mircea_popescu: i'm just saying - there's nothing lost here. can emulate naked pastebin in v style np.
asciilifeform: mircea_popescu: what does the cafe key provide ?
mircea_popescu: phf yes, in principle. in practice i expect "centralizing" rebasing vpatches to happen all the time to prune trees of their noncontroversial joints.
mircea_popescu: you can literally come up with an idea for a thing while travelling ; go to internet cafe ; spin up gpg to make you a new key while you bash it down ; then sign the patch with that key which you don't even bother taking from there.
phf: mircea_popescu: correction in this case can come in a form of another vpatch either by original author after the discussion or by anyone who wants to make annotations, etc. in which case the dialog becomes valuable on its own, because it preserves the differences that have didactic value
mircea_popescu: ie, item X enters life signed by "doubful matter key Z" and then a year later is signed by "ok this is it key Z" and makes it to the stable press
mircea_popescu: you ~can~ after all rebase a patch ; not just because of the patch, but also because of the sig.
asciilifeform: phf: recently mircea_popescu suggested a cleaner scheme where everybody has multiple signing keys, and they rate one another, as if they were people. which is probably as mechanized as this will ever get.
asciilifeform: so we're at the chalkboard still.
asciilifeform: phf: i stabbed at the problem of formalizing a way to specify what it is you actually commit to when signing a patch. it was the 'vectorized' thing, and everybody barfed.
a111: Logged on 2016-02-20 22:45 phf: "i, ordained computron mircea lifeform, in the year of our republic 1932, having devoted three months to the verification of checksums, with my heart pure and my press clean, sit down to transcribe vee patch ascii_limits_hack, signed by the illustrious asciilifeform, mircea_popescu, ben_vulpes and all saints."
phf: but the reason i made those statements yesterday is because i think that like saying things in log is an opportunity to be corrected, so does posting a vpatch, it could be a learning experience. instead the mindset seems to be http://btcbase.org/log/2016-02-20#1411214
mircea_popescu: software design as god intended. FIRST you do by hand ; THEN you automate the parts worth automating.
mircea_popescu: phf ah, yes, but note that the item contemplated there is specifically not discussion ; but its results.
a111: Logged on 2015-09-12 20:28 mircea_popescu: which... you know, i slept on this, and it occurs to me that ACTUAL MATHEMATICS might be the perfect thing to implement over V.
phf: actually the subject is an ongoing, year old conversation about about V as a way to represent knowledge starts with http://btcbase.org/log/2015-09-12#1271598 and then gets revisited every couple of months
a111: Logged on 2015-08-31 14:33 asciilifeform: http://log.bitcoin-assets.com/?date=31-08-2015#1257082 << i loathe python per se. but the only realistic alternative was perl. (my original attempts at 'v' were in awk/sed, and did not work very well)
a111: Logged on 2015-08-05 14:00 mircea_popescu: there's no reason for this to take a week to do.
a111: Logged on 2015-08-05 13:38 mircea_popescu: one is USE. specifically - hanbot must be able to put into work the theoretical advances b-a produces. and ima use her as a stand-in for "intelligent and willing to work, but not able to grow a beard".
a111: Logged on 2015-08-05 04:42 asciilifeform: also the continuity of identity. if, for example, my patches at some point go from tiny to elephantine, from single-purpose to 'omnibus', from deadly-simple to 'wtf is that' - folks are to presume that i have been finally killed and key is in hands of hitler. and should then rate accordingly.
a111: Logged on 2016-12-22 17:46 phf: i'm already spelunking logs for v history, to go on another trip at the moment :)
mircea_popescu: kinda brings to mind this lulzy discussion from 2013 : http://trilema.com/2013/the-wonder-of-inflation/#comment-93905 "dude, if inflation is this high, us gdp must be shrinking ?" "certainly is".
mircea_popescu: aha. the way this has to go.
phf: word on the street, that u.s. alibaba office is hiring like crazy, and that the salaries are so high, microsoft/amazon/apple no longer attempt to match compensation in the event of the poaching. so "blacklist"
asciilifeform: then i have monumentally nfi
mircea_popescu: the boys are shockingly shy, it surprised me. 10 years ago used to be about 100x more virulent.
