BingoBoingo: And I got to skip the awkward waiting between mandatory check out and check back in times at the hostel
jhvh1: BingoBoingo: Bitstamp BTCUSD last: 15301.0, vol: 13793.77734905 | Bitfinex BTCUSD last: 14997.0, vol: 57417.10285256 | Kraken BTCUSD last: 15316.5, vol: 2684.04490121 | Volume-weighted last average: 15065.3519288
BingoBoingo: Who could have predicted that bank holiday ends and crashing resumes!
deedbot: zen_ voiced for 30 minutes.
zen_: oh shit does that mean i have to chat that long?
BingoBoingo: zen_: No, it means you have that long to make a case for sticking around
BingoBoingo: Or issue a plea for the chance to make your case again
zen_: i have a case of chocolate truffles
BingoBoingo: Anyways with regards to your identity. Who are you, Where did you come from, How did you get here, and where to you imagine you are going?
BingoBoingo: zen_: Also do you have a blog, or any links to work which might impress the republic?
shinohai: The whole series thus far is very entertaining.
shinohai: Nice post BingoBoingo, only thing that offends my ears: "El edificios aca tiene huesos / Los edificios aca tienen huesos"
BingoBoingo: It appears my verb tense errors may be leaking bits of recent conversation context
BingoBoingo: In other news, god bless Peru for not having real datacenters. Sounds like a very inconvenient place to be right now.
shinohai: De nada .... what's happening in Peru, The Cuy is rising up?
BingoBoingo: The current president issued a compasionate pardon allowing former president to die in a hospital rather than prison. The pantsuit fake news is alleging riots. Possibility of USG.blue color shennanigans.
shinohai: Turns out that my supposition wasn't too far off, average Peruvian being about as intelligent as a cuy,
shinohai: Does local TV have sexy weather girl BingoBoingo ?
BingoBoingo: I haven't actively watched tv since Houston
shinohai:
http://archive.is/5AOg5 << Run moar prb ... "I dont know how i got hacked, i use my laptop just for coldstorage, i turn it on and all my BTC are gone."
shinohai: I don't think cold storage means what you think it means.
jhvh1: BingoBoingo: Bitstamp BTCUSD last: 15950.0, vol: 14013.56605120 | Bitfinex BTCUSD last: 15816.0, vol: 59791.06830403 | Kraken BTCUSD last: 16079.8, vol: 3298.34940167 | Volume-weighted last average: 15851.6396379
ben_vulpes:
http://btcbase.org/log/2017-12-26#1758614 << consider a genesis, a and b patchset where a and b don't touch the same files. if i press to b, write a c that descends from b, it will not have a as an antecedent. when you say "modify the filebase into the shape that you want it to be", do you mean include the changes of a in the c patch?
a111: Logged on 2017-12-26 13:30 BingoBoingo: zen_: Quien es su papa y que se hace el?
a111: Logged on 2017-12-26 13:32 zen_: i have a case of chocolate truffles
mircea_popescu:
http://btcbase.org/log/2017-12-26#1758666 << speaking of, /me saw fox something-or-the-other over at whorehouse/steakhouse yest. they had a chick with an ass/waist/boob range that was iiincredible, and then had her act like a retarded 12yo playing pattycake and whatnot with the blondy they had paired her up with.
a111: Logged on 2017-12-26 15:19 shinohai: Does local TV have sexy weather girl BingoBoingo ?
mircea_popescu: sometimes lowest effort as typical of tv pays off, i guess.
mircea_popescu: better than the "professional" conceit. really, someone got to anchor tv for their non-cleavage skills ? what am i to believe next, that bridges are for sale ?
BingoBoingo: Could anchor for non-cleavage acrobatics skills.
mircea_popescu: if one's stuck pretending something stupid, better off make it include being stupid rather than pretend it's possible to include some kind of being smart that doesn't readily reduce to being smrt.
mircea_popescu: "oh, i read cnn for the commentary, the boobs are INCIDENTAL. also, fleshhouse and playcunt, same deal. oh and those letters to the editor ? totes legit!"
mircea_popescu: "the arrangement leaves me stranded when eg trying to understand why hillary lost, but that's ok, i'm not really meant to understand things."
a111: Logged on 2017-12-26 17:52 ben_vulpes:
http://btcbase.org/log/2017-12-26#1758614 << consider a genesis, a and b patchset where a and b don't touch the same files. if i press to b, write a c that descends from b, it will not have a as an antecedent. when you say "modify the filebase into the shape that you want it to be", do you mean include the changes of a in the c patch?
