log☇︎
51600+ entries in 0.334s
mircea_popescu: an autistic child you didn't kill at the age of 7 will support a whole comicbook industry until age 70 ?
asciilifeform: there was a famous incident of a d00d who went around looking for bears
mircea_popescu: this has apparently become a major psychological liability for the crowd, they imagine that since everything they ever encountered was safely behind bulletproof glass, that's "just how life is".
asciilifeform: it was in a trilema
mircea_popescu: there's a great gif with a stupid little girl confronting a lion from behind a glass pane.
mircea_popescu: but anyway, apparently taking a few heads here and there creates exactly no conception of own vulnerability in the rest of the heads. ☟︎
asciilifeform: http://btcbase.org/log/2017-10-13#1725038 << every time i sit down to think about this megaproblem , inevitably come to the conclusion that the threeway cut was a mistake ☝︎
asciilifeform: it is, to lift from another mircea_popescu article, 'a thrown brick, not airplane'
mircea_popescu: though honestly, may take a while for it to become a problem.
shinohai: I can't remember what I did with it lobbes .... but there was a plugin that let you announce a msg only to folks subscribed to it.
lobbes: That is a point. I can always tweak the announces to be less frequent as people see fit anyways
mircea_popescu: the forum happens in the marketplace for a reason.
mircea_popescu: tbh, i don't think doing it in #trilema is either unfeasible or a bad idea.
lobbes: Also gotta figure out where auctions would happen. I'd imagine #t wouldn't be keen on the auctions taking place in this channel. Would be a lot of added noise.
mircea_popescu: mental warming is a thing.
trinque: mhm. node 1 sits on a domestic line; 2, commercial. I could see it
mircea_popescu: and EVEN IF maintaining a large (and, obviously, expensive) network of nodes etc, you can still run into surprises, as the bitbet issue very well illustrates.
trinque: mod6: I did lose a transaction in a node's mempool the other day using it. was clearly there, sitting, going nowhere.
mircea_popescu: when we're discussiong classes we're not automatically discussing an impl. it's a data model.
asciilifeform: sooner turn pig into a professor of mathematics, than cpp trb into sanity.
trinque: I do owe a few patches at this point. btw, am using sendrawtransaction with deedbot, works so far.
mod6: Submit me a vpatch, Sir.
mod6: ah yeah, i do have a rawtx on sitting here -- just needs testing. also, still have some makefile changes that are needing some testing/review. and further, want to put in some requested changes to import/dump priv key.
mircea_popescu: a, was he ?
mircea_popescu: anyway, i tried /invite polarbeard but * polarbeard :No such nick/channel ; anyone know him irl drop a line ?
mod6: iirc I had created a proper vpatch of the thing last year, but we never wrapped up all the testing. i'd like to revisit that.
asciilifeform: trinque, mircea_popescu : this is actually a variation on an ancient and imho interesting theme, the 'aeternity service' or what was it called, early attempt at 'bitcoinism' and p2p warez combo, 'proof of storage'
trinque: must also provide a web endpoint for the archive, etc
mod6: mircea_popescu: understood. it is more a learning process than anything at this point.
trinque: actually mircea_popescu's solution to forcing miners to hold all blocks could conceivably used to hire hosting for the archive from whoever has space. folks register with eg deedbot, must produce hash of all archived items at interval to receive a small bitcoin payout, flowing from subscribers to the archive service.
asciilifeform: and understand it the way a man packing a parachute understands.
mircea_popescu: http://btcbase.org/log/2017-10-13#1724903 << this is why i wouldn't recommend writing tests for it for any other reason than learning / checking. it's a prototype yet, only the growth spur should be checked as it were. ☝︎
mod6: anyway, i don't wanna be a distraction. am simply, hopefully, on a useful side-quest.
mod6: <+asciilifeform> so there has to be a synthetic (i.e. Unsigned_32 ) type , that on archs that have narrower bitness than that, ends up a doubleword << ah, i see.
asciilifeform: archives are a necessary thing, but really oughta be auto-mirrored
mircea_popescu: asciilifeform afaik it was a scriba feature. Framedragger !!1111
asciilifeform: so there has to be a synthetic (i.e. Unsigned_32 ) type , that on archs that have narrower bitness than that, ends up a doubleword
a111: Logged on 2017-10-13 13:14 asciilifeform: http://btcbase.org/log/2017-10-13#1724860 << trinque was this a deedbot feature ?
asciilifeform: because 8 bits is not enough to index a useful number of 8-bit words ☟︎
mod6: a way of doing necessary work, and forcing myself to do the maff at the same time.
mod6: to understand the thing, one can simply do the: B ^ ((S - 1) & (A ^ B)) on a calc
mod6: so, that's what I'm spending time on now anyway. and, making good progress in a short amount of time. however, ya, many gaps i msure.
mod6: i believe there to be a lot of merit to having unit tests around the specific procedures and functions in ffa.
mod6: that's more of a higher-level, functional style test.
asciilifeform: mod6: the key imho re a ffa tester, is that it gotta exist ~outside~ of ffa, and use a traditional, known-to-produce-correct-numberz ( and ideally, more than one ) arithm stack
mod6: probably won't even start until we're closer to a finalized version of ffa.
mod6: once we have some solid unit tests in place, i'll look at some functional tests as a higher-level way to ensure we don't regress as well. but that's a bit later yet.
mod6: again, these are unit tests, not functional, integration, or performance tests. meaning: i simply call a (so far public) procedure/function within ffa with specific parameters, and expect specific outputs.
mod6: im certain that even the tests that I do have are missing certain cases -- but its a starting point. and when I publish these, the hope is that people here can help fill in the gaps of the coverage so we can use this as a means to ensure that we do not regress as changes occur to the code.
mod6: i don't have a ton of coverage yet -- just mainly around the fz bitwise ops, fz_add, fz_mult, handfull of others.
shinohai: Awesum ... I'm going to take a look at the thing asciilifeform posted last night later, gotta get Excedrin for preemptive strike on headaches lol
mod6: that's a mighty useful feature too btw.
shinohai: asciilifeform: think it was a scriba feature .... though bot appears to be present
asciilifeform: http://btcbase.org/log/2017-10-13#1724860 << trinque was this a deedbot feature ? ☝︎☟︎
mircea_popescu: this shit's a wonder to watch at work.