asciilifeform: maybe http://trilema.com/2016/how-the-other-half-lives-a-very-seriously-funny-article/ ?
asciilifeform: no bot there?
mircea_popescu: incidentally, speaking of misfortunate subhumans : anyone care to guess the most frequent answer my bot gets ? it's "most frequent" in the sense that the next-frequent is like 2 std deviations rarer. hint : it's a question.
phf: asciilifeform: actually Esperanto was big in su. it was not an anathema to say that in the communist future the peoples of the word are going to speak not just Russian, but also some other, better, universal language. certainly interstellar trade would need something better. or when you talk to machines.
mircea_popescu: on the other hand i could readily notice the havoc this had wrought upon her poor brain.
mircea_popescu: but fwiw, serbian chick stands out in my mind, she was 19, her poor head had just come out of 12 years of which 6 were spent doing cyrilics and the other 6 doing ~BOTH~ latin and cyrillic and who genuinely thought that you know, this is BETTER EDUCATION!! because where other people got one, she got both!!1
phf: i still remember how my hired driver crossed the state line and had to speak a very broken english with the locals to ask for directions
mircea_popescu: well, india is a sort of china's africa. except the chinese were too lazy to properly fuck theirs.
asciilifeform: the most, afaik, egregious babel on planet3.
asciilifeform worked in a lab with two dozen folx from india, no 2 spoke same language other than english
mircea_popescu hasn't fucked that many indian womenz has not much clue as to their native barking system.
phf: indian children still mostly speak conlang "by this measure". there's something like 15 ~official~ dialects, all of which have a significant post panini sanskrit incursion so that in say malayalam you have two different ~grammatical~ structures available to express the same thing, malay and sanskrit.
asciilifeform: afaik esperantism was one of those crackpotteries utterly thermonuked by ww1, so i have nfi where one would even begin to look for remnants.
mircea_popescu: but w/e, maybe my very dim view of "scholarship" in the field is entirely unwarranted, i'm just a meadhater and library contains "a river of gold", to quote obama.
mircea_popescu: you have a better shot at finding well written c-s implementation in the library.
asciilifeform: which is why i'd like to see ~how~ they were fucked, there has to be some remaining crater to look at
mircea_popescu: asciilifeform the problem is that for most people, language is the only thought-amplifier they have. once you teach them a non-language they're very severely fucked.
mircea_popescu: natural language dictionaries are usually in the 100k symbols range ; however natural languages altogether are very large graphs, and i'd venture in the yottabyte range.
asciilifeform: mircea_popescu: sorta why i was specifically curious re what it was that the esperantist ~children~ spoke. the resultant creole, not the original 'megalomaniacal engineer' item.
a111: Logged on 2016-12-11 19:54 mircea_popescu: phf suppose you make an ai expert system to beat us at go. this gives you two practical options : either include 10gb worth of binary flags preset ; or else have us beat it at go for 10 centuries before it gets to where it plays like a freshly fucked 19yo.
mircea_popescu: here's a view that may help alf : natural language is the summarized http://btcbase.org/log/2016-12-11#1581451 calculated over something like 10*5 * 10*7 * 50*24*365*50 ~played~ instances of prisoner's dilemma.
asciilifeform: imho the jury is out on conlangs.
mircea_popescu: it's just not clear if it has much substance on the jaguar's side. but clearly when ~you~ look into its eyes, ~you~ perceive a soul there.
asciilifeform: phf: why not one op that dispatches, correctly ? is it as if the compiler ever did not know the actual operand's type ?
mircea_popescu: i'm not terribly unhappy with the analogy.
asciilifeform: common lisp committee was hobbled by babelism, had to stuff in the redundant idiocies, and never got to, say, actually developing streams properly (why the fuck are we stuck with hacks like gray streams? and where is the commonlisp with ~working~ mop ? etc)
phf: elt can potentially escape in the error clause to handle extensible sequences (and it does on SBCL)
asciilifeform: and if this were the only such example!
phf: asciilifeform: last time i said that cons structures and sequences solve two entirely unrelated problems. cons specifically is a fundamental memory management abstraction for a von neumann machine (it solves the insert problem), so has its own set of operations that can predictable performance and behavioral characteristic.