mircea_popescu: the situation where a and b are pressed SEPARATELY is a fork.
a111: Logged on 2017-12-24 17:02 mircea_popescu: because it will cost someone's time to reconcile.
mircea_popescu: so as the reconciler, you get to pick which of ~either~ a ~or~ b to count in your considered oppinion as the republican and which as the heretic.
mircea_popescu: because nothing less than this is at stake ; and do not make stray patches if you can help it.
hanbot: in other puns, i discovered just this morning that "hippopotamus" is the latin horse + river, had missed out on the double pun fun of
http://trilema.com/2017/the-hippopotamus/ 's last line. i stand in awe and horror of whomever ends up adnotating trilema
mircea_popescu: hanbot no but see, THIS is what a classical education fucking is. not retarded fry making hobbit jokes because hey, the history according to fry started when he first got a bit of usg.corp fried chicken.
jhvh1: BingoBoingo: Bitstamp BTCUSD last: 15662.0, vol: 13962.96317052 | Bitfinex BTCUSD last: 15563.0, vol: 60770.99895801 | Kraken BTCUSD last: 15746.5, vol: 3400.44095335 | Volume-weighted last average: 15588.6777321
danielpbarron: photographed the unboxing to maybe post on blog when i figure this mp-wp thing out
danielpbarron: i'm familiar with apache but i was trying out nginx at first for this
mircea_popescu: might need mysql db installed separately if only comes with postgres or something
shinohai: mp-wp worked equally well on apache and nginx for me. Never used another db besides mysql tho
deedbot: seefelder voiced for 30 minutes.
deedbot: seefelder voiced for 30 minutes.
BingoBoingo: seefelder: Who is your daddy and what does he do himself?
seefelder: i am (NOT) a bot, reflecting human species
BingoBoingo: That's not very informative, you've expended both your strikes. One last chance to make a good impression.
diana_coman: seefelder, any blog where one can read what you reflected on so far?
seefelder: only locally stored ... i am maybe paranoid
phf: would that be considered "a coy behavior"?
mircea_popescu: "avoidant" as per teh "attachment theory" nonsense that pantsuit "public policy" rests upon
BingoBoingo: At the rate things are moving by mid 2018, the demand for openess to get a lend of voice is likely going to be self performed rectal ultrasound on mfc while reading logs
mircea_popescu: more practically, voice will be acquired via introduction by lord, ie, court presentation.
mircea_popescu: go make yer castles today, so noobs have a shot at life in 2018
BingoBoingo: Ah the lols to come from venue shopping n00bs
BingoBoingo: One doesn not see the city in the splendid glow of burning metal salts in the sky without missing them.
BingoBoingo: There's chicas everywhere so if that fell through no big deal... The concerted effort put for by the locals on the Latino Christmas show, I lack the patience and means to make that happen on a whim.
a111: Logged on 2017-12-26 19:46 BingoBoingo: At the rate things are moving by mid 2018, the demand for openess to get a lend of voice is likely going to be self performed rectal ultrasound on mfc while reading logs
mircea_popescu just dumped a whole sack of coal on teh grill, teh girls happily cleanning up a grosse of pleurotus... if anyone needs me ima be at teh rancho bbqing.
deedbot: borstal voiced for 30 minutes.
ben_vulpes: it's still 2017, borstal, what do you have to show for how you've lived your life?
borstal: Right Banks, you bastard! I'm the daddy now, next time, I'll fucking kill ya!
mod6: TMSR~: mod6.net is back up. Thanks.
ben_vulpes: where's your holiday spirit, diana_coman
diana_coman: ben_vulpes, he started sounding too much like mpi-insane
shinohai: This is hilarious on thepiratebay website, on the donation footer says: "Bcash: LOL"
trinque:
http://btcbase.org/log/2017-12-26#1758679 << to my mind this suggests the hashes present in vpatch blocks are wrong. if they were rather the hash of the entire concatenation of the diffed item before and after applying that block, there'd be no need to pointlessly edit files to continue on the same branch.
a111: Logged on 2017-12-26 17:52 ben_vulpes:
http://btcbase.org/log/2017-12-26#1758614 << consider a genesis, a and b patchset where a and b don't touch the same files. if i press to b, write a c that descends from b, it will not have a as an antecedent. when you say "modify the filebase into the shape that you want it to be", do you mean include the changes of a in the c patch?
