciilifeform: much of what i've been doing in the mathematics room for past year has been to try to give this frying pan a handle.
Framedragger: semi-orthogonal (but not too orthogonal): ditching the conceptual level for a second and thinking about mundane reality, would a new transport layer even *work* given current internet infrastructure? there are problems with ICMP traffic on some ISPs (sure, ISPs should die anyway, and esp. those i hear you say.) this can be tested to some extent, hm.
PeterL: what keeps you from moving a signed message thorugh gossipd?
mircea_popescu: now, taking a tiny incremental step by providing redundancy for v while at the same time trying a udp implementation si a very sensible move at this juncture.
mircea_popescu: but all this stuff aside, back to the important point here : the "gossipd-like" thing contemplated for moving signed material (ie, v stuff) around is very much a different beast from the actual gossipd, which doesn't work on signed material ; presumably doesn't work on tcp etc.
asciilifeform: fleanode gives us a cheap plastic imitation of gossipd currently.
mircea_popescu: what, you're going to make a tank ? "new from mercedes. six gun ports."
Framedragger: then the argument does admittedly slide a little: yes okay, this is great that it can't be DoS'd that easily when the time comes; and yet it still has to parse multiple packets before it determines that hey, i don't know this fingerprint.
mircea_popescu: just because owner currently automated a hole into the ship doesn't mean that hole doesn't come with a pre-made hatch.
asciilifeform: it is unfortunately not (yet) a separable concern.
Framedragger contemplates a PoC `register` attack on deedbot
mircea_popescu: how would it send a challende encrypted to a key it does not have ?
mircea_popescu: you still can't answer, for lack of a way to form your answer.
mircea_popescu: asciilifeform well yes restate the whole thing. apparently there's a knot somewhere.
mircea_popescu: how would you send otps to check to a person you don't know ?
mircea_popescu: in a word, no.
scriba: Logged on 2016-09-09: [12:37:26] <asciilifeform> mircea_popescu: deedbot is not a working model of 1packetron omfg
scriba: Logged on 2016-09-09: [12:36:29] <mircea_popescu> except a working model of 1 is already both deployed and theoretically understood, as described above.
mircea_popescu: a network between a and b who have BOTH been replaced ?
asciilifeform: mircea_popescu: deedbot is not a working model of 1packetron omfg