mircea_popescu: this otherwise defensible stand (at least in general) is why computer languages may never become more than shopping lists ; but then again fashion-behaviour (as seen in both languages and github/reddit) may have the last word.
asciilifeform: and there are entirely good 'scheme is skeletal to the point of complete unusability in factory state' arguments
asciilifeform: i can definitely see the pov of, e.g., lisp2ists
asciilifeform: other than bloat
phf: lisp natively, you find all these nooks and crannies in the language that facilitate. nothing like that exists in scheme
phf: another take on this might be common lisp vs scheme. cl was standardized after the fact, existed and evolved on lisp machines. i'm looking at mit's cadr at the moment, and at a certain point you have maclisp, interlisp, zetalisp and "common lisp" all coexisting on the same machine, 10 years before the standard was written. scheme on the other hand was esparantoed for a purpose. the result is that as you move closer to speaking common
mircea_popescu: they're covering it too huh
mircea_popescu: you will ~never provide a better solution to the waterflow via running equations than via running the fucking water.
asciilifeform: then again finnish had ~0 written word until 19th c
asciilifeform: mircea_popescu: aha, urk 'ebonics' is perfectly speakable, just like the american variant, but (until recently) 0 written word.
mircea_popescu: yes, but they were idiots in another way : much like stalin's central planning committee (as implemented by the stuart court in london) can't beat the disorganised merchants of the low countries - just so five dorks with nary a clue can't compete with the combinatorial experimentation of the actual pigdin.
mircea_popescu: and your objection to "ukrainian" which is a lulz much like the "croatian" for instance doesn't come and isn't informed nor supported by "the poor quality of conventions employed". it's its lack of history that marks it, correctly, as nonsense.
mircea_popescu: but the historical matrix of language (on which for instance myth is merely encoding artefact of lived experience) is the key ingredient to language - which is why romanian, a language spoken by a people for many years is much more powerful than enlgish, a language spoken by much more mongrels, but not so long really.
mircea_popescu: yes, eventually, once v is so old to have been forgotten like the original calculations for the pyramids, robots will perhaps have enough linguistic history that you could pretend idiomatic-c is no less a language than latin.
mircea_popescu: language ~is a convention~, yes, but it is ~made from~ experience ; not consensus. people don't say "people" to denote people because ~they~ agreed to, but because people in the past ~have~.
asciilifeform: meat vitalism?! what happened to the robocalypse?!
mircea_popescu: the substance used*
mircea_popescu: viable conlang is impossible for the reason woman-cut-from-marble is impossible : the substance uses lacks the capacity for the intended usage.
mircea_popescu: anyway ; language reform sometimes works - but this because it was a language, not because it was reform. similarily to how cutting fruit trees works, but this is because they are trees, not because of the cutting. if you cut a broom whatever way it's not gonna make plums.
asciilifeform: i hate to distract mircea_popescu from a good gloat re the impotence of engineer, but i'd like to see where he cuts the subj : viable conlang is impossible because no one remotely understands meat well enough? or, useless, even if it were made ? ... or both ?
mircea_popescu: eh, hiding behind that to salvage your pretense to engineering power is exactly how and why engineer kids are always the ones who don't see any cunt at slut parties.
asciilifeform: the kibbutz was a bonsaikitten and deservedly vanished. but their conlang! worked!111
mircea_popescu: asciilifeform ok so then if you know, what are you on aboot!
mircea_popescu: nor do the average old women who put forth their usual menopausal engineering pretense within my earshot and get treated to a large helping of stfu and make sandwiches.
mircea_popescu: the result was that 3rd generation girls reverted naturally to the coy behaviour, and the whole thing is today forgotten to the point you don't know about it
mircea_popescu: bear in mind that at the height of jewish idiocy (well, 2nd height, after being stupid in the 30s) manifested as israel socialism, uppity bitches pretended to "teach" their daughters to be naked in public because "no shame" because hey, old woman engineer thinks herself powerful enough to repress the male gaze.
mircea_popescu: asciilifeform re the "taught their children" : engineer-parents are no fucking better than engineers generally. and again the dr moreau reference rules supreme.
