mircea_popescu: Mocky these schmucks i swear...
mircea_popescu: BingoBoingo argentine sub was diesel.
mircea_popescu: http://btcbase.org/log/2018-11-28#1875895 << all that's left of the us pretense to an economy is government wankolade.
a111: Logged on 2018-11-28 22:17 Mocky: so many of the jobs out there want secret government clearance now, jeez. who do they think they are kidding with their super secret shit
BingoBoingo: mircea_popescu: Sub was diesel, but sub and reactor are both outside Argentina's maintenance ability
mircea_popescu: along with a toothpick factory.
mircea_popescu: incidentally, the whole usg "clearance" wank narrowly mirrors the quatari "sponsorhip employment" wank.
mircea_popescu: http://btcbase.org/log/2018-11-29#1875920 << in the quaint words of the virgin mary, "yes."
a111: Logged on 2018-11-29 01:07 Mocky: "must be open sores contributor, love python 3" what's next? "line cook, must love toasters with cancel buttons"
spyked: http://btcbase.org/log/2018-11-27#1875118 <-- point well taken. my schedule is still a bit erratic at the moment, so what do you think about making the switch on monday (3rd dec)? I'll be tuned in starting cca 6pm utc.
a111: Logged on 2018-11-27 14:38 trinque: spyked: point is a simple "hey trinque, lets make the swap at $date, and get the PM subscriptions ported over meanwhile"
spyked: re pm feeds: if no one objects to this, just gpg me a list of the feeds and the recipients and I'll add them manually.
spyked: http://btcbase.org/log/2018-11-27#1875503 <-- /me ordered a c101pa from shitazon recently, will make prime target for cuntoo testing.
a111: Logged on 2018-11-27 19:51 mircea_popescu shall buy a lappy just to try trinque's thing on!
spyked: http://btcbase.org/log/2018-11-27#1875439 <-- this has been on my mind after seeing http://trilema.com/2018/this-gns-thing/#selection-317.47-317.193 . the next immediate step in my queue after publishing rss bot vpatches will be to look at automating communication with deedbot.
a111: Logged on 2018-11-27 19:12 mircea_popescu: i imagine that'll be spyked's next thing, bot-driven vtrees mirror service.
asciilifeform: http://btcbase.org/log/2018-11-29#1875935 << last i knew , there were still a coupla open problems re actually using that box ( in particular, i did not yet succeed, and presently haven't the time to continue, in baking a bootloader for it that loads non-googlistic kernels )
a111: Logged on 2018-11-29 13:35 spyked: http://btcbase.org/log/2018-11-27#1875503 <-- /me ordered a c101pa from shitazon recently, will make prime target for cuntoo testing.
asciilifeform: iirc phf did turn his c101pa into ~some~ form of usable
asciilifeform: ( and yes google's loader ~will~ load custom kernels, but only if you sign'em with their tool, so then gotta keep whole shitchain around )
a111: Logged on 2018-09-08 19:07 asciilifeform: http://btcbase.org/log/2018-09-08#1848650 << update : flashing in the uboot with the dram turd from vendor fw, did nothing detectable
asciilifeform: it's a solvable, of course, problem , theoretically 100% of the iron init coad is published . but somebody would have to put in the sweat to port it into e.g. stock uboot.
asciilifeform: then and only then , c101pa will turn into an item like rk, where you can operate strictly with self-built contents
asciilifeform: presently, cuntoo won't run there (even if trinque's process can be coaxed into building arm64 bins) cuz it has no way of blessing the kernel.
asciilifeform: spyked: if you're interested in arm64ifying cuntoo, i recommend to begin with rk.
asciilifeform: http://btcbase.org/log/2018-11-29#1875930 << hilariously, i recently tried to run a simple numeric proggy on a box infested with python3 , and found that it utterly breaks the numeric stack : once you do e.g. x = x / 3 you can no longer e.g. x >> 1 -- because the / converted it to 'float' ! holy mother of fuck.
a111: Logged on 2018-11-29 06:10 a111: Logged on 2018-11-29 01:07 Mocky: "must be open sores contributor, love python 3" what's next? "line cook, must love toasters with cancel buttons"
asciilifeform: presumably they've introduced some new demented syntax for integer divides ? i have nfi, and could not be arsed to find out.
asciilifeform: http://btcbase.org/log/2018-11-29#1875928 << it does indeed ; usg ministry of whateverthefuck in fact charges corp 20-30k $ for the privilege of having a new d00d pierced into secre-pederasty
a111: Logged on 2018-11-29 06:09 mircea_popescu: incidentally, the whole usg "clearance" wank narrowly mirrors the quatari "sponsorhip employment" wank.
asciilifeform: ( and iirc, the cost is in fact ~open-ended~, and is proportionate to how many hours the pinkertons actually feel like sinking into particular d00d )
asciilifeform: traditionally, folx who show symptoms of being able to survive outside of the reich, disqualified.
