log☇︎
124100+ entries in 0.074s
diana_coman: I always read asciilifeform's "meat" along the lines of "call of the annoying reality of existence in human form"
asciilifeform: eyeball-powered diff, when there is a mechanical one, is nuttery
a111: Logged on 2018-01-18 20:53 mircea_popescu: nothing prevents you from saying "this is the genesis of apeloyee's fantastical funicular, it consists of a cog i stole from alf's exceptional exsudator, some other bits i wrote myself and various parts i forgot where i stole" ; but something QUITE fundamental prevents you from writing down "include #fucksticks" and to have a helpful paperclip pop out of nowhere in your emacs guts and declare "this so far looks like you're pu
asciilifeform: http://btcbase.org/log/2018-01-18#1772477 << in ideal vtron, the 'and i stole x from a, y from b...' is protocolic, rather than promisetronic. i.e. abolition of eyeball-powered diff. ☝︎
a111: Logged on 2018-01-18 20:33 BingoBoingo: From the WaPo: "Six buses operated by Silicon Valley tech companies to shuttle employees to their offices have been attacked on the highway over the past week. The California Highway Patrol reports that four Apple and one Google bus had their windows shattered on Tuesday by unidentified projectiles"
asciilifeform: http://btcbase.org/log/2018-01-18#1772453 << this has been a regular thing since at least '14 ☝︎
asciilifeform: ( and it ain't too far from the troof... )
a111: Logged on 2018-01-18 20:31 phf: i half expect it to be some kind of kafka narrative, where, when not engaged with tmsr business, alf actually does some macabre butcher work involving big meaty hands and slabs
mircea_popescu: point is it's enough to choose <
diana_coman: in the "it takes human hand"? the thing put together needs taken apart of similar nature
a111: Logged on 2018-01-18 20:58 mircea_popescu: the problem is that later on, eucrypt's smg_keccak will be pulled into the main to be used for purposes ; if EVEN LATER it gets a patch, phf will then not actually have a way to seamlessly "get just the patch", he will have to regrind at that time anyway.
diana_coman: http://btcbase.org/log/2018-01-18#1772485 <- after things are put together by human hand it takes human hand to take them apart, yes, unsurprising; point is it's enough to choose and pick out of the desired vpatch what is relevant (i.e. for keccak in this example) and that's as far as it goes ☝︎
mircea_popescu: which is why the original "do not link across v trees"
mircea_popescu: ie, this "independent parts in an automatic fashion" is a hope impossible in practice. the only way he can have it is if HE reads it, as it is found wherever it is found (eucrypt as it happens here), and then HE puts it in, as a regrind, ie, yes, "de novo" item.
mircea_popescu: the problem is that later on, eucrypt's smg_keccak will be pulled into the main to be used for purposes ; if EVEN LATER it gets a patch, phf will then not actually have a way to seamlessly "get just the patch", he will have to regrind at that time anyway. ☟︎
mircea_popescu: consider concretely the case of eucrypt's keccak. diana_coman is writing it as a direct derivation off genesis, meaning on extant v impls if one wanted to import it they could import JUST it, without the rest of eucrypt (it'll be pulled in later through the usual procedure in eucrypt itself). superficially this may seem like it encourages phf to go "o i know, i'll just link keccak patch into my codebase rather than regring (i ☟︎
a111: Logged on 2018-01-17 19:56 asciilifeform: however it DOES mean even ~more~ work for folx using v, than ever before. and not less.
mircea_popescu: http://btcbase.org/log/2018-01-17#1771941 << it means the bar for what may be acceptable primitives is slightly lower than before ; this is not necessarily avoidable though, the advantage of having alphabet starts around having twenty or so letters only ; fifty letter "alphabet" is not much better than 5000 hieroglyphery. ☝︎
mircea_popescu: in other words, the only maker of independence in the realm of thought is the hand of man, and this can't be fixed.
mircea_popescu: because it'd need guile to do and guile is shit. if for no other reason. comprende ?
mircea_popescu: nothing prevents you from saying "this is the genesis of apeloyee's fantastical funicular, it consists of a cog i stole from alf's exceptional exsudator, some other bits i wrote myself and various parts i forgot where i stole" ; but something QUITE fundamental prevents you from writing down "include #fucksticks" and to have a helpful paperclip pop out of nowhere in your emacs guts and declare "this so far looks like you're pu ☟︎
a111: Logged on 2018-01-17 19:55 asciilifeform: iirc mircea_popescu's argument was that it is wrong to say that they could ~ever~ be properly independent. and that if they could be shown to be independent, they ought to be separate v-trees.
