trinque: http://logs.nosuchlabs.com/log/alethepedia/2020-12-18#1001739 << this is an incorrect reading of the opposability-of-ratings thread.
snsabot: Logged on 2020-12-18 17:25:21 asciilifeform: thimbronion: not even speaking of him specifically. but imho the psychology of 'mustn't sign anything, must not commit to a statement unless absolutely unavoidable' is destructive and in fact symptomatic of 'dark triad'
trinque: imagine oneself inside a system of law built in a society running on wot-tronics.
trinque: "asciilifeform does not suck" cannot be a statement for which I can be hung if later asciilifeform is found to suck.
trinque: or it's straight to the worst north korea ever built for anyone.
trinque: the narcissism angle cuts the other way also: imagine dorks running around with signed ratings from asciilifeform like they're a badge of honor, or "good credit" etc
trinque: or take another, "sally is a great prostitute". not all interesting statements are either beneficial or desirable to be sworn.
trinque to bed
trinque: actually one more thing. this line of thinking leads me to the "swearing" being optional, as in our thread the other day.
trinque: for *all* statements one may make
trinque: perhaps I swear my rating of you, and chuckle my rating of sally
trinque bed for real.
thimbronion: Just saw a >3 inch "grasshoper." Local name for it translates to "Lobster"
asciilifeform: http://logs.nosuchlabs.com/log/alethepedia/2020-12-23#1001756 << i'm perplexed by the conflation of 'signed' and 'must hang'. imho is exactly same situation as with e.g. vpatches and antipatches.
snsabot: Logged on 2020-12-23 02:49:39 trinque: "asciilifeform does not suck" cannot be a statement for which I can be hung if later asciilifeform is found to suck.
asciilifeform: so, for instance, there are bug fix patches in ffa. and then there are several people hosting old trees that dun have'em, cuz they tuned out. i do not see these folx as 'attackers'.
asciilifeform: even mp was not consistent re subj. for instance, published a signed 'castle deed' for # of people, incl. asciilifeform . and w/out any such thing as expiration date or whatever. iirc didn't even bother to publish a signed 'antipatch' for these.
asciilifeform: imho is a safe bet that if you still work with someone, you will have an accurate picture of his ~current~ ratings.
asciilifeform: the 'opposable' thread, as i understood it, was about whether ~third parties~ ought to have access to ratings. and imho a wot system is less useful than could be, if they do not. i do not grasp the appeal of the fabled freedom to lie and tell X 'my rating of Z is +1' and tell Y 'my rating of Z is -1'.
trinque: eh, you're glossing over the part that makes the whole point.
asciilifeform: trinque: hm?
trinque: I proposed that signing *any* statement ought to be a choice
trinque: if I wanted to extrude a trail behind me of every action I ever took, I'd accept the NSA's panoptical god as my lord and savior
trinque: some, yes, certainly
asciilifeform: trinque: i'm not 100% sure i understand the notion -- e.g. does it also bother trinque that vpatches are signed ? also 'trail for nsa' ?
trinque: not at all, and I don't mean that it's that kind of problem.
trinque: lets go upstack; what's the meaning of a signature?
trinque: broadly, in historical context
asciilifeform wrote a piece re the general case, but not walked away w/ impression that solved
trinque: signature means exactly "this is a statement to which I can be held" to my mind.
asciilifeform: obv. in the broadest possible sense -- signature links a payload (which may contain human-readable comments) to a pubkey
trinque: so I think if there's a place of difference it may be philosophical, in the sense that you believe it's useful to hold men to all statements.
trinque: or do we disagree that this would be the meaning of signing *all* ratings?
trinque: I just mean the social meaning of signature, aside the impl
asciilifeform: trinque: my q, posted in '15 and still unanswered, is 'to what does signing a vpatch behold the signer'
trinque: that on their honor the item is as they observed it.
trinque: which is I think appropriate for vpatches
trinque: and most *serious* ratings
asciilifeform: simply 'is' i.e. as found ? or also implies comprehension ? (in e.g. mp's cosmography, it did, 'if you sign, is same as if you wrote from 0 and published as original' roughly)
trinque: I was starting from where I figured we'd most likely agree
asciilifeform: it is sadly an unanswered puzzler from the very start, so i cannot even say 'agree' or 'disagree'
trinque: in its historical context doesn't signing mean this?
trinque: obviously men's games can be declared to mean anything, but I think this is the traditional interpretation
asciilifeform: the historical context on which asciilifeform based orig. v, was this one.
snsabot: (trilema) 2014-11-14 asciilifeform: there was a three-man system
asciilifeform: fwiw.
trinque: leaving my own opinion aside, I think the mptronic objection (and I'm not here as a talmudist to retroactively make the scriptures consistent) was that if one takes s electronic signature as an extension of the ink one, ratings cannot be attributable in their published state, because this exposes the rater to claims about the truthfulness of their rating, which whould either chill or
trinque: bureaucratize ratings.