Mocky: looks like Roger Ver's bitcoin.com is looking for java and js developers for their mining team http://archive.is/MlomS
Mocky: "74 applicants last week" heh
a111: Logged on 2018-11-20 03:38 mircea_popescu: i have nfi, it's pretty fucking lulzy though. i mean, there's a long list of these defeateds by "fate", but he's one of the most hysterically humiliated humbugs.
asciilifeform: apparently at least 4 still-not-gone-for-glue horses in usg's stable ( gavin, buterin, wright, ver )
Mocky: so many damn recruiters. they're like mosquitos in a swamp. they all want to 'have a call' and then 'a skype' 'see if you'll be a good fit'. ugh. two years out of college with a degree in hr, you're gonna decide if i'm worthy to be submitted for a job, and then pressure me to take low ball offer.
Mocky: this is why i worked for the same company for 14 years, i hate this shit. but also what made me soft target for their betrayal
Mocky: i'm convinced this whole ecosystem of outside recruiters is 100% byproduct of startup "investing" that props up the illusion of "high growth" companies that burn thru cash by hiring people so they will qualify for more money
asciilifeform 100% in agreement with Mocky , saecular work suxx
Mocky: the only activity in life that makes me understand why people turn to the bottle
asciilifeform: Mocky: i've even worked in a salt mine where the bottle was Officially (i shit thee not) issued, erry friday at noon.
asciilifeform: 'happy time'
asciilifeform: srsly there was a bar, of sorts, built in, on the grounds
a111: Logged on 2018-11-28 22:46 asciilifeform: ( d00d had pretty good gig: opened safe in the morning, closed in evening, and played 'wow' all day... )
Mocky: reminds me of "20% time", "work on your own projects on friday since we know you'll be doing that anyway"
asciilifeform: afaik this went away with the 1st 'dotcom' hysteria
Mocky: haven't seen it with my own eyes, even then
asciilifeform: ( and only even existed on paper in west coast sv shitholes; where stipulation was 'IF you've shat out your quota before friday', which typically is approx as likely as in gulag in 1950 )
Mocky: I interviewed with a job once that offered "unlimited paid time off" but inside reports were that anyone who tried to use was instantly on the shit list
asciilifeform: it's unfortunate that mircea_popescu is asleep right nao, otherwise he could tell us about how we're idjits and how working for living is for lusers, and how troo hero can run on sunlight and do as he pleases
Mocky: in truth i do feel like idjit for being old man without the brains to have saved anything. i don't even have a workshop full of shit like asciilifeform
asciilifeform: Mocky: as mircea_popescu is fond of pointing out, workshop is very much 2edged sword
asciilifeform: arguably it is better to be '100% theoretical' and have 'workshop' that fits in briefcase, as e.g. phf does
BingoBoingo: https://www.elobservador.com.uy/nota/-como-estan-hoy-los-precios-para-alquilar-en-la-temporada-de-verano--2018101920160 << Argentina's pesos falls, but the attempted controlled descent taking Uruguay's peso down is "The dollar strengthening"
asciilifeform: Mocky: to a 1st approximation, 'saving' in the reich dun actually work, see e.g. http://btcbase.org/log/2018-08-23#1843888 , http://btcbase.org/log/2017-06-05#1666161 , http://btcbase.org/log/2016-11-29#1574483 , elsewhere.
a111: Logged on 2018-08-23 17:37 asciilifeform: and indeed usgulag is powered by not only printing press but confiscating savings from ~randomly-picked engineers, little-league 'ceo', etc . a la mr stack.
a111: Logged on 2017-06-05 19:09 asciilifeform: it is interesting how phrase 'life savings' already has an archaic 'patina coat' -- same flavour as 'workhouse', 'savings & loan', 'joint stock'
a111: Logged on 2016-11-29 02:36 asciilifeform: there was iirc something about how an economic landscape stable enough for 'savings' is an ephemeral and quite unnatural thing, rather like billiard table
asciilifeform: it dun even matter so much what the supposed 'savings' are denominated in -- eventually you're out of work and they get wiped.
asciilifeform: reich is set up to ensure that this happens with clockwork regularity to ~erry worker bee.
asciilifeform: ( another great way to lose shirt is 'start company , try to sell useful product ' , a+++ worxxx )
asciilifeform: some folx get to the bottle after 20yrs in workschwitz and cowardice, some after 1,2,3,4 attempts to break out, end result same
asciilifeform: and (at least in asciilifeform's observation) the competent engineer types get to bottle ~sooner~ than the chair warmers.
Mocky: i can't even complain, i ended up with exactly what I wanted as a 20yo, "write software, have a lot of kids"
Mocky: as long as I still had kids at home, never tried to get out
mircea_popescu: http://btcbase.org/log/2018-11-29#1875956 << in similar news, there's a gautier, mississippi.
a111: Logged on 2018-11-29 16:17 Mocky: looks like Roger Ver's bitcoin.com is looking for java and js developers for their mining team http://archive.is/MlomS
asciilifeform: oh hey guten tag mircea_popescu
mircea_popescu: http://btcbase.org/log/2018-11-29#1875954 << whole shitshow runs like that, "open ended", to make the most of "moneyed". i recall back when i was looking at shooting lawsky imbecile in the head, had brief discussion with "legitimate" new york "detective agency".