mircea_popescu: http://btcbase.org/log/2018-01-17#1771939 << this is quite it, for half of the problem ; for the other half -- ANYTHING is independent if you make it so, just put it in your genesis. ☝︎
a111: Logged on 2018-01-17 19:43 asciilifeform: fenómenos que "no tiene antecedentes" en el país << orly?
mircea_popescu: http://btcbase.org/log/2018-01-17#1771920 << quite rly ; the south american south is remarkably peaceful, nobody needs (or wants) them for much. ☝︎
mircea_popescu: sadly, until phf's improved diff hits the deck there's nothing we can do here.
mod6: http://p.bvulpes.com/pastes/ZvDPR/?raw=true << diana_coman has all the details.
mod6: yah, food for thought here maybe.
mircea_popescu: well, this is the curse of magic words.
mod6: yah, that's actually what it added up to: "+++ ", but no diff info / hash after it.
mircea_popescu: mod6 the magic string diff uses is "+++ " ie three plusses AND A SPACE
shinohai tips hat to mats
deedbot: shinohai updated rating of mats from 2 to 3 << #trilema - Never has let me go into battle with an empty war-chest.
mod6: mircea_popescu: also, fwiw, we might need to adjust our "NO '--- ' or '+++ ' to begin a line in a vpatch to "NO '-- ' or '++ '". There was a vpatch in development where my vtron choked on a line being added into a source file that began with '++', and with the diff '+', became '+++'. My vtron correctly choked here. But maybe a bit of an adjustment to the rule?
BingoBoingo: From the WaPo: "Six buses operated by Silicon Valley tech companies to shuttle employees to their offices have been attacked on the highway over the past week. The California Highway Patrol reports that four Apple and one Google bus had their windows shattered on Tuesday by unidentified projectiles" ☟︎
mod6: <+mircea_popescu> if anyone is wondering, alf lives in a sausage plant. there's lots of meat. << been wondering if I should move in with him, or take bets on when he gets scurvy
shinohai: Heh, in my my mind I always imagined asciilifeform living in abattoir because of the "brb, meats"
phf: i half expect it to be some kind of kafka narrative, where, when not engaged with tmsr business, alf actually does some macabre butcher work involving big meaty hands and slabs ☟︎
mircea_popescu: if anyone is wondering, alf lives in a sausage plant. there's lots of meat.
mircea_popescu doesn't think greek tragedy is party to this discussion, but is overwhelmed by explaining why.
phf: or lovers, like what is it, the two gentlemen of verona
mircea_popescu: i can't think of major examples outside of the star wars thing (coulkdn't think of that either until pointed out). commedia is mostly brothers.
BingoBoingo: <asciilifeform> or is that just hollywoods << The Greeks did it!
asciilifeform: or is that just hollywoods
asciilifeform: i thought it was typically fathers
mircea_popescu: but that's only one pov.
phf: ascii vs brother doesn't strike me as that kind of relationship. more like "you have to learn much little one, and i have N years on you"
mircea_popescu: imagine, there's two of these out there!
phf: that would be quite a twist
mircea_popescu: (but since we're on it -- the enduring interest in obfuscated-c is strictly this, "let us try and write a novel in orthogonal language ; this should be done in c because it's very much not orthogonal". became a self-recursing joke, that the practitioners don't even properly understand, just sorta-feel. BECAUSE they talk about it in natural languages.)
mircea_popescu: and all this goes right into that older thread of ambiguity, orthogonality and language -- you can never make a language that's orthogonal and ~useful~ in the natural sense. let alone "that anyone'd want to use". ☟︎
mircea_popescu: you could not have written such a tagging mechanism as it'd have permitted a machine to recognize the substance of what you were saying when discussing "pointers" which weren't.
mircea_popescu: and this identity is also fluent, because different things are the same thing, such as no further than earlier the "you really want a lisp stack". this recognition is only there because of personal knowledge, and absent otherwise.
mircea_popescu: the strong statement here is that the ~only~ possible identity bits of code have is based on ~personal memory~. to revisit the oft used bubblesort example -- any particular implementation of bubblesort IS bubblesort because ~you~ recognize it as such and for ~no other reason~, factual ~or possible~. consequently fits in head as the basis of code identity.