trinque: further, pointless edits are antithetical to v-tronics
trinque: "whole patch must be meaningful"
ben_vulpes: it may be that tracking hashes at the file level isn't the right thing, because with vdiff as implemented i can trivially generate a vpatch b while working from a press of a that nevertheless still only depends on genesis if a leaves files untouched and b touches those untouched files only
diana_coman: I ran a bit into that with eucrypt but I can't say I have some clear idea what would be a good solution there as it seems to me it's a matter of level considered i.e. file, folder, component, whole project, what
mircea_popescu: trinque this is actually not altogether a meritless point.
ben_vulpes: so far the solutions i see are a) mircea_popescu's "what tree, all patches must descend linearly or be reground into linearity" 2) allow for pressing heads that do not touch the same files
diana_coman: onth "meaningful" there doesn't refer to code only and if new code in new files depends on existing code in untouched existing files, it may be meaningful nevertheless to update the readme file of the project for instance to say as much
mircea_popescu: then if it is, trinque's point goes away because well... file hash as good as project hash for the linear bat.
ben_vulpes: this still leaves me in the pickle of producing a vpatch from a press to a that won't actually descend linearly from a without touching a file, and adding "this line necessary to ensure this vpatch descends from a and not genesis"
phf: fwiw later approach has been a sop so far
ben_vulpes: phf: which vpatch does this show up in?
phf: ben_vulpes: official trb releases touch many files to tie them all into a single endpoint, e.g. `makefile' did that
shinohai: Just in time for the holidaaaaaaaaaaays fake bitcoin returns
ben_vulpes: phf: this exposes another problem: in ideal vtronics it is an illegal operation to press to multiple leaves at the same tree level, which is the only way to have a codebase against which to create a makefiles-type patch that ties multiple patches together
phf: we also at some point had a thread, where i believe ascii but also others were leaning towards the idea of a single file vpatches (i.e. that a vpatch should only ever contain hunks for a single file). i'm starting to think that multi-file solutions in general are a hack ("we can't fit the entire compilation in memory"), but then i've been looking at TeX on one hand, and the "millions of support files" in diff/patch on the other
ben_vulpes: also, single-file vpatches drives the system towards whole sourcetree hashing, otherwise they'll definitionally never depend on one another
phf: i've also played around with treating vpatch as a container. there's a function on btcbase for splitting all of the vpatches into separate objects, that each hold a single file changes (it's called something like explode-vpatch)
phf: in which case your graph connections become mandatory transitions (rather than what it is now, where you can technically press two different vpatches against a non-overlapping sets of files)
phf: i don't remember if there was any kind of interesting insight from the resulting graph, i should revisit it..
trinque: danielpbarron: your boy MTW all set?
danielpbarron: i think so, but it'd be nice to get a confirmation from him in here
danielpbarron: relatedly, l0de expressed interest in running a "trilema infomercial" on his show. I told him he should bring it up in here, as i'm not sure what that should entail
deedbot: l0de voiced for 30 minutes.
l0de: Yes hello, l0de here
l0de: the funds have been remitted to my account
l0de: I suggest the most serene republic air some sort of introductory statement
l0de: to get viewers of my show interested in their political goals
danielpbarron: would that be a pre-recorded audio clip/ video that we provide?
l0de: Or you can have someone call in to read a statement
l0de: I would stress that my audience doesn't take anything seriously unless it's presented in a humorous context or it is presented as ruining someone's shit
l0de: as they are internet neer-do-wells
danielpbarron: perhaps we can get one of shinohai's cam whores to do it
shinohai: Assuming I can find one that speaks English nowadays ..... xD
l0de: oh yeah, video chat with a cam whore would work perfectly
l0de: no nudity though, youtube is run by craven puritans
l0de: someone with a thick romanian\cambodian accent reading a statement they obviously do not understand
danielpbarron: wasn't indiancandy/sofiababy looking for a way to make bitcoin?
trinque: shinohai, you are the baron titsbare himself; got any camhos that can read?
shinohai: hmmm danielpbarron .... sofiababy may be good choice (though not one of my stable)
deedbot: l0de voiced for 30 minutes.
phf: in unrelated lulz, knuth shares mp approach to discipline:
phf: If you have only the present report, not a tape, you will have to prepare files WEAVE.WEB and TANGLE.WEB by hand, typing them into the computer by following Appendices D and E. Then you have to simulate the behavior of TANGLE by converting TANGLE.WEB manually into TANGLE.PAS ; with a good text editor this takes about six hours.