a111: Logged on 2018-11-29 15:07 asciilifeform: ( and iirc, the cost is in fact ~open-ended~, and is proportionate to how many hours the pinkertons actually feel like sinking into particular d00d )
mircea_popescu: was pretty lulzy, typical-to-TV-altreality new york "contractor", "well... we'll have to put 8 men on it in three shifts... that's 54876983769287563984798634... and then..."
mircea_popescu: i was like... lolok.
mircea_popescu: http://btcbase.org/log/2018-11-29#1875962 <<< was even dood here for a while, did ~jack shit but loudly for a year or so.
a111: Logged on 2018-11-29 16:42 Mocky: so many damn recruiters. they're like mosquitos in a swamp. they all want to 'have a call' and then 'a skype' 'see if you'll be a good fit'. ugh. two years out of college with a degree in hr, you're gonna decide if i'm worthy to be submitted for a job, and then pressure me to take low ball offer.
mircea_popescu: the problem with a country consisting of doing-other-people's-laundry expects is that... well...
mircea_popescu: http://btcbase.org/log/2018-11-29#1875963 << speaking softly ~withouy~ carrying a big stick is a very poor strategy.
a111: Logged on 2018-11-29 16:43 Mocky: this is why i worked for the same company for 14 years, i hate this shit. but also what made me soft target for their betrayal
mircea_popescu: http://btcbase.org/log/2018-11-29#1875964 << quite exactly. you ever read http://trilema.com/2015/you-know-what-gets-no-airplay-unflattering-truth/ ? there's a complex GRP-like system that folded the us economy, and yes the red hot core was mis"investing".
a111: Logged on 2018-11-29 16:48 Mocky: i'm convinced this whole ecosystem of outside recruiters is 100% byproduct of startup "investing" that props up the illusion of "high growth" companies that burn thru cash by hiring people so they will qualify for more money
mircea_popescu: led to the usual symptoms of imperial decay -- monetization of real estate (a bad idea throughout, even if it repatriated the money japs made selling toyotas/sinking detroit in the 70s -- sorta like arsenic is bad for you even if it "keeps syphilis inactive" and so on) ; the star model driving and being driven by the fashion cycle, and so the fuck on.
mircea_popescu: (grp, glucose-regulated protein, "chaperones" in microbiology)
mircea_popescu: http://btcbase.org/log/2018-11-29#1875967 <<< most of wall street works like this, actually.
a111: Logged on 2018-11-29 17:00 asciilifeform: Mocky: i've even worked in a salt mine where the bottle was Officially (i shit thee not) issued, erry friday at noon.
asciilifeform: mircea_popescu: by my reckoning, ~90% of the software worx in usa ( just about all of remainder being -- direct usgisms )
asciilifeform: fit this description, that is
mircea_popescu: i meant re the bottle. your "boss" will take you "partying" at approved venues etc.
asciilifeform: 'we are hiring developers' 'why' 'we got 100m series a'
asciilifeform: iirc in e.g. jp you can't even work (if yer a d00d) if ya dun drink
asciilifeform: and makes sense that 'partying at approved place', otherwise grunt could party at who-knows, maybe run mouth..
asciilifeform: ( 100% iron reich logic )
mircea_popescu: esp the sort of scum they hire.
asciilifeform: they hire 'the scum we have, not the scum we wish we had'(tm)
mircea_popescu: the work-on-wallstreet is rather visible a collegiate track, sorta like "nigger".
mircea_popescu: back in the days of pam grier's youth, her brother going "i'm black. i don't dance. i don't athlete. i don't sing. wtf should i do ?!" was kind-of a joke. now though...
asciilifeform for most part innocent of that racket, has nuffin to add re subj
asciilifeform: those folx do occasionally pay for software to be written, but the work (perhaps deservingly) is ranked as janitorial and paid accordingly (esp. given where the grunts are to live)
mircea_popescu: considering the shit they get...
mircea_popescu: iirc i reviewed a typical example.
asciilifeform: i have nfi what they get, for all i know 'wall street' is 100% microshit-excel powered even now.
mircea_popescu: let me dig it up.
asciilifeform: ( is there a prb plugin for excel yet ?1 )
mircea_popescu: http://trilema.com/2013/why-mpex-is-better-than-fiat-institutions-part-349085-we-dont-use-excel/#selection-77.0-81.426 << standard of wall street software.
mircea_popescu: meanwhile in totally unrelated news, https://dod.defense.gov/About/Biographies/Biography-View/Article/1281505/ellen-m-lord/
mircea_popescu: what the fuck do they do now, cross jews and hares ?!
mircea_popescu: http://btcbase.org/log/2018-11-29#1875978 << fuck sunlight, i spent hours in the tropically warm bay waters slutwrestling, i am so burned i'm peeling.
a111: Logged on 2018-11-29 17:08 asciilifeform: it's unfortunate that mircea_popescu is asleep right nao, otherwise he could tell us about how we're idjits and how working for living is for lusers, and how troo hero can run on sunlight and do as he pleases
mircea_popescu: http://btcbase.org/log/2018-11-29#1875979 << and you're quite right to feel that way.
