log☇︎
1323 entries in 0.628s
asciilifeform: shitgnomes lie, lie, but that would be inconsequential if the lies did not also take the form of bogus code that supposedly functions, but go and try to replicate. not one thing in ten works as described or to spec.
mircea_popescu: im not even sure pgp has a spec.
gabriel_laddel: As for PGP, I'm assuming that I'm going to have to spend a decent amount of time with the codebase and reading the spec. Perhaps not.
punkman: artifexd, maybe some relevant ideas for gossipd in that spec
trinque: as per spec
asciilifeform: trinque: finished in the sense of there not remaining a single peripheral that could be described as not entirely working to spec ?
Vexual: GAMMPool spec are a little thin, what do you actually wanna do?
BingoBoingo: mircea_popescu: I have no idea for a better spec.
mircea_popescu: and now im off to trilemize the dictionary/spec discussion, if anyone wants to be in ask now.
mircea_popescu: i defy you to find a spec anywhere in use in the shitstack today that manages to not include things it really has no business specifying.
asciilifeform: but the only proper spec is a) actual spec b) fits in head c) because was created by actual person
asciilifeform: not that a spec arrived at via the usual idiot committees is any better
asciilifeform: mircea_popescu: that's how you end up with 'the spec is the implementation' crapolade
mircea_popescu: that's the main task of spec work : avoiding doing anything as much as humanly possible.
mircea_popescu: complex specs are in no sense better. the only thing a very complicated spec shows is that the designer did not understand the overwhelming importance of not specifying.
asciilifeform: decimation: i can't be arsed to wade through spec, but didja find who generates key?
decimation: but I don't know if the spec is open
jurov: &> file (aka >& file) is not part of the official POSIX shell spec, but has been added to many Bourne shells as a convenience extension (it originally comes from csh).
mircea_popescu: nubbins` it's forever on the router dude. that was the spec. for a box to plug into your spare router hole and sit there forevert.
decimation: mircea_popescu: well, the way it works now, to my understanding, is that there's a kind of wot among high-spec clocks
mircea_popescu: standards are racist. the spec is the code. WE ARE BRINGING STUFF TO AFRICA
ben_vulpes: spec a node, i guess.
jurov: well mircea, then make a spec. many people unsuccessfully tried various mixing proposals that were supposed to do what you propose
mircea_popescu: <mircea_popescu> jurov i thought we were discussing bitcoin as a spec, rather than bitcoin as a hack. <<
mircea_popescu: jurov i thought we were discussing bitcoin as a spec, rather than bitcoin as a hack.
decimation: whether they want to create it or not, there is a spec
davout: gavinandresen: "okey dokey, then we might not have much to talk about if you want to stick with OpenSSL bugs that were included in the protocol by mistake." <<< actually i distinctly remember mike hearn telling me how that particular bug was part of the protocol and how it somehow justified not putting any effort towards actually specifying anything, in a spec, not in code
BingoBoingo: latest diff https://code.google.com/p/plan9front/source/detail?spec=svn8cd112f650b1e4f831de26631de627465bceb219&r=e263d58bfeda6b441179595bc44d0ef00374c75f
mircea_popescu: <davout> just wondering if there was a specific reason it wasn't mentioned in the spec << so people can have fun.
davout: just wondering if there was a specific reason it wasn't mentioned in the spec
davout: BingoBoingo: sure, but if your timestamping respects deedbot's spec just get it included along
mircea_popescu: anyway, that spec'll hafta be revised, but needs more thinking.
undata: I'm just taking the spec as is, implementing around the go bot afore-linked
undata: the spec intends assbots fingerprint is one param?
undata: mircea_popescu: I get "No data" from the link in your spec article, and when I substitute my key's fingerprint
davout: that's how is spec'd yes :-)
undata: I'm hacking on something now, but will wait for an updated spec
mircea_popescu: PeterL original spec didn't call for that.
mircea_popescu: so i guess ima have to modify the spec instead of finishing my "on terrorism' article.
assbot: Logged on 17-01-2015 22:34:54; davout: asciilifeform: yeah, that's what i was reading, it mentions user ids in the subpackets spec, but i'm unsure whether that includes an actual key fingerprint, i tend to understand that it doesn't
mircea_popescu: accidental spec!
davout: well, i was just reading about them in the deedbot spec
davout: asciilifeform: yeah, that's what i was reading, it mentions user ids in the subpackets spec, but i'm unsure whether that includes an actual key fingerprint, i tend to understand that it doesn't ☟︎
asciilifeform: davout: the version 4 signature subpacket spec << what do you want to know about it ?
davout: lol, where does the spec even mention a second party?
davout: undata: i didn't spec it, ask mp for the rationale -> http://log.bitcoin-assets.com/?date=30-08-2014#815284 ☝︎
davout: the version 4 signature subpacket spec isn't that clear to me, maybe asciilifeform has some insight
davout: - alter spec and add a gribble dependency
davout: oh, and it's ppl who have L1/L2 trust from assbot, a subset from the wot members sez the spec
davout: thing is, i was also reading mp's deedbot spec, the part i was wondering about was the "extract keyid from signed message, and use it in w.b-a.link URL"
ben_vulpes: undata: if you're interested in implementing, talk to MP for the spec
mircea_popescu: kakobrekla works for me, editing this spec.