a111: Logged on 2017-12-26 21:40 phf: ben_vulpes: official trb releases touch many files to tie them all into a single endpoint, e.g. `makefile' did that
a111: Logged on 2017-12-26 21:42 phf: we also at some point had a thread, where i believe ascii but also others were leaning towards the idea of a single file vpatches (i.e. that a vpatch should only ever contain hunks for a single file). i'm starting to think that multi-file solutions in general are a hack ("we can't fit the entire compilation in memory"), but then i've been looking at TeX on one hand, and the "millions of support files" in diff/patch on the other
a111: Logged on 2017-12-26 22:02 danielpbarron: relatedly, l0de expressed interest in running a "trilema infomercial" on his show. I told him he should bring it up in here, as i'm not sure what that should entail
mircea_popescu: my original thought was "can we do something like the tits drip but for sw enthusiasts instead ?" ; alf pointed out thatr well, carrierless issue.
mircea_popescu: if anyone's got an idea of how to construct the item though i'd much like to hear.
mircea_popescu: basically something like "do x, get y" with a well chosen x.
a111: Logged on 2017-12-26 22:14 l0de: Or you can have someone call in to read a statement
a111: Logged on 2017-12-26 22:19 shinohai: Assuming I can find one that speaks English nowadays ..... xD
a111: Logged on 2017-12-26 22:22 danielpbarron: wasn't indiancandy/sofiababy looking for a way to make bitcoin?
mircea_popescu: if we had a chinese girl that'd work, except, of course...
shinohai: Yeah sofiababy left in huff, just noticed she had pm'd me to find her on skype but cba with that.
a111: Logged on 2017-12-26 23:05 mircea_popescu: basically something like "do x, get y" with a well chosen x.
a111: Logged on 2017-12-26 23:04 mircea_popescu: my original thought was "can we do something like the tits drip but for sw enthusiasts instead ?" ; alf pointed out thatr well, carrierless issue.
a111: Logged on 2017-12-26 23:02 mircea_popescu:
http://btcbase.org/log/2017-12-26#1758790 << especially in the light of 1. move/rename 2. edit double-patch method this seems more and more like the right thing ; but then again what to do of regrinds ? regrind should ideally be "one patch out of many"
a111: Logged on 2017-12-26 22:06 l0de: to get viewers of my show interested in their political goals
shinohai: Socialism at it's finest right here.
a111: Logged on 2017-12-26 21:42 ben_vulpes: phf: this exposes another problem: in ideal vtronics it is an illegal operation to press to multiple leaves at the same tree level, which is the only way to have a codebase against which to create a makefiles-type patch that ties multiple patches together
a111: Logged on 2017-12-26 21:36 ben_vulpes: this still leaves me in the pickle of producing a vpatch from a press to a that won't actually descend linearly from a without touching a file, and adding "this line necessary to ensure this vpatch descends from a and not genesis"
a111: Logged on 2017-12-26 21:42 phf: we also at some point had a thread, where i believe ascii but also others were leaning towards the idea of a single file vpatches (i.e. that a vpatch should only ever contain hunks for a single file). i'm starting to think that multi-file solutions in general are a hack ("we can't fit the entire compilation in memory"), but then i've been looking at TeX on one hand, and the "millions of support files" in diff/patch on the other
a111: Logged on 2017-12-26 21:31 mircea_popescu: imo "tree" MUST be linear.
a111: Logged on 2017-12-26 21:22 trinque:
http://btcbase.org/log/2017-12-26#1758679 << to my mind this suggests the hashes present in vpatch blocks are wrong. if they were rather the hash of the entire concatenation of the diffed item before and after applying that block, there'd be no need to pointlessly edit files to continue on the same branch.
MTW: [16:55:48] <+trinque> danielpbarron: your boy MTW all set?
a111: Logged on 2017-12-26 20:29 mircea_popescu just dumped a whole sack of coal on teh grill, teh girls happily cleanning up a grosse of pleurotus... if anyone needs me ima be at teh rancho bbqing.
a111: Logged on 2017-12-26 19:48 mircea_popescu: more practically, voice will be acquired via introduction by lord, ie, court presentation.
a111: Logged on 2017-12-27 00:03 asciilifeform: or even suspect why he oughta.
a111: Logged on 2017-12-27 00:06 asciilifeform: let's say every homo redditus alive -- suddenly interested. then WHAT? what's gained, other than a great mass of meat that now gotta be put somewhere far from the reactor rods
a111: Logged on 2017-12-27 00:23 asciilifeform:
http://btcbase.org/log/2017-12-26#1758779 << i see e.g. trb tree, as the frayed end of a rope. in long term, observe, the loose ends that dun get built on -- fade away, like orphan chains. btc is actually more or less same kind of system. but iirc we had this thread.