a111: Logged on 2018-11-29 17:10 Mocky: in truth i do feel like idjit for being old man without the brains to have saved anything. i don't even have a workshop full of shit like asciilifeform
mircea_popescu: ahahaa argentine peso 37 to the dollar. fucktards.
mircea_popescu: and i bet you they're STILL going around pretending.
mircea_popescu: "oh, dollar '''strengthened''', this building's value DOUBLED TO MATCH brekekekekekeke"
BingoBoingo: It takes more than 37 argentine pesos to get a dollar on this side of the river
mircea_popescu: im sure.
asciilifeform: http://btcbase.org/log/2018-11-29#1876036 << sounds like asciilifeform would last 10min in mircea_popescustan. asciilifeform burns to crisp on 39th parallel , in ~20m
a111: Logged on 2018-11-29 18:15 mircea_popescu: http://btcbase.org/log/2018-11-29#1875978 << fuck sunlight, i spent hours in the tropically warm bay waters slutwrestling, i am so burned i'm peeling.
asciilifeform: ( on at least 1 occasion, even in winter )
asciilifeform: http://btcbase.org/log/2018-11-29#1876034 >> '... degree in chemistry from the University of ...' << tovarasul elena ceausescu !111
a111: Logged on 2018-11-29 18:12 mircea_popescu: meanwhile in totally unrelated news, https://dod.defense.gov/About/Biographies/Biography-View/Article/1281505/ellen-m-lord/
Mocky: I got my first "so burned i'm peeling" in 2018, 3 hours on motorcycle in tank top mid day, late june.
asciilifeform: in unrelated lulz, $ wc -l libffa/* >> 3930 , $ wc -l ffacalc/* >> 1184 ; and story not even finished yet. ( tho ch1's 'RSA occupies around 3000 lines, incl. comments' was not a lie, it's exactly what the minimal rsa of ch9 weighs... )
asciilifeform: perhaps fyootoor folx will find ways to cut sumthing.
asciilifeform: ( for comparison: e.g. http://btcbase.org/log/2017-07-08#1680705 ; or, current trb is ~22k loc, ~not~ incl. the dep balls )
a111: Logged on 2017-07-08 03:49 asciilifeform: i just counted gpg 1.4.10 : 156,436 loc -- and that ain't counting the autoconf liquishit, or the libs it pulls in
asciilifeform: ( ffa has 0 deps outside of gnat itself and system character i/o, currently posix's )
asciilifeform: 1 of the reasons i wrote the series with the 'and here we have egypt', 'and now we remove egypt', etc. is specifically to show 'well you can cut 1000 ln here if you're ok with 500x longer run times' , i can picture applications where one might want this
asciilifeform: ( and on the other end of the digestive tract -- inlining. i'ma quite certainly issue a v-branch that removes inlining, on microcontroller it typically wins 0 , while making 100x bulkier bin )
asciilifeform: ideally, at some point we actually get that 8192-bit ALU chip, and whole thing can be... 0 ln
asciilifeform: !#s best machine is no machine
a111: 9 results for "best machine is no machine", http://btcbase.org/log-search?q=best%20machine%20is%20no%20machine
asciilifeform: ^ see also.
a111: Logged on 2018-10-25 19:15 asciilifeform: at the risk of repeating ancient thread -- 'the best machine is no machine', it weighs nuffin, needs no maintenance. and the best proggy, is no proggy at all, if a problem can be solved without writing proggy, it ought to be. erry line of coad can be rightfully pictured as an act of intellectual littering. y'know, like throwing cig butt or bottle on the ground in the park.
diana_coman: asciilifeform, and the loc is not the whole story either; I'd much rather read *your* 1000 loc than Koch's 100 loc
asciilifeform: why ty diana_coman
diana_coman: although ofc it's more like Koch's 1mn loc ~always
diana_coman just ate ffa 3, will sign
asciilifeform: oh hey
asciilifeform: on other front, phf : on occasion of the most recent bolix thread , i went and looked again at the http://btcbase.org/log/2018-01-17#1771955 artifact ; it is interesting that they provided 128 iron types, incl. bignum, but not a 'bignum of fixed N words'. i guess in '80s ~nobody was thinking of crypto at all. ( and i was prolly unduly pessimistic to the orig finder of $item, it is prolly 95% of what's needed for cycle-accurate clone.
a111: Logged on 2018-01-17 20:04 fromloper: asciilifeform: someone uploaded "I-Machine Architecture Specification" to Bitsavers three days ago, I thought you might find it interesting: http://www.bitsavers.org/pdf/symbolics/I_Machine/I-Machine%20Architecture%20Specification.pdf
asciilifeform: the tricky bit is that remaining 5% -- ops introduced since that draft, loadable microcode, init logic, etc )
asciilifeform: err, 64
asciilifeform: diana_coman: nitpicks / spiffy karnaugh simplifications / etc. welcome as always.
diana_coman: asciilifeform, the only thing re ch3 that I keep circling because not entirely clear why so is the "overflow in means *or* on word"
asciilifeform: diana_coman: btw you may find it entertaining that i found several vers of the carry & borrow eqns in my orig notes -- including yours. but for some reason i originally rejected that variant because it needed 2 accesses to the result D . but i neglected to write down why, possibly was simple brain fluke.