BingoBoingo: Now that's a crashing!
gribble: Error: "!Up" is not a valid command.
BingoBoingo: How many mac minis are necessary to bond together a 14 Gbps link to the worl?
saot: i'm not fan of a specific meme but throwing communists out of helicopters sounds fun
BingoBoingo: Ah, why not register a key? Make yourself at home?
BingoBoingo: So, who's your favorite? Any memes here you are a fan of?
BingoBoingo: Maybe one of these days hitting a rock bottom will give him an opportunity for a spiritual experience? Anyways, what are you lurking for?
trinque: mircea_popescu: I'm gonna post a tutorial on deedbot.org tonight, and also hook up the newb nick+key conveyor
BingoBoingo: <mircea_popescu> i don't get it, i can make a plain light colorized image of an ultraviolet source, but i can't make a sound pattern of some microwave ? << AHA, like "traffic safety" radar
ccmtacks: Noted, I've read a lot of the help material, so I actually get it how it all works now, well a good part of it :)
mircea_popescu: pay a coupla dumb hos who don't even hooker for a living to sit among some plastic shits at a "conference", spend 10x that much ?
mircea_popescu: trinque zinx ran me a grand or so so far, but i would way the fuck rather blow a grand on that than what the fuck is the alternative
trinque: this thing obviously pairs up with a buy/sell board quite well
mircea_popescu: anyway, simple way to sell anything for bitcoin : get a !!deposit from deedbot, give out that address.
trinque: nbd. there needs to be a whole tx weighing mechanism in order to let people know what that amount will be in the response.
trinque: ccmtacks: actual amount will be less the tx fee. I'll fire that off in a bit.
a111: Logged on 2017-10-12 20:52 mircea_popescu: trinque here's a thought : write me a branded single page "how to" somewhere, and ima modify the pay for your tits thing to only work with deedbot invoices. can then use it to push teh service, how's that sound ?
trinque: idea is to tighten the loop until it is a matter of <1min
mircea_popescu: ccmtacks sorry for all the hassle ; re-do once more, use a 1something address like god meant it.
trinque: asciilifeform: there's no hot wallet here; I'm marching my ass over to another box and making signed tx, then popping into a trb with polarbeard patch for sendrawtransaction
trinque: not eggog, it's a string
ccmtacks: mircea_popescu, I've received a message from deedbot that the amount of coins will be transfered to desired btc address
asciilifeform: how would trb pay a 3, lol
mircea_popescu: this way you have a passive "your books and my books gotta match yo" extra accounbting check there.
trinque: that sends a defined amount to a person, hasn't to do with invoicing
mircea_popescu: maybe have a magic !!pay joe all ? and btw, why is it not simply !!pay ?
ben_vulpes: asciilifeform: tiny set of scripts to get all of the source for a project from box a to box b
mircea_popescu: i don't get it, i can make a plain light colorized image of an ultraviolet source, but i can't make a sound pattern of some microwave ?
asciilifeform: wat's a source delivery pipeline
asciilifeform: 'questions of a possible microwave or radio wave device .... If that were the case, the sounds wouldn’t show up on a recording' << and anyone should believe the clip is genuine article precisely why.
shinohai: I'll be back in a week when the vomiting, dizziness, and diarrhea subside.
asciilifeform: they are almost certainly unnecessary on a pc, but can't put it past intel & friends to 'optimize' shl/shr, at the very least in the 'do nothing if operand is 0' sense.
BingoBoingo: Maybe can count a bit on directionality of airflow provided by other fans, but... cable induced turbulence is a thing
BingoBoingo: <asciilifeform> 'pcengines' racks a-ok << Gotta mind hot/cold aisle distinction for self defense from other's heat if otherwise passively coolable
shinohai: http://btcbase.org/log/2017-10-12#1724571 <<< I was almost afraid to look down, but nope - not a trap. ☝︎
asciilifeform: that's a max, for if you fill the disk cage
mircea_popescu: lumea s-a schimbat
asciilifeform: when asciilifeform was a sysop in early 2000s, the only racked item in the dc he ministered to that drew < 300 W, was a thermometer.
asciilifeform: mircea_popescu: you need a mb that actually has the sockets for ecc dimms
mircea_popescu: (note that the whole sert-up, ie that mobo, a modest am3+ chip, a coupla mem banks and some minimal diskage should draw <1oo W)
asciilifeform: ( you don't even strictly speaking need a u-case for it, can bolt 'ears' to the stock case )
asciilifeform: 'pcengines' racks a-ok
asciilifeform: i've drilled various rackmount for custom internals ( currently fitting FG, incidentally , into u1 ) and it is a manual carpentry
asciilifeform: mircea_popescu: in principle almost anything can be stuffed into a u1; ~literally~ anything can stuff into a u2
asciilifeform: while we're on subj -- the 'dumpster for ip, disks/cpu-horse elsewhere' gambit is (afaik pioneered, at least publicly, by mircea_popescu in mpex) imho a monumental thing