phf: asciilifeform: so was Sanskrit, but at that point the line of "engineered" becomes very blured
mircea_popescu: i have no fucking idea why the postmenopausal worldview is even tolerated by actual engineers ; if i thought myself one i'd actually spend half my time having people a la "we'll make a language" whipped at the public stake
asciilifeform: catholic latin was every bit the engineered.
mircea_popescu: anyway. an airplane is very strictly not a bird without feathers ; there's no relation to a bird whatsoever, an airplane is a sort of bus not a sort of bird.
asciilifeform: ditto what the ukrs speak on orc tv (~none of the public heads speak anything but ru at home)
asciilifeform: i'll point out that there is also an 'if it prospers, none dare call it treason' effect going on. modern hebrew, for instance, is nearly as much a conlang as esperanto.
asciilifeform: i still wonder about the supposed 4 mil or how many esperantists, who taught their children, etc. was it disinfo ? and if not, where did they go ?
mircea_popescu: aha. and the results - informative.
mircea_popescu: asciilifeform a language is not a car nor an aeroplane. also engineers are not nearly as important ; or powerful as they like to pretend.
asciilifeform: that there lived no one remotely qualified to touch the problem, does not make it a nonproblem
mircea_popescu: the ignorance of random john smiths with "progressive" delusions of self importance is scarcely a basis for existence.
asciilifeform: or airplane without feathers
mircea_popescu: it's not a ~real~ language, but a very pathetic ersatz.
mircea_popescu: asciilifeform tbh, the reason a "constructed language" is readily forgotten is precisely the fact that it was constructed.
asciilifeform: in other vintage lulz, https://tonyarcieri.com/volapuk-a-cautionary-tale-for-any-language-community << esperantists had their own python3, apparently.
mircea_popescu: and in other argentine wtf, /me felt like putting some of the grandiose coffee available here to good use as coffee liqueur. sent girl to farmacy, where she bought 1 liter bottle of pure ethylic alcohol (98%). for like 6 bux. it's fabulous, tastes mildly of corn.
mircea_popescu: lol this is like cinderella on monkey island already ; they keep bringing her all sorts of offerings. "boogers ?" "how about some feces ?" "dried feces from yesterday ?" "how about this corn kernel" "ok how about dead molluscs"
asciilifeform: thestringpuller: 'people'
thestringpuller: because whitewashing security is the only way to make people have trust these days?
trinque: Teechan is similar in design to the Lightning Network, save for one crucial differentiating factor: it leverages trusted execution environments (TEEs), that is, secure hardware components found in recent commodity processors such as the latest batch of Intel CPUs with Software Guard Extensions (SGX). << why the fuck even bother talking about this
asciilifeform: iirc this is what they moved mike hearn to
asciilifeform: 'next, bestest ethertardium, now seeek000red with fritz chip'
asciilifeform: thestringpuller: i have sincerely nfi what is emin or what is an sgx
thestringpuller: asciilifeform: Emin is pushing SGX like solutions. Seems everyone is infected with security theatre these days.
mircea_popescu: anyway, yes, it'd seem iris fs took over the disks meanwhile, who knew.
asciilifeform: unlike the db switch
asciilifeform: then O(1) !!11
mircea_popescu: also the issue of performance discussed by dave chinner is worthy of consideration (in trinque 's link)
mircea_popescu: impressive. alrighty then
mircea_popescu: trinque ah, you can recompile with flag and it makes them 64 ?
asciilifeform: afaik there is ~no~ fs that gives 'as many of ANYTHING as disk will hold'
asciilifeform: so it'd appear that there is no suitable fs in common use.
a111: Logged on 2016-12-22 07:32 jurov: also, there is limit of 2^32 inodes, even in 64-bit mode
mircea_popescu: (for the record, in practice the utility of prepared statements is often nil, and can be mostly captured through saying eg insert in x values y, values z rather than insert in x values y, insert in x values z. literally the whole benefit is that it compiles the part prior to values just once.
a111: Logged on 2016-12-22 07:32 jurov: also, there is limit of 2^32 inodes, even in 64-bit mode
mircea_popescu: http://btcbase.org/log/2016-12-22#1588254 << can this be rewritten ? not that we're liable to have over 4bn txn at any point, but more of the principle of the thing "fuck you and your fucking magic numbers. if i run 64bit processors i'll run 64bit disks also motherfuckers!"