asciilifeform: diana_coman: plox to link to line ?
diana_coman: i.e. if one shifts right by 2 bits but provides an overflow of 8 bits than 6 of them get simply or-ed, it's not like they get pushed in
asciilifeform: diana_coman: how would a shift by 2 provide overflow of 8 ?
diana_coman: it doesn't provide it, no
asciilifeform: you mean, what if one were to feed it one when cascading ?
diana_coman: yes, feed it one, whether cascading or not; the point is: what is the meaning of overflow in exactly for a shift op?
asciilifeform: it will produce garbage, yes. i considered to make OF_in a limited type, but it would slow down the place where the item is actually used, substantially ( ada does not offer a fast bit-count-on-word operation )
asciilifeform: observe that the operators of fz_shifts are not exported in ffa.ads ( ch11 unified api ) , they are strictly for internal use in the lib.
asciilifeform: this is one of the reasons why.
diana_coman: aha; imho this sort of thing would be great added to the comments there
asciilifeform: diana_coman: i'ma absolutely add it to comment and discuss in ch14 mail bag.
diana_coman: thank you!
asciilifeform: and this is prolly not the only instance of the item. ( i discussed it briefly in ch11 , in the section where preconditions -- ended up moving many preconditions to the exported wrappers, they slow things down quite substantially when present on inner-looped invocations, as they prevent (for obv reason) inliner )
asciilifeform: idea being , exported routines ( current set is shown in http://www.loper-os.org/pub/ffa/hypertext/ch13/ffa__ads.htm ) are to be 'safe on all electrically possible inputs' , with the exception of div0 (user is commanded to test for div0, as example in http://www.loper-os.org/pub/ffa/hypertext/ch13/ffa_calc__adb.htm#172_17 )
asciilifeform: but the same guarantee cannot be given for erry internal component, not without geological runtime.
asciilifeform: i'ma add commentary/warnings when diana_coman et al point out good additional places where; but no one should live with expectation that ~all~ possible ways to break ffa by monkeying with internals, will be listed
asciilifeform: ( the only guarantee i can offer in good conscience is that nuffin can be broken by operating the ~external~ controls -- but even there user is required to see whether his cpu has barrel shifter (see ch13 discussion) , constant-time mul ( see ch9 discussion ) )
diana_coman: asciilifeform, it's not about exhaustive list of ways-to-break, certainly not
asciilifeform: and in principle i'm not averse to adding detail to comments. ( this is exactly the reason why comments exist -- show what is not necessarily evident from the coad )
asciilifeform: incidentally even the currently given 'external api' is eventually to be 'internalized', in the sense that user input is expected to be in P-code (presently named as 'ffacalc') and output ditto
asciilifeform: i.e. one will call, roughly, P(pcodestring, ffawidth, ffaheight, maxsteps) and get output.
asciilifeform: (this is, for most part, already what cmdline does. aside from 'maxsteps' not having yet been introduced Officially, cuz no loops yet)
asciilifeform: diana_coman: i'ma also note, _O_I is used strictly in fz_mod : http://www.loper-os.org/pub/ffa/hypertext/ch13/fz_divis__adb.htm#83_14 ; prolly oughta be inlined ~there~ and abolished as a global (even internally) function.
asciilifeform: 'best rake is no rake'
asciilifeform: at 1 time, was used in the constant entry ( nibble inserter ) routine, then the latter was replaced with rewritten http://www.loper-os.org/pub/ffa/hypertext/ch13/fz_io__adb.htm#29_14 , nao sole remaining use is in the knuth divider.
asciilifeform: orig was writing from pov of 'most general possible set of subcomponents' , but this is not necessarily what is wanted in the end product, esp. from loc-cut pov.
diana_coman: that's fine, note that my feedback is given as I read them so without actual knowledge of what happens later/where exactly they are used
asciilifeform: my current aim is that in ch22 ( see also http://www.loper-os.org/?p=2735 ) any subcomponents not used, are to be cut; and any that have strictly 1 invocation, are to be made sub-functions and remain visible solely in the scope where used.
asciilifeform: diana_coman: this is a+++ good, it is exactly how it is meant to be read
asciilifeform: anyffing that dun make immediate sense in ~that~ light oughta be either changed or commented.
asciilifeform: on 2nd pass of log, noticed that i did not answr http://btcbase.org/log/2018-11-29#1876084 : answr is, it is for shifting ~into~ a FZ . ( if this wasn't clear from thread ) ftr.
a111: Logged on 2018-11-29 19:30 diana_coman: yes, feed it one, whether cascading or not; the point is: what is the meaning of overflow in exactly for a shift op?
diana_coman: aha, that was my understanding of it
asciilifeform: right. ( and as presently stands, i've abolished all but 1 of the algos where this actually takes place )
asciilifeform: unrelatedly, to whom does trb noad 18.104.22.168 belong ?? it's been wedged for year+
asciilifeform: !#s 22.214.171.124
asciilifeform: srsly, whoever's poor neglected box this is -- fucking put in 'aggression' or pull the plug, thing is eating mains current to no useful end
asciilifeform can't even count nao how many time had convo, 'hulp! my noad is hanging' 'didja aggression' 'no...' '...'