artifexd: I believe the point is that Snowden didn't say gpg was compromised. However it could have been compromised since he left. Only mircea_popescu knows mircea_popescu's motivation for that spec though.
mircea_popescu: asciilifeform i have no idea, it's an open spec.
asciilifeform: "oh, we included code but we had no idea what it does" sort of approach to code-is-spec specwork << this is actually a very serious boojum, beyond what most realize. i'll give example.
mircea_popescu: davout if proof was needed that the 'teh-code-is-teh-spec' approach is fundamentally braindamaged, well, look no further <>< clearly. not to mention the recent ecdsa openssl debacle, which is pretty much exactly a replay of the bdb stuff. "oh, we included code but we had no idea what it does" sort of approach to code-is-spec specwork
davout: sgornick: an implementation detail in some older client does not count as a protocol rule in my book, if proof was needed that the 'teh-code-is-teh-spec' approach is fundamentally braindamaged, well, look no further
mircea_popescu: can you implement the deedbot spec ?
BingoBoingo: One crack rock and a text file can test against an actual spec.
mircea_popescu: why not ? spec it.
mircea_popescu: asciilifeform i thought it's implicit in the spec, in the "receiver decides if to relay" : you can configure a client to pass, to user or other clients, whatever you wish
mircea_popescu: artifexd you will notice the spec is broadcast-oriented.
mircea_popescu: in both cases, we are discussing user Panopticon, who sees all and says nothing. now, in spec as is, it is true that user P will know... nothing. correct ? whereas in your proposed spec, he would know... everything. correct ?
mircea_popescu: spec was simpler on the assumption that since the helo package is keyed to the supposed key of the server, there's no need to challenge
mircea_popescu: i don;'t see what in the spec would prevent anyone from so doing.
asciilifeform: let's work out, using spec as given in mircea_popescu's article: what does an inquisitor know about a particular utterance
mircea_popescu: not with the spec as given, imo.
mircea_popescu: there is no such thing as "sign for friends" outside of the model detailed in my spec.
artifexd: I still haven't processed the wot part of the spec. I'm still trying to understand why you wouldn't sign the messages. It seems to be asking for evil actors.
asciilifeform: when part of mechanism, by spec and by reliable practice sans hand-holding
mircea_popescu: o you meant in your signed alt-spec. mkay.
mircea_popescu: mebbe "heyo" ? make an argentine joke right in the spec!
RagnarDanneskjol: punkman has a newer much more detailed spec
assbot: 1 results for 'deeds spec' : http://s.b-a.link/?q=deeds+spec
davout: !s deeds spec
mircea_popescu: ascii_modem ima spec it in a few
mircea_popescu: ima make a spec for it.
punkman: davout: who decided, and why, that the contents should be gpg signed ? << it was mp's spec. even if you don't want to tie messages to identity, need some kind of access control and WoT works well for that
davout: guess i didn't see the spec
punkman: davout, spec
decimation: asciilifeform: if you got it to fully synch, the answer is 'no' as far as the '0.5.3 spec' is concerned right?
decimation: I just want the spec and docs (with talmud of errata)
asciilifeform: folks who write spec often forget that a spec is a cheque that someone's pair of hands needs to cash
undata: asciilifeform: seems like there's just far too much spec for that to happen
ben_vulpes: +asciilifeform:kakobrekla: i take it turdatron functions to spec now? << what spec
asciilifeform: kakobrekla: i take it turdatron functions to spec now?
gribble: You rated user punkman on Sun Oct 26 16:38:08 2014, with a rating of 2, and supplied these additional notes: improved on my spec!.
assbot: DosVectors · http2/http2-spec Wiki · GitHub ... ( http://bit.ly/11JuDyq )
mircea_popescu: https://github.com/http2/http2-spec/wiki/DosVectors << incidentally, what do we think of http/2 ?
assbot: Logged on 31-08-2014 02:17:56; assbot: Logged on 27-07-2014 19:05:02; asciilifeform: mircea_popescu: in 2010 i concocted a variant of the test that could be carried out for approx. 50k USD, from scratch. schwartz answered that, in his estimation, the test would work to spec, but results would not be accepted by the field unless carried out on one of the two existing eotvos balances. one - adelberger's, one - chicom.
mircea_popescu: supposing you have a person printing machine, can make them up to spec.
decimation: yeah that's all 'mil-spec' hardware, with the 'round' connectors
mircea_popescu: nubbins` no lol, cuz no \n in spec :D
mircea_popescu: by spec it should.
mircea_popescu: borlandc++ sucked by comparison. maybe because i was too dumb to properly grok the language, or too lazy to learn the spec
decimation: unlike intel, the spec should be published before the turd is made
mircea_popescu: mandarin was never specced, because, well... the speech is the spec.
mircea_popescu: bounce o.o. secret spec ?
bounce has a phone spec in mind that could stand building
Pierre_Rochard: Ok, will do! I’ll write up the spec and pass it around