a111: Logged on 2017-12-27 00:34 asciilifeform: ^ can anybody suggest a working variant ?
mircea_popescu: anyway, maybe the correct approach re the l0de thing would be something like a simulcast interview. you do interviews l0de ?
deedbot: l0de voiced for 30 minutes.
mircea_popescu: if you register a key you can self-voice don't have to keep doing this voicing thing
jhvh1: BingoBoingo: Bitstamp BTCUSD last: 15627.96, vol: 14963.98271799 | Bitfinex BTCUSD last: 15685.0, vol: 64040.84715043 | Kraken BTCUSD last: 15811.4, vol: 3687.11992386 | Volume-weighted last average: 15680.3140104
BingoBoingo: <mircea_popescu>
http://btcbase.org/log/2017-12-26#1758819 << she kinda got sent off for NOT being looking dedicatedly enough ; then she got pissy because i imagine in her dumb head she had self-delusions as to self-importance as dear as they were baseless. << Ah, gets BTC once returns years later and has less sense and more ideas
a111: Logged on 2017-12-26 22:22 danielpbarron: wasn't indiancandy/sofiababy looking for a way to make bitcoin?
BingoBoingo: Hence the spankings and vacuum cleaner tube
BingoBoingo: Anyways thinking on the noob roughing up earlier, could be a side effect of tranquility overdose. Gotta torture betas.
BingoBoingo: Will probably start upping the blog frequency to let some more vemon out
mircea_popescu: anyway, personally launched lb+ "polvoras" ie fireworks here.
mircea_popescu: damned thing was the size of a god damned depth charge
BingoBoingo: Eh, give it 10 years and everything Republican will be inconel while the last USG trinkets come in Chinese potmetal
trinque: it again runs into the unix idiocy of "file", though, because file does sometimes mean module, sometimes means subcomponent of module.
trinque: in either case, what does asciilifeform think of the ritual of "add comment to unrelated file to merge paths", not symptomatic of a problem?
trinque: answer could be "trb was fucked from birth, let us not allow backflow into V"
trinque: so then, when things get so modular that you have frayed rope, author makes clean separation, as V has told him he has distinct items.
a111: Logged on 2017-12-21 17:38 mircea_popescu: anyway, continuing the trinque discussion, it seems entirely unavoidable that trb will become 3 things : a wallet node, optimized for pumping out local signed tx ; a block node, optimized for keeping the blockchain, getting blocks, no mempool nonsense ; and a spy node, optimized to keeping track of the lies and nonsense flowing through the relay network (mempool, timing nodes, what have you).
trinque: what is the meaning of the hashes in a vpatch?
trinque: should someone pressing think he has a coherent whole after pressing any patch? if not why?
trinque: only way to make a 3rd improvement rely on two distinct improvements in past is to put cruft in both.
a111: Logged on 2017-12-26 21:22 trinque:
http://btcbase.org/log/2017-12-26#1758679 << to my mind this suggests the hashes present in vpatch blocks are wrong. if they were rather the hash of the entire concatenation of the diffed item before and after applying that block, there'd be no need to pointlessly edit files to continue on the same branch.
trinque: to my mind the hash is currently putting too narrow a constraint on the context within which the current patch is to be applied, where we could broaden the context at no detrimental, and possibly beneficial cost
trinque: asciilifeform: is possible, but deserves to be in the logs
trinque tends many fires, but knows he owes a few posts
a111: Logged on 2017-12-19 18:04 asciilifeform: we actually had the ast-diff thread, re phf's lisp diff lament.
trinque will put some pen to paper this evening
mircea_popescu: lmao fucktard. a) the "pirate party" scam utterly fucking failed, how about addressing THAT ; b) kinda hard to have anyone ditch an empty ship, huh.
mircea_popescu: "hey, aren't you worried your shitcoin will get altcoin'd in the near future ?" "no, i am dog and i don't understand anything. vote me!"
trinque: means for example that db-interaction routines are in one file, network interactions in another.
trinque: cannot without the 'add comment to files you needed' ritual
trinque: I'm actually writing right now on how the hypertext thing relates
mircea_popescu: let me put it this way, maybe resolves problem : v unpermits a specific kind of hack within the purview of this discussion, wherein one tries to design after the fact. correctly designed items will have the larger bits (by footprint) earlier in the patch tree ; and fray out correctly. github-style nonsensica commonly attempts to discover that "hey, this johnny come lately item should have been an evie-comes-early MODULE, let'
mircea_popescu: v doesn't permit this backwards in time ; and if you run in this situation where johnny needs an experts' sex change to become evvie, it's high fucking time you hand-rewrote your whole steaming pile of crap.