mircea_popescu: asciilifeform quite evident ; but yes, i believe wrong problem to solve. "oh lord, how could bits of code have an identity ?" "why ?" "so they could be my girflriends" "Come on!" ☟︎
a111: Logged on 2018-01-17 19:34 apeloyee: files are NOT INDEPENDENT. despite CVS and v pretending they are. this is a problem. you could have required some form of cryptographic commitment to either the tree state or even the antedecent patches themselves, but didn't
mircea_popescu: http://btcbase.org/log/2018-01-17#1771907 << v is not at all pretending they are ; that's accidental misconstruction. ☝︎
asciilifeform: worse, was thinking of very different ( and probably unsolvable ) problem ( 'how to give arbitrarily small pieces of code, a perma-history' )
a111: Logged on 2018-01-17 19:32 asciilifeform: apeloyee: the 'liable to change' thing was very much NOT part of my orig design for v.
mircea_popescu: http://btcbase.org/log/2018-01-17#1771903 << because didn't seriously think of the particular problem, which happens with prototypes. ☝︎
mircea_popescu: diana_coman cool deal then.
phf: never mind then! i gotta figure out how to do the whole "file moved part" anyway, and i don't need a hashing function yet. i'm using a sha512 implementation from busybox
mircea_popescu: a damn, seems i shot too quickly. sorry phf
diana_coman: padding, bitstream, sponge, oaep to follow; currently still in the works
mircea_popescu: asciilifeform don't tell me you didn't take your volkskomputi to the commissariat for the mandatory urine injections or what was it.
diana_coman: mircea_popescu, yes, all that there is still only... first part
mircea_popescu: diana_coman ah, is it a two part item then ?
diana_coman: phf, those are the keccak transformations; do you need the actual sponge?
mircea_popescu: http://btcbase.org/log/2018-01-17#1771889 << that's ok, pipeline permablown anyway. ☝︎
phf: aye, i'll marry keccak to vdiff today or tomorrow to get a poc out.
mircea_popescu: seems one of the most promising.
asciilifeform: i got a megatonne of those . ( e.g. unrolled comba )
mircea_popescu: some possible things need to be made before it can be decided.
asciilifeform: aha, it's on the side-point list.
asciilifeform: it's a variant. not currently convinced that it is The Right Thing.
mircea_popescu: seems your stack wants to meld into that anyway.
asciilifeform: ( i considered lispism, rejected for the given application strictly on account of moving-parts-count , 'p' gotta have parachute-level simplicity )
mircea_popescu: why not actually do it then, and first construct an ada circular ring lisp-memory stack first ?
asciilifeform: it already lives in the forth hole, lol
mircea_popescu: note how ffa ends up satisfying the greenspun rule, through the predictable entry point.
asciilifeform: it's next on asciilifeform's conveyor after ffa. ( this should not discourage anybody else from writing adatronic lisps of whatever type, however )
mircea_popescu: http://btcbase.org/log/2018-01-17#1771869 << thinking about this, it seems that both a) a re-implementation of the lisp memory model is unavoidable for any serious purpose and b) the place where the rewrite will start is the stack ; are true. ☝︎
asciilifeform: ( you have to satisfy the compiler that it has exactly same 'life' as the item pointed-to. and in practice this is quite difficult. )
asciilifeform: ada working in normal mode typically won't even let you take a pointer of an object on the stack.
asciilifeform: ( the ada passing by reference semantics give you almost anything that c programmers typically want from pointerism )
asciilifeform: so long as you don't use the heap, you more or less can dispense with traditional pointers.
mircea_popescu: whereby the jump is in the item pointed to, not in the mechanism of pointing
asciilifeform: ( spoiler : you can trivially implement linked list without -- properly speaking -- pointers, but with integers, this also came up in the earlier 'lisp in ada' thread )
mircea_popescu: ah you don't actually mean to use pointers, just a lisp-style memory emulator for the stack itself
mircea_popescu eats then
asciilifeform: right. but... i'ma let mircea_popescu eat rest of the thread
a111: Logged on 2018-01-17 18:07 asciilifeform: however this introduces explicit pointerism. ( though, i will add, NOT pointer-arithmetism )
mircea_popescu: http://btcbase.org/log/2018-01-17#1771854 << will be hard to keep at bay. ☝︎
deedbot: http://www.dianacoman.com/2018/01/18/eucrypt-chapter-6-keccak-transformations/ << Ossasepia - EuCrypt Chapter 6: Keccak Transformations
BingoBoingo: By all appearances though the Englishman is functionally taking his vacation on Facebook.
BingoBoingo: And there's an old USian chasing 80 degree temps for his arthritis and going to AA meetings that at least made an effort at memorizing phrases before he fucked off to Buenos Aires
BingoBoingo: And the Brasileros without spanish can usually get by on the latin language mutual intelligibity deal
asciilifeform: even on pnoje there are language gamez etc
shinohai: Maybe he is still searching for an app for that!
BingoBoingo: Apparently one can have fun in Punta Del Este with zero spanish provided they have a Trump budget, but it is a serious WTFism.