mircea_popescu: no but srsly, people think this. as per http://trilema.com/2014/la-florida-and-other-places/ ; the whole unwashed orc hordes of the third world actually believe there's something to usgistan.
mircea_popescu: (while the store-by-hash txn scheme promises at most 256 directories OR 65536 files however in practice finding 65536 contiguous hash txn seems rather unlikely for a while yet)
mircea_popescu: unless you want to store them by hash, in which case of course it's at most 65 chars, though because of difficulty i expect it should be made 40 or less
mircea_popescu: jurov in principle the address of a block is something like /44/4600 for the latest, so 2char dir + 4 char name
jhvh1: shinohai: The operation succeeded.
davout: ben_vulpes: what's the sqlator?
shinohai reads the giant logz
shinohai: neato ben_vulpes (on the lisp V)
ben_vulpes: BingoBoingo, pete_dushenski: in re graphite, "meaningfully stand behind their Q" perhaps?
ben_vulpes: dun worry, no such is on the table
jurov: also, there is limit of 2^32 inodes, even in 64-bit mode
jurov: mircea_popescu: ext4 has 256Binodes, and "The target of a symbolic link will be stored in inode if the target string is less than 60 bytes long."
mircea_popescu: moreover if you say press kimkardassian and it says "failed - parishilton not found" then it is somewhat likely you forgot to sign blondy.
ben_vulpes: i expect it to be a royal pain to run down "which patch is missing sigs and breaking the flow from genesis to whatever"
mircea_popescu: http://btcbase.org/log/2016-12-22#1588097 << dun sweat it mod6 ! tomorrow is another friday!
mircea_popescu: should it also show you the contents of your catpics folder, in case maybe you wanted to do some clicking ?
mircea_popescu: http://btcbase.org/log/2016-12-22#1588093 << why the fuck. if they're lacking sigs they're not even patches.
a111: Logged on 2016-12-22 02:28 mod6: it sounds like everyone wants instead, a general overhaul to get to the 'wot variant press' instead, which would also fix the bug, because these vpatches, without a corresponding seal, would simply be ignored.
a111: Logged on 2016-12-22 02:04 ben_vulpes: and sure, if .wot is the empty set, return true ;)
mircea_popescu: http://btcbase.org/log/2016-12-22#1587972 << it's altogether unclear where did you get this "celebrated is its ability to support a scientific dialog" ; got a link or something ? seems more the case what was celebrated was its ability to put a wrench through imperial-style "scientific dialog" where "we dindu nuttin wrong".
BingoBoingo: Don't forget the dirty Pascal panties!
ben_vulpes: someone else entertain the man
ben_vulpes: they're barely-compiling c tools in unix land
ben_vulpes: there *is* an ansi standard for sql
mircea_popescu: in practice i'm in two waters about it, wouldn't be surprised if the whole thing catches fire.
ben_vulpes: that is a very liberal use of the acronym
mircea_popescu: profile seek times, whatevs reasonable testage once the artefact's prepared.
mircea_popescu: aaaanyway. mimi's in a fine position to try this out, attach old drive, write script to spit them out as per schema, see what occurs.
mircea_popescu: well infinite here means "to fill the disk".
ben_vulpes: fill, sure. but an /infinite number/ thereof?
ben_vulpes: because every open source anything i've ever touched has failed in precisely these sorts of extreme use cases
mircea_popescu: afaik this is entirely unwalked ground, for all the foss bs about million eyes.
ben_vulpes: sure yeah, i remember the design pretty well
mircea_popescu: you'd end up with (as proposed there) a directory tree like 20 layers deep, with thousands of symlinks in each.
a111: Logged on 2016-06-13 23:48 mircea_popescu: and there's symlinks if anyone wants to alias.
mircea_popescu: ben_vulpes the bright idea was http://btcbase.org/log/2016-06-13#1481832 and the test needed is to check wtf happens if you symlink the txn to the block tree
a111: Logged on 2016-12-20 13:18 mircea_popescu: btw mod6 ben_vulpes trinque re the whole db/fs etc discussion, anyone recall the symlinks method / proposed tests ?