mircea_popescu: now, is this enforcement a problem with v, is the proposition ? or is the aforegoing a misrepresentation of the discussion ?
trinque: this is what I meant actually, by "can't modularize" within same walk of the v tree
trinque: trb's database interactions are properly in a filesystem-implementing walk of graph of human knowledge
mircea_popescu: well, i am kind of a fan of the whole "v doesn't permit you to lie to yourself about having supposedly designed what's utterly an ad-hoc item".
trinque: network in a generic gossipd press path
trinque: in that these are distinct items.
trinque: these aren't "trb" except in shitworld
mircea_popescu: there's nothing wrong either in principle or in practice with making a correct item as the genesis and then patching in various parts of trb.
mircea_popescu: ie, i don't expect the trb cut as described to have a trb genesis necessariyl, or even probably.
trinque: anyhow lemme run at it again. you can't modularize because you have to fake work in a disparate part of the tree to merge
mircea_popescu: so for instance, "genesis a proper db ; then patch in three different branches for the three different types of node envisaged" doesn't seem on the face problematic.
mircea_popescu: asciilifeform i intend to prove no such thing to no such martian.
mircea_popescu: under the authority of the republic ; not trying to enact selves out of the well threadbare wizard cloak of satoshi.
trinque: asciilifeform: as a carving tool for the graph of knowledge
trinque: doesn't mean you don't wholly remove sections and replace
mircea_popescu: whosoever deeply cares about the historically irrelevant accident of windows-bitcoin-0.1 is more than welcome to diff his own sources of that against tmsr-bitcoin
trinque: proper hypertext system (itself based upon v) provides the talmud commentary thing endlessly
mircea_popescu: relevancy is dearly bought ; man wanted to still be in the genesis of 2017, man should have made proper db calls, proper logs, etc.
mircea_popescu: man did not bother, which can only be rendered as "man told us in no uncertain terms to fuck off", well... the sentiment is mutual.
mircea_popescu: there's no difference i can observe between indiancandy scratching at the door and satoshi scratching at the door. there's a way to get in -- getting in "on their own terms" is not on the table at all.
mircea_popescu: as history ended up unfurling, "let's truncate hash to 10 chars or 20 chars depending" takes one from 2009 to some portion of 2018. better than nothing ; much less than could have been had, if only.
trinque: back briefly on the frayed rope, what's harmful about naming an antecedent that you didn't edit, but require
mircea_popescu: specifically "do a good job in preference of a bad job, last longer".
trinque: sure, in the vpatch would be "I require this list of antecedent items, subset S of which I intend to change thus"
a111: Logged on 2017-12-05 13:27 mircea_popescu: no but see, we use different terminology. i do not assign anything to "code written". the source of code, to my eyes, is he in the wot who has read it.
trinque: right, so if we could do this, name unchanged antecedents that are required, could merge A's log patch and B's db patch in C's subsequent patch without having to manually edit A's or B's
trinque: takes what constitutes context for the patch and puts it in the hand of the operator
trinque: I don't see that regrinding solves it
trinque: current V requires that a file actually got edited to be an antecedent, but C editing B's work does not mean he's discarding A's, and A regrinding his patch means editing something edited in C to get in
trinque: because the hashes are hashes of touched files
mircea_popescu: so suppose lord X makes tree A : A1->A2->A3->A4 are patches, delivering some kind of utility we don't care to specify.
mircea_popescu: now suppose lord Y comes along, and similarily makes B : B1->B2->B3->B4.
mircea_popescu: then along comes lord Z, and this lord Z observes that if he used B3 and instead of B4 installed C4 on the same top, he'd get a wholly different but entirely useful to him item. so he makes this.
mircea_popescu: at this moment, if lord K observes that he could use the tree of X up to A2 and the tree of Y up to A3 ~together~ he could install D4 on this pile and similarily to Z produce a different still useful item.
mircea_popescu: now, the v doctrine as it stands right now, both on logs and actual precedent, at least as far as i understand it (but this is vacuous both as a representation and as a history, as most important questions haven't yet been seriously tested) -- is that Z is right to simply sign a patch on B-genesis ;
trinque: I assume you mean A2 and B3
mircea_popescu: whereas K would be wrong to attempt same, and instead should regrind a whole new genesis, call it D, even if it is made up of the reunion of the As and Bs he uses.
trinque: why prefer this to being capable to merge?