asciilifeform: ( observe -- sans agression, the moar reliable is your hosting, the ~more~ certain your noad is to get perma-wedged, as it'll never reboot and never satisfy shitoshi's 'catch up only on cold boot' idjit condition.. )
lobbes: asciilifeform: heh that's my trb box. funny thing is, I -do- have aggression on that one >> http://btcbase.org/log/2018-07-14#1834350
a111: Logged on 2018-07-14 00:16 lobbes: http://btcbase.org/log/2018-05-22#1816438 << in other news, you were not kidding! my trb (hdd + aggression) has moved a mere 20K blocks in one month
lobbes: however, I have not yet tried timer patch on that one yet
lobbes really needs to get hands into that thing and feel around
mircea_popescu: http://btcbase.org/log/2018-11-29#1875991 <<-->> http://btcbase.org/log/2018-11-29#1876120 remarkable similarity. i suppose at this point to repeat that "socialism is about hindering the worthy to prop up the unworthy" is too much of a truism.
a111: Logged on 2018-11-29 17:23 asciilifeform: and (at least in asciilifeform's observation) the competent engineer types get to bottle ~sooner~ than the chair warmers.
a111: Logged on 2018-11-29 22:50 asciilifeform: ( observe -- sans agression, the moar reliable is your hosting, the ~more~ certain your noad is to get perma-wedged, as it'll never reboot and never satisfy shitoshi's 'catch up only on cold boot' idjit condition.. )
mircea_popescu: http://btcbase.org/log/2018-11-29#1876053 << the dependency shitballs are very much the problem though. dragging in all sorts of crap, by the mn loc, including wxwidgets or w/e, not to mention crapossl and so on.
a111: Logged on 2018-11-29 19:10 asciilifeform: ( for comparison: e.g. http://btcbase.org/log/2017-07-08#1680705 ; or, current trb is ~22k loc, ~not~ incl. the dep balls )
a111: Logged on 2018-11-29 19:21 diana_coman: asciilifeform, and the loc is not the whole story either; I'd much rather read *your* 1000 loc than Koch's 100 loc
asciilifeform: mircea_popescu: wx liquishit , qt, jettisoned loong ago. but bdb, openssl , boost, remained
mircea_popescu: a right, qt.
asciilifeform: 100x fatter turd mass
asciilifeform: ( than wx were )
asciilifeform: !#s zx2c4
asciilifeform: zx2c4: long time no see. what brings ya back ?
zx2c4: two things
zx2c4: one benign, one bothersome
zx2c4: the first is curosity
zx2c4: the second is wireguard needs funding for 2019 and thought this nation might help carry the weight
zx2c4: also i'm wondering what the usual trilema party line is on rust vs ada
asciilifeform: zx2c4: didnt mircea_popescu send you a coin ?
zx2c4: he did yea
asciilifeform: !#s rust
a111: Logged on 2017-05-10 21:29 phf`: i'm rewriting everything that asciilifeform is releasing in Ada in Rust, because it's secure AND modern!
a111: Logged on 2017-03-30 14:13 asciilifeform: why would anybody even name a comp lang 'rust' ? how about a surgical antiseptic named 'putrescence' ?
asciilifeform: in all seriousness, i don't even presently know of a more leprous pile of shit, either on pedigree or technical pov.
zx2c4: you dont think the borrow checker eliminates large classes of problems in a performant and somewhat elegant way?
asciilifeform: the what?
zx2c4: rust's ownership and borrow semantics
zx2c4: hey ffa in ada
zx2c4: nice post
asciilifeform: zx2c4: weren't you here last yr for a 'the technical cannot be separated from the political' and 'if program+all accessories doesn't Fit In Head, it is garbage, not proof' thread ?
zx2c4: no i dont think so
zx2c4: presumably the premise is something along the lines of complexity making things impossible?
a111: Logged on 2018-05-04 17:23 asciilifeform: zx2c4: as a matter of fact, is IS a conclusion i can jump to trivially. because your supposed 'person' is actually a nameless cockroach beneath my feet. because he is not in the wot, and thereby not distinguishable from the 90000..+ faux 'humans' usg manufactures on daily basis to further its psyops.
asciilifeform: zx2c4: point being, i don't practice haskellism. and nit from being illiterate yokel who has nfi how. it so happens that i know how. but consider the whole approach to be braindamaged .
asciilifeform: and, in parallel, to be an organized campaign of sabotage against ~concept~ of actual correctness in programs.
zx2c4: _organized campaign of sabotage_
zx2c4: that's a much larger accusation
asciilifeform: wtf is the point of 'here's a proof but you need this here 100MB of gnarl to ~run~ it and of course you will trust output, or are you a terrorist'
asciilifeform: proof is proof when it fits in head, now just as in euclid's time.
zx2c4: have you seen HACL*?
