log☇︎
257100+ entries in 0.169s
ben_vulpes: now that you mention it, 'log in' entails a link in yr email.
ben_vulpes: that just makes it harder to get their money later.
ben_vulpes: why would you ever want to log users out?
ben_vulpes: today, i am electing to not implement a logout button.
shinohai: Sorry BingoBoingo I left out "would be required" after the " invoices each year.” quote
BingoBoingo: In classic trilema http://trilema.com/
jhvh1: shinohai: The operation succeeded.
shinohai: !~later tell BingoBoingo http://ix.io/1ovm
mircea_popescu: and in other "we forgot how to do it" news, http://66.media.tumblr.com/9bb75d2478bb05786cfff6a5ba46278f/tumblr_mm8fu3y9ky1rmxgp0o1_1280.jpg
scriba: Logged on 2016-09-19: [19:58:22] <asciilifeform> 'If the miners had half a clue they'd have told Idiot Co-op exactly where to stuff it, and here's the beauty of it: with the arrival of ASICs and their significant capital cost, the odds that the sort of feelgood ninnies currently involved in mining will still be around are nil. ' << if only this had been !
mircea_popescu: http://log.mkj.lt/trilema/20160919/#655 << she was spot-on, actually. i don't think anyone who wasn't here can even grok what sort of copumpkin-esque imbeciles counted for miners back then. it's cleared immensely.
mircea_popescu: <mircea_popescu> http://log.mkj.lt/trilema/20160919/#652 << remarkable how ahead of her time she was.
shinohai: I think the whole place is inhabited now by human shannonizers that make mindless posts for 50 cents in satoshi a day.
shinohai: I know but there aren't even lolz any more, nary an ingenious scam to be found.
ben_vulpes: (another horribly formatted import from the old site, complete with broken images)
ben_vulpes: http://cascadianhacker.com/21_a-tour-of-bitcoind-booting-to-its-first-thread << i should do another of these!
asciilifeform did not frequent tardstalk when it was alive, never had acct there
asciilifeform: lel, i never even read this before.
asciilifeform: 'What there's need for is people to sit down with a cup of coffee and a (preferably printed) copy of the code and just read it through. This can be done in bits as long as the bits aren't arbitrarily segmented (it's ok to summarize a procedure, it's not ok to summarize between lines 520 and 545). Once we have a few of these completed we're already very far down the road.' << mpoepr
asciilifeform: 'If the miners had half a clue they'd have told Idiot Co-op exactly where to stuff it, and here's the beauty of it: with the arrival of ASICs and their significant capital cost, the odds that the sort of feelgood ninnies currently involved in mining will still be around are nil. ' << if only this had been !
asciilifeform: n in fiascoes like this one. Specification is the way out of it, and most importantly specification is the way out of having you idiots create fiascoes like this one randomly, one at a time, for the unforeseeable future. ' -- mpoepr
asciilifeform: 'Do you grasp this? Bitcoin will never exist as a toy for five idiots. You will never get to matter inasmuch as what you want to do is have this little black box the world reveres that only you are allowed to peer inside. This is not how the world works, currently (and past about 1800 or so). This is not how the world should work, either. Specifying the code does not "result in fiascoes like this one". Your idiotic codebase results i
ben_vulpes: contacts are definitely toast.
asciilifeform: pull the hose off the shop wall, wash ?
ben_vulpes will ooze hydrocarbons from the eyes for days, probably
ben_vulpes: found, last night. discovered its contents, this morning.
ben_vulpes: in other satoshisms, i found a bottle of contact lens solution that turned out to have high vitamin e oil for topical use in it
ben_vulpes: menial wwwtronix are just a matter of time
mircea_popescu: also true.
asciilifeform: and to call it mushroom is insult to mushrooms; it is fungus on public toilet.
asciilifeform: convince somebody to mine.
asciilifeform: well that there's ye olde 'high S' problem, aha
asciilifeform: mircea_popescu: the failure to generate 1MB tx is promisetronic, not protocolic, though.
scriba: Logged on 2016-09-19: [18:25:45] <asciilifeform> tx can't be any bigger than block.
phf: from the "wat" department https://www.thesun.co.uk/news/1799205/sex-change-soldier-is-britains-first-female-to-fight-on-front-line-after-being-born-a-boy/
asciilifeform: https://archive.is/1zGfH << the imho more readable original. for now.
asciilifeform: ben_vulpes: the formatting of ^ suffered greatly from the wordpressification.
scriba: Logged on 2016-09-19: [18:25:12] <ben_vulpes> did some late night back-of-the envelope on script length and miminum transaction size and i don't recall it breaching u32
mircea_popescu: http://log.mkj.lt/trilema/20160919/#534 << not since we drove stake through hearn-gavin shambler heart. but "in the future" of retardation, i'm sure there will be.
asciilifeform: it was closest thing there was to a 'textbook of v'
scriba: Logged on 2016-09-19: [18:24:07] <ben_vulpes> and the whole "if less than 253, that's the number, if equal then read the next octet, if blablabla" is that consistent across c-land?
mircea_popescu: http://log.mkj.lt/trilema/20160919/#528 << first time i ever heard of it. also fucking stupid, 253 = 11111101 lord have mercy.
asciilifeform: mircea_popescu: imho section ix ought to link to ben_vulpes's mega-article on subj
asciilifeform: it was and remains the best starting point for sanity.