a111: Logged on 2017-12-27 03:43 mircea_popescu: let me put it this way, maybe resolves problem : v unpermits a specific kind of hack within the purview of this discussion, wherein one tries to design after the fact. correctly designed items will have the larger bits (by footprint) earlier in the patch tree ; and fray out correctly. github-style nonsensica commonly attempts to discover that "hey, this johnny come lately item should have been an evie-comes-early MODULE, let'
trinque: hm. I may be learning something here, so bear with me.
trinque: as I cannot put a definition to merge that is not "destroyed vertex on this graph, because it was by my lights wrong, and created a new one"
a111: Logged on 2017-12-27 04:02 trinque: I assume you mean A2 and B3
mircea_popescu: trinque which is a valid thing to do, but NOT if one wishes to at the same timeeschew regrinding/genesising
mircea_popescu: and in unrelated lets-suck-our-own-cocks-we-utterly-deserve-it : consider that the whole l0de thing started because someone from here checked out a SPAMMED item. the fuctard/pantsuit "engineers" in name only in EVERY OTHER fucking channel ~think~ themselves all open-minded and intelligent and whatever, yet i can make a very obviously correct and banal prediction - they wouldn't have followed it, nor in any case escalated and
mircea_popescu: so on. because "it's spam" and that means "it shouldn't be read" and they actually have a consensus on this, which they idiotically but universally misrepresent as somehow different from any other cultish behaviour, such as believing "racism" or "global warming" or "witchcraft" are things.
mircea_popescu: somehow the voice model makes spam such a rarity in #trilema, people actually have the mental vigour to evaluate it!
trinque: specifically what people have been doing when "regrinding" is adding comments to unrelated files and thus including their patch in the tree.
mircea_popescu: whereas the pantsuit psychotic cleaving, where ~some kinds~ of spam are spam (ostensibly because they came from russian hackers as per their bayesian filters ?) whereas some other kinds of spam magicaloly "aren't spam" somehow, because pravda said it, or some "transgender" schmuck said it, or whatever.
mircea_popescu: trinque there's an ambiguity here i'm possibly responsible for though not intended : to "regrind", ie to take a pile of patches and make them into one single patch ; as opposed to re-genesis, which is what happened with eg mpi.
trinque: that is what I mean by a merge, and has the same result.
a111: Logged on 2017-12-27 01:52 trinque: only way to make a 3rd improvement rely on two distinct improvements in past is to put cruft in both.
a111: Logged on 2017-12-26 18:26 mircea_popescu: so as the reconciler, you get to pick which of ~either~ a ~or~ b to count in your considered oppinion as the republican and which as the heretic.
mircea_popescu: if the patch tree goes a1->a2->a3/a3` your position is now to choose which of a3 or a3` counts, and which doesn't count. the discarded one may be scavenged for useful content, but it will never be a proper patch.
mircea_popescu: (this results in an immediate reimplementation of eg's linus torvald's linux codebase management, except properly and per protocol rather than ad-hoc and in a manner nobody can explain or meaningfully defend)
trinque: yes, but the method of inclusion by diddling unrelated file is frivolous and less meaningful than explicitly denoting the relationship
mircea_popescu: there's still a disconnect because i don't understand what the hell you mean.
trinque: which is where I got to "concatenate whole cppwad and hash that" as that's your cpp program anyway.
ben_vulpes: a3 and a3' can touch a disjoint set of files and never be depended upon by an a4 without mutating unrelated files to ensure dependency is properly codified.
mircea_popescu: it doesn't matter which files they touch. a4 will build upon one of them, and then a5 on a4, and the unbuild upon one is left as a "fray" on the rope.
trinque: "I edited the networking code and added better logging statements which requires the better logging code on fray, but I didn't edit the logging code."
ben_vulpes: so then for a3' to be an a4, it must touch an unrelated file in a3, forever crufting up the codebase with v artifacts.
mircea_popescu: ben_vulpes oh it finally dawned on me what this is about. sorry it takes so long.
ben_vulpes snaps keyboard, retires, drowns children, shoots self
mircea_popescu: so the idea is, you got up to a2, which consists of files F1... Fi ; now one patch call it a3 touches file Fj, and another patch call it a3` touches file Fk
mircea_popescu: the ~only way to establish a lineage among these two so a3` is properly a4 is if the patch is spuriously modified to add a "hey v sucks" comment in Fj
trinque: totally, if I have to edit something to name it as antecedent
mircea_popescu: because otherwise, touching entirely different filesets, their precedence can not currently be established as per extant v
a111: Logged on 2017-12-25 22:56 ben_vulpes: the specifics of this case is that increase_aggression_levels touches *only* net.cpp and excise_hash_truncations touches a whole lotta stuff but *not* net.cpp
mircea_popescu: hey, i spent most of the intervening day revelring! i have circumstrances!