ben_vulpes: http://btcbase.org/log/2016-12-20#1586531 << i don't recall the proposed tests, actually. i mulled on this for a bit, am reluctant to try any sort of implementation until i finish the sqlator which a) is probably just sunk cost fallacy rearing its head, as i've done not much there but design the schema and prep a massive ingest job and b) has now been bumped down my todo list *again* in favor of vtronic
a111: Logged on 2016-12-22 01:49 phf: well, last time i brought it up with mod6 he said something along the lines of "i'm not ready to sign, because it's still work in progress"
ben_vulpes: http://btcbase.org/log/2016-12-22#1587935 << upon review of mine, it simply sucked horrendously and i suspect that i knew at some level at the time.
BingoBoingo: locally Praxair is known for exploding (literally) in the 1990s
a111: Logged on 2016-12-22 01:52 asciilifeform: trinque: https://upload.wikimedia.org/wikipedia/commons/thumb/1/10/Elektronika_MK_52_with_ERROR.jpg/400px-Elektronika_MK_52_with_ERROR.jpg << them
mircea_popescu: http://btcbase.org/log/2016-12-22#1587945 << ah there it is. so alf, of course my head read "kleptronica" for some reason.
a111: Logged on 2016-12-22 01:52 ben_vulpes: r sounds like g to them or something
BingoBoingo: Seriously. Anyways the Granite City and Gary Mill are as described, opened enough to get machines warm.
mircea_popescu: BingoBoingo there's absolutely no way in hell anything in the us can compete with mittal.
mircea_popescu: (not that this consideration has much practical value - outside of #trilema the community to do such thing doesn't exist, and so wasn't at the time an option)
asciilifeform: fwiw i deliberately did not polish my vtron, wajted to give folx a turn in the sun
mircea_popescu: bitcoin'd certainly be infinitely stronger were it the result of a concurrent development effort in this style.
mircea_popescu: so there's no real strong categorical difference from theory. but in practice it sure as fuck exists.
mircea_popescu: and on the other hand, i suppose it's entirely possible that if a years-made-wiser satoshi tried to release bitcoin, it'd have been done in the manner of how we did vtrons not in the manner of how he did bitcoin, ie, "here's what should happen - anyone who wants to participate make one"
mircea_popescu: i suppose it's possible that as technology matures it goes from this stage to "yawn, whatever, just use the cannonical version". so there's possibly that path there.
a111: Logged on 2016-12-22 01:58 asciilifeform: there is a 'harem-v' and a 'forum-v'
mircea_popescu: i perceive no benefit to, eg, getting everyone to use mod6's or alf's or anyone else's v implementation, as opposed to the present situation ; just as i perceive no advantage to getting everyone to make "his own" trb.
a111: Logged on 2016-12-22 01:47 phf: ben_vulpes: this subthread since your response to my original statement is one example of what i'm talking about. in this case none of the v implementations are on btcbase, because nobody wants to sign own hacks, because the cost of failure is too high.
mircea_popescu: http://btcbase.org/log/2016-12-22#1587923 << i am fwiw satisfied that it's qutie mroe than this : vs aren't on btcbase because they don't fundamentally belong on btcbase, because unlike public trb "we all use this" they're private "my girl will dance the way ~i~ want her to dance and stfu". there's a much more limited set of rules re what vtrons should do ; than re what trbs should do.
a111: Logged on 2016-12-22 01:35 asciilifeform: trinque: are you thinking of mod6's minor bug , or of some other
mircea_popescu: http://btcbase.org/log/2016-12-22#1587892 << fwiw first girl (there quoted) put my sig in the dir ; but for subsequent passes the item was rewritten to skip unwotted sigs rather than die.
a111: Logged on 2016-12-22 01:22 ben_vulpes: just because mircea_popescu didn't complain about the failure at the time doesn't make it right.
mircea_popescu: i was saying "this is working correctly", did it end up reading the opposite ?
mircea_popescu: http://btcbase.org/log/2016-12-22#1587847 << fish out the links, should be informative.
deedbot: mircea_popescu rated phf 4 at 2016/05/17 03:19:21 << his lordship the lord chancellor
mod6: i mean, we could do it that way... where the printable flow tells you which do not have sigs, but then impl-wise, they must be two different lists as to not inadvertantly press out a WILD vpatch. that, or re-check the flow at press time.