asciilifeform: trinque: my intent wasn't to talmudize, but to actually work out this binomial;
trinque: if one does not take this view of signatures, or yes, had some mechanism to declare intent as signed "I signed to capture forevermore that I saw the enclosed item as steaming shit", would be entirely different
trinque: certainly
trinque: this difference wouldn't however remove the problem of "now derps have a place to accumulate legalisms"
asciilifeform: what i dunget, is, how is a signed rating a problem in the case of a future conflict where had to e.g turn a + to a - ?
asciilifeform: imho is exactly same situation as a vpatch
asciilifeform: i.e. reader is responsible for making an effort to get ~current~ (per his set of trusted signatories) tree
trinque: I think the charitable interpretation of the perspective is that "opposable ratings are a risky precedent that later could accumulate loads of legalistic nonsense"
trinque: perhaps the better solution to this is deniable keys
trinque: i.e. "this is my drinking with my buddies subkey, and you can't prove it's me"
asciilifeform: trinque: no one even now prevents 1 fella from using N keys
trinque: yep!
trinque: this was also comtemplated back when
trinque: *contemplated
asciilifeform: aha
asciilifeform: trinque: 1 other point asciilifeform considers quite important, is that the insanities of whatever totalitarian hell you, i, or other user of public key crypto may happen to inhabit, are neither caused nor cured by the choice of cryptosystem or custom for signing
asciilifeform: as dan mocsny wrote, 'anyone can in principle be held responsible for anything, given the existence of a force strong enough to hold him responsible and wishing to do so'
trinque: sure, and I'm satisfied that sub-identities are a good solution to the problems I see.
asciilifeform: imho not only solution, but in fact what everyone has been doing by default all along. (there were folx known only by pubkeys/nick, at all points, e.g. punkman)
asciilifeform: asciilifeform, interestingly, did not (as naive lamers could suppose) but heavily win from publishing under human name. on numerous occasions my www demonstrably helped to drum up biz in saecular meatspace.
asciilifeform: *did not lose but
asciilifeform: but naturally erryone has the choice of whether to do this, and to proclaim 'this-here is my pub' etc
trinque: yep, I dropped the hipster nick shortly after arriving
trinque: this space to choose is exactly freedom imho.
trinque: develops a better mind than otherwise
asciilifeform: upstack, i still hold to the position stated in my '16 piece re subj -- that signatures ~whenever possible~ ought to have standardized meanings from which signer may choose, and cement the choice into the payload
asciilifeform: e.g. 'this is just as if i had written it myself' 'this is steaming shit which i advise against touching with barge pole' and similar (see subj re examples) .
trinque: hm, if you had a standard ontology of signings, what'd you build atop that?
trinque: instead of say a wrapped signature with your penned introduction at the top?
asciilifeform: trinque: see here re rationale.
trinque: right, the disapproval is useful.
trinque: brings my project to jar linux to mind
trinque: *all* of the early patches have my unequivocal disapproval
asciilifeform: specifying an ontology is less than half the battle -- gotta get ~other~ people to agree on the spec, ~and~ to discuss in sufficient detail that all can be certain re the shared meaning
trinque: would be interesting to look at how much "hated" code a particular codebase contains
asciilifeform: my ontology didn't make it even past stage 1
trinque: sure
trinque: all the best things die in the womb; satan reigns
asciilifeform: imho would be a win tho, to be able to 1) publish 2) with caveats or even condemnation 3) while statement carries signature
trinque: brings forth the idea of a condemnatory patch that changes nothing, but simply marks substrings in src as s damned, etc
trinque: speaking of damned, this keyboard
asciilifeform: trinque this aint the 1st time you mentioned a dud kbd
asciilifeform: what happened there, outta curiosity?
trinque: shitty modern laptop
asciilifeform: aa
trinque: the mac ones break after a year tops
asciilifeform: a these. funnily enuff i have one, but dun recall having pressed the keys on it many times, i had with it a little portable kbd thing that in fact fits on top of the machine when unfolded
trinque: hm, that's an idea
trinque bbl, going to go absorb some sunshine
asciilifeform: there's 'over 9000' of these. the particular example, i liked especially, had 'topre' (jp) waterproof switches
asciilifeform: trinque: laters
trinque: nice to be chewing on the old notions, even if still mostly unsolved.
asciilifeform: http://logs.nosuchlabs.com/log/alethepedia/2020-12-23#1001840 << ftr i avoided deliberately any non-machinereadable outputs in mechanisms in v.
snsabot: Logged on 2020-12-23 13:34:12 trinque: brings forth the idea of a condemnatory patch that changes nothing, but simply marks substrings in src as s damned, etc
asciilifeform: i.e. 'if you want to mark line 17 as 'damned', ok, but write a patch that actually removes that line and inserts comment re why'. rather than merely 'string'
asciilifeform genuinely bbl