zx2c4: Kind of an interesting project
zx2c4: They write proofs in F* showing equivalence between some functional description and some imperative description
asciilifeform: have seen
zx2c4: And then they're able to lower the F* down into C
zx2c4: Which is pretty readable
asciilifeform: zx2c4: try to think about what i actually said tho
zx2c4: The proof, however, is gone from the C
zx2c4: Well of course everyone prefers simpler proofs that fit in the head
asciilifeform: if you need a 'proof system' to prove $assertion, you have NOT proven it. not to me.
zx2c4: Simpler code that you can read in a sitting
zx2c4: And so forth
zx2c4: But it turns out now all things are so easily provable as such
zx2c4: Not all math is as simple as Elements
asciilifeform: then they aint proven, basta.
zx2c4: Hasn't writing always been a tool to expand our knowledge / understanding / assurance beyond a single mind?
asciilifeform: not fits in head ? not proven. ( i did not say ~whose~ head, but imho safe to suppose that whole f# microshitiana garbage pail fits in no head )
zx2c4: Write down thoughts from one day, use writings to your benefit the second day, and now you're two people in essence
asciilifeform: zx2c4: theorem that dun fit in any 1 skullcase, aint proven.
asciilifeform: this is a political declaration, not an astronomical observation, zx2c4
zx2c4: Oh, I mean, if you're willing to allow for anybody's head (as i presume you mean by mentioning ~whose~ head), then you just posit an incredible genius
asciilifeform: not enuff to 'posit'
asciilifeform: gotta actually exist ( and 'genius' btw is the fella who makes the 'fit in 1 head' fit in 9000. )
asciilifeform: !#s heaviside
asciilifeform: ^ see also.
zx2c4: I'm not even quite sure what you mean by "fitting"
zx2c4: There are some things I could only prove if I had a paper
asciilifeform: fortunately there is a convenient demo
asciilifeform: ffa. see esp the passage re parachute.
zx2c4: I more or less know the process involved but don't have all of the steps in my head. And as I wrote it down, I forget the details of previously written steps while working out current new steps. Yet I have faith in the process of writing it down systematically and having intermediate results from pages prior
asciilifeform: the statement is quite serious, i and some yet-unknown number of other people will literally bet arse on our grasp of ffa correctness.
asciilifeform: the attribute which permits this approach, vs haskellism and other idolatrous rituals to mechanical molloch, is called 'fits in head'.
zx2c4: > At this time we will walk through the mechanics of our Karatsuba multiplier, so as to cement in the reader’s head the correctness of the routine, and lay groundwork for the optimization which is introduced in Ch. 12B.
zx2c4: Certainly a good pedagogical practice
asciilifeform: let's take from other end of digestive tract. consider 1 AND gate. idealized, no metastability, no breakdown voltages, literally 'ands' two logical values. ( can have physically limited to % of c speed, if you like, but this is uninportant )
asciilifeform: with me so far ?
zx2c4: so this state of "being proven" for you -- it requires some kind of intuitive bullet shot from start to finish of all particulars of a certain logical progression? and any deference of that to outside tools (like paper, or intermediate results with forgotten details) ruin the intuition?
asciilifeform: lemmas ok, paper ok
asciilifeform: so long as in the end it clicks in your crankcase and is as obvious as that and gate.
zx2c4: alright and you'd freely admit that when you write things down and solve intermediate lemmas, its often the case that by the time you get to the end, you don't have in the forefront of your brain the details of all intermediate steps anymore
asciilifeform: grasp obvious in its totality, as in there is literally NO hesitation to strap the thing on and 'if mistake -- any mistake -- i hit cement at terminal velocity'
asciilifeform: zx2c4: i like paper. and even chalk.
asciilifeform: but they are devices for eventually rearranging contents of your head, permanently, rather than substitutes for head,
zx2c4: so perhaps you prefer chalk and slate to paper then
asciilifeform: i sometimes even use computer algebratrons, in exploratory work. but would never consider presenting output straight from one as 'here, correct'
zx2c4: a temporary buffer, rather than a rolling log
asciilifeform: ( on some occasion, also macsyma )
asciilifeform: but is immaterial, next will ask what kinda chalks.
zx2c4: haha, was just curious. i presume you're ancient
asciilifeform: depends how you consider
asciilifeform turns 35 next wk
zx2c4: oh. then derive and macsyma surely are jokes
asciilifeform: deadly serious.
asciilifeform: ( wait till he learns that asciilifeform has a fleet of e.g. dos boxes. )
zx2c4: just install Coq for heavens sake
asciilifeform: or that FUCKGOATS rng is built to speak rs232 at 115200 baud and always will be
asciilifeform: or that ~ nuffing in asciilifeform's machine room postdates '09
zx2c4: hey i like good ol rs232 too
asciilifeform: zx2c4: whole reason i bother with this thread is to explain why no, i would ~not~ like to suck the coq, if you will
zx2c4: I agree with you in spirit -- I think learning details of a proof enough that you have all of it in your head is a good pedagogical approach and gives you more mathematical agility as you progress forward. I just don't know about the ontological statement regarding the proof's validity
asciilifeform: haskellism is like dope for intellectually-bent folx. it is addictive but ultimately ruin.
zx2c4: (coq is from the late 80s. its still in use, but i guess you could use the 2009 version...)