ben_vulpes: and this can only be done by calculating the dag and then indicating who signed which patches.
asciilifeform: i am failing to discern what, if anything, is in dispute here, so i will bbl. possibly it will be more evident from the log.
mod6: because why would i want gavin's vpatch stuck in the middle of my flow, if he's not in my wot?
mod6: <+ben_vulpes> flow is simply the directed acyclic graph of patches, and is *not* predicated on wot contents. << disagree.
asciilifeform: the only time a vtron must hard-fail, is when it is impossible for it to operate in finite time, i.e. the case where it detects a graph cycle.
ben_vulpes: flow is simply the directed acyclic graph of patches, and is *not* predicated on wot contents.
mod6: if you were to print the flow, it would still not show up.
mod6: <+ben_vulpes> pressable being operative word there. << it is inconsequencial.
ben_vulpes: the antecedent chain can be constructed without ever needing to refer to the signing of patches, and imho should show all patches in the flow, alongside which keys signed them and if none then marked as wild
mod6: this is why they exist in the first place, these WILD vpatches, because my impl wasn't written with this in mind. i was more written with the idea that a guy would place things in .wot/.seals/patches by hand and would know what is what.
ben_vulpes: "flow" refers to the antecedent chain, nothing else.
ben_vulpes: pressable being operative word there.
mod6: they will never, ever show up.
ben_vulpes: well, hang on. if patches with no sigs are omitted from flow, they won't show up as WILD in the flow, correct?
asciilifeform: mod6: 'leave out of all possible flow' is the correct answer.
ben_vulpes: (and in the absence of a WILD boolean, asciilifeform's pain-receptor-switch)
ben_vulpes: provided the implementation fails if any patch in the flow has *no* signatures from keys in .wot, that sounds right.
mod6: i apologize for this oversight about the WILD patches.
mod6: it sounds like everyone wants instead, a general overhaul to get to the 'wot variant press' instead, which would also fix the bug, because these vpatches, without a corresponding seal, would simply be ignored.
asciilifeform: trinque: there were 2 separate cases where this happened. mod6's buggy vtron (which he fixed today), and mine when the self-appendectomy-time-turn-off-pain-receptor switch flipped.
trinque: and I was all hot and bothered
asciilifeform: a seal, on the other hand, floating about without an active pubkey, for for that matter without a corresponding patch, is inert.
mod6: i think, he's saying, what is the benefit of V honking when it doesn't find a key in your wot that matches a seal in your seal dir, provided that you don't pull a mod6.
asciilifeform: trinque: a patch, so long as there exists 1 seal for it, and that seal corresponds to a key in your active wot -- is valid.
ben_vulpes: let's rewind: what does trinque miss when v finds a seal for a vpatch for which it doesn't find a key and proceeds merrily, provided it does find *a* seal for the patch that corresponds to a key in .wot?
trinque: make another tool
trinque: "push this button to make it stop hurting" has no place in the tool
trinque: V is a harsh constraint upon the programmer that says that his acts will be unavoidably attributable to him, and those that vouched for him.
trinque: I put it there and chemical reactions happened and so on.
trinque: one can say the purpose of the bomb was to explode over there but that's ridiculously backwards
trinque: attribute changes to the definitions of words with your ass
ben_vulpes: these things?
ben_vulpes: what purposes are these
ben_vulpes: i will motherfucking *not* shuffle both patches/* and .wot/* around when i want to press. this is stupid and carves off a whole space of adjacent possible.
ben_vulpes: i get the impression i disastrously fail to understand your point, trinque .
trinque: but first the definition
ben_vulpes: one can have either "die on discovery of seal for which wot lacks key" or "wot variant press"
trinque: otherwise yes, asciilifeform is right that if this doesn't matter, just have a thing that presses patches with hashes in the m
trinque: http://btcbase.org/log/2016-12-22#1587898 << this destroys the ~who~ of each seal, which matters
asciilifeform: when we discuss in the logs, it can be many things, but not formal definition.
trinque: I see other folks here discussing in logs.
asciilifeform: formal definitions don't live in the logs