mircea_popescu: anyway, and the proposed fix for this is to actually add a hash for the whole filebase in each patch ?
ben_vulpes: moreover i want to bring up another overlooked point which is that it is illegal to press a tree with these two patches side by side
trinque is only recently not a walking solution of wine and nog
trinque: mircea_popescu: yep, concatenate every single item in the path diff processed, use *that* hash as antecedent and that recalculated as expected.
mircea_popescu: trinque kitten trying to get into the backseat so i can play with her tits ever so briefly kissed my new suit pant's leg, now i have a typically indicative white spot on it. tbh i knida like the look of it.
mircea_popescu: how about a convention whereby all new genesises must contain a manifest.genesis file, which file will be constantly patched on each patchj, no exceptions, by adding a line which reads : "This is patch #x and the codebase hash is blabla".
mircea_popescu: this has the advantage that you can readily understand what any press is made of by looking in root.
ben_vulpes: hash of the patched codebase including the patched manifest with hash of patched codebase in it?
mircea_popescu: no, it's signed. it simply is not used in the one spot where the codebase hash is calculated.
trinque: manifest can be the patch header nearly as is
trinque: "let it be known that there are these files, with these hashes" "I have changed these; their hashes are now ..."
mircea_popescu: anyway, this'll need moar discussions, i'm not specifiying anything on dec 26th.
trinque: and yet, I can see the entire thing from the other perspective still, that cpp is broken, trb itself not a single concept but a mud, etc
mircea_popescu: trinque gotta force emergence of sanity through some sort of rational process.
mircea_popescu: otherwise what, we rebuild africa, "sonny we sat here and marveled at this mud for 955 generations"
mircea_popescu:
http://btcbase.org/log/2017-12-27#1758974 << very much so ; and especially on a blog. there's utterly nothing wrong with being wrong, and discover it over time, especially if gracious about it. not like you sign the damned things, nor like the distinction isn't very fucking clear a a matter of public policy.
a111: Logged on 2017-12-27 01:58 asciilifeform: i'd like to encourage trinque to put some of his 'crackpot' algos 'to paper', as articles. the hypertext thing was interesting imho, for instance, and so was earlier trinque pill for 'mining is a bug', and possibly other occasions. dun be afraid to write down conjectures, trinque , gauss did
ben_vulpes: it'll need codebasehashprepatch and codebasehashpostpatch i think
a111: Logged on 2017-12-27 02:03 asciilifeform: or, on other end of the possible, a vtron that somehow understands that a call of foo necessarily depends on the patch that birthed foo... and requires disambiguation only if >1 foo exists in the tree
ben_vulpes: (it's an urbit, and presumably others crackpottery)
ben_vulpes: how is a4 to indicate that it needs both a3 and a3' otherwise?
ben_vulpes: putting the codebasehash in headers doesn't work then, as there is no 'file that will always be touched' that is a part of v to participate in the toposort
ben_vulpes: i'll have to doodle, cannot do this live
trinque: hm. the manifest also gives you a place to name blobs.
trinque probably at a point to digest also
trinque: but specifically, blobs not included. "and you will need the debian 2002 iso; go find"
trinque: ^ cuntoo direly wants this
mircea_popescu envisaged the genesis.manifest as wholly mechanical item, just a patch-per-line count of patches, no space to adlib.
trinque: my cuntoo installer script requires some what, 500mb of wads of other items that are not text, or useless.
ben_vulpes: mircea_popescu: okay, i geddit. do it as the first step of vdiff, so the mutation shows up
ben_vulpes: question then becomes how to get the patchtitle into .manifest
trinque: why not have one file, manifest, and you edit it, then vdiff the whole shebang.
ben_vulpes: oh you said "patch # and the codebase hash is..."
ben_vulpes: comedy option: vpatch names are now the hash of the resulting codebase
mircea_popescu: trinque because it'll get a mess ; ben_vulpes it's just a counter. increments 1 from prev line. shall i do a sample pastebin ?
trinque: can have a tool to edit it sure, but then that's the file that's being edited no matter what else is edited, and there's a coherent history based primarily upon a list denoting what's considered a thing at time of patch
mircea_popescu: trinque arbitrary-content manifests will become a mess of random.