scriba: Logged on 2016-09-19: [18:15:46] <asciilifeform> well the thing that presumably drew ben_vulpes to using 'binary-types' is the notion that 'describe the type and never have to manually craft readers/writers'
mircea_popescu: http://log.mkj.lt/trilema/20160919/#507 << iirc what drew him was that you told him to.
scriba: Logged on 2016-09-19: [18:13:11] <phf> the type is called compact size (in bitcoin src parlance), it is simply an number. the reader reads a byte, decides what to do, ultimately returns ~the number~
mircea_popescu: much like a "press", it's a personal take on the lightcone as-it-is.
asciilifeform: rather than the form in which it is representable in real time.
asciilifeform: my original point was that the 'linear' blockchain is very much an after-the-fact flattening
mircea_popescu: doesn't work, of course, but then again lemmings aren't looking for solutions ; merely for the appearance thereof.
mircea_popescu: in a sense getting rid of historical "orphans" is very much community-trying-to-insure against the nature of the blockchain.
scriba: Logged on 2016-09-19: [17:53:56] <asciilifeform> (you don't know that it will be orphaned until later)
mircea_popescu: http://log.mkj.lt/trilema/20160919/#461 << technically you don't know it will be orphaned ever, because "being orphan" is not a quality of a block/chain. if tomorrow we decide to extend an "orphan" from 2014 and in the process strand extant bitcoin, we ~can~.
BingoBoingo: <mircea_popescu> "dad i wanna get married" "honeybunch, you'll get old and your tits will sag and it'll suck. don't get married, it's a dead end" << LOL
asciilifeform: phf: interestingly, the ada folks got this right.
phf: i agree though, the numbers are unnecessary, i'll have to borrow the whole sizeof concept :>
jhvh1: asciilifeform: The operation succeeded.
asciilifeform: !~later tell mircea_popescu your new page's section IV oughta link to trb
phf: well, you write your heathen-octet-stream wrapper around your default sb-internal::binary-trit-tryte-input, so that read-byte on a stream gives you the right things
asciilifeform: well yes but the '4' !11111
phf: it doesn't matter, because on your box with 7-trit trytes you make your reader/writer infrastructure read it correctly
asciilifeform: !#s tritcoin
asciilifeform mutters 'what if i'm on a box with 7-trit trytes'...
phf: i think it aws en.bitcoin.it
phf: i'm trying to remember what i used as reference
asciilifeform: l0l! was this auto-generated from something ?
phf: (define-proto-structure txin ()
phf: (nil txout (compactSize-uint txout))
phf: (nil txin (compactSize-uint txin))
phf: (define-proto-structure tx ()
asciilifeform: would be interesting to read side by side with ben_vulpes's
phf: i lisp pasted it a year ago, but i'll dig it up if you're interested. it's hairy though
phf: err, it's worthwhile attempt because binary types core is small and easy to understand, and likewise easy to write own
phf: it's a worthwhile attempt anyway, because implementing binary types from scratch with necessary parts to support bitcoin is not that hard. mine is 129 lines
ben_vulpes: something like that
phf: so you're either stuck with that, or you figure out how to make it general purpose :>
ben_vulpes: but yeah, anything that has compact size parts needs custom readers and writers.
ben_vulpes: http://wotpaste.cascadianhacker.com/pastes/d6kbl/?raw=true << i have to run for a bit
phf: ben_vulpes: ok, so that second paste is i guess not "general purpose", you'll have to write a reader/writer for every structure that has compact size'd parts in it
ben_vulpes: (there's a fair amount of garbage in the thing i didn't feel like cluttering the discusison with)
ben_vulpes: well yeah we were only talking about compact size!
phf: the entire thing you posted is and only is compact size.
ben_vulpes: phf, yes, that's why i pass a 'byte-count' keyword argument to the read-binary method for script bytes
ben_vulpes: generally try to keep the snr-damaging wailing and gnashing to my self.
phf: ben_vulpes: i still think there's some misunderstanding. once you have a compact size reader, you don't automatically get "read N objects of compact size count"
ben_vulpes: i have thought about this for at least ten seconds.
asciilifeform: but idiot author wanted 'generality', so we now have this.
asciilifeform: tx can't be any bigger than block.
asciilifeform: and think for 10 seconds
ben_vulpes: did some late night back-of-the envelope on script length and miminum transaction size and i don't recall it breaching u32
ben_vulpes: i don't even think you can have max compact size anything in trb.
asciilifeform: it's a very sad pattern though. promisetronic. in that you have a format ~in your head~ and then write 'reader' and 'writer' and there is no machine-assured isomorphism between the one and the other, and definitely not with the format.
ben_vulpes: i only intended to write the compact size reader, not take it out of bitcoinland.
ben_vulpes: and the whole "if less than 253, that's the number, if equal then read the next octet, if blablabla" is that consistent across c-land?
ben_vulpes: perhaps i misunderstand you, but once i have the type, sizeof, read-binary and write-binary implemented, then i'll have a "general purpose 'binary type object of count `compact size`'"
phf: oh, then i don't grok your question. "yes, it's exactly like any other hand-rolled variable length integer in c land"
ben_vulpes: at least from where i'm sitting today.
ben_vulpes: asciilifeform: the maximally compact description is 3-4 'binary classes', and some hairy braindamage around the varint and scripts.
ben_vulpes: once i have a stake in the varint it will go precisely nowhere.