asciilifeform: intellectual bankruptcy.
asciilifeform: zx2c4: lemme ask, on what authority am i to accept the correctness of a proof generated via e.g. coq ?
zx2c4: i can see that critique of haskellism. endless intellectual masturbation (hello trilema?) that doesn't actually drive at any essential truth. there's a particular benefit in demanding truths remain small rather than large
asciilifeform: i have not read src of coq, to do so would take 20yr ( and i dun mean 'skim', but to find total proof of correctness of whole shebang and load into MY head )
zx2c4: well, as it turns out, coq had some amazing fallacy in its core code a few years back resulting in the ability to prove any statement true...
asciilifeform: and i'm to take it seriously after that ? pray tell why ?
zx2c4: it's convenient?
asciilifeform: as it is i have less nausea for even honeopaths than for mecha-proof folks
zx2c4: more seriously, if you're mostly after small computer programs to help you out when exploring a field but eventually do the proof by hand, z3 and sage wind up being super practical as day to day work horses
asciilifeform: homeopaths at least only bamboozle the irredeemably stupid
asciilifeform: zx2c4: you may find it interesting to learn that i once worked in a dour 'salt mine' where shat out 'correctness proof' all day. in 'sage.'
zx2c4: (i learned algebra back in the day from a professor who wrote a haskell program to generate our textbook... presumably in your mind, my foundational education could not be more screwed up https://en.wikipedia.org/wiki/Dave_Bayer )
asciilifeform: 'darpa' paid.
asciilifeform: sucked beyond my ability to describe.
zx2c4: z3 is nice for things like
zx2c4: that z3 script gives a little more assurance we didn't screw up the radix conversion here
asciilifeform: it was a misery. and 'sage' made it ~more~ of misery, if yer gonna commit 'machine proof' atrocities, has no biz doing in a non-homoiconic lang (i.e. any non lisp)
asciilifeform: zx2c4: lemme ask you this -- those mecha-proofs of yours, how much do you actually believe in'em ?
asciilifeform: i.e. wouldja bet life of self, wives, pets, whole clan of yours, on there being no boojum in coq, z3, other claimed brain substitute of the day ?
zx2c4: no. they're mostly just tools
asciilifeform: i'd hope that i dun have to explain that when you write a cryptosystem, you are in fact asking other people to do this.
zx2c4: to do what
asciilifeform: to jump on ~your~ parachute, that you sewed.
zx2c4: oh. yea, as i said, they're mostly just tools
asciilifeform: tools for what ?
zx2c4: an additional check against my feeble human reasoning
asciilifeform: how wouldja describe for what
zx2c4: it also lets me try things super fast without thinking
zx2c4: and then think backwards in order to reason about its correctness later
asciilifeform: so like paper ?
zx2c4: sort of
zx2c4: more efficient paper
asciilifeform: then why do you advertise and publish the 'proofs' ?
asciilifeform: i dun publish my 'derive' barf
asciilifeform: or chalk board contents
zx2c4: i think in crypto most people refer to it as "verification"
zx2c4: rather than proof
asciilifeform: whereas 'here is my z3 proof' is typically published for cachet among haskellist co-religionists
zx2c4: i save my z3 script so that
zx2c4: later if im changing the code
zx2c4: in addition to reasoning about it
zx2c4: i can quickly re-check my work
asciilifeform: it dun verify jack shit. is machine barf. just like the raw output of your compiler, except less meaningful
zx2c4: make sure i didnt so something immediately dumb
asciilifeform: if you can follow the proof by hand
zx2c4: in the example above, its pretty easy to prove that by hand
asciilifeform: can you, for the one for that routine you linked ?
asciilifeform: how much does it weigh ?
zx2c4: it's not bad at all
zx2c4: its a radix conversion
zx2c4: highschool contest math
asciilifeform: lessee the proof ?
zx2c4: fair enough
asciilifeform: what i see in the link, is a buncha c code, with pointerisms
asciilifeform: does your proof demonstrate that none of them can overflow, regardless of what happens , say ?
zx2c4: right. shows that the carrys work out without overflows given certain bounds on the input
asciilifeform: ... or that the #define ULT(a, b) ((a ^ ((a ^ b) | ((a - b) ^ b))) >> (sizeof(a) * 8 - 1)) macro dun turn to barf in the preprocessor on acct of some esoteric beard shaving from dennis richie ?
asciilifeform: and i could continue..
zx2c4: "unsigned less than"
zx2c4: dennis probably wants those variables surrounded by ( )
asciilifeform: consider, zx2c4 , i can show that nuffin in my arithm routines can overflow a buffer. even if you were to turn ada's overflow checks off. without any complicated tooling.
zx2c4: with good comments? or some nice feature of ada? or easy arithmetic youre implementing? or what?
asciilifeform: i can show this because the inputs , for given size, literally do not affect the program branch flow.
asciilifeform: zx2c4: with bare hands.
zx2c4: carrys/overflows are handled explicitly or something?
asciilifeform: consider to read the series.
zx2c4: what you've described sounds good. im wondering if it's the result of comments, the result of just better organized code, or the result of some nice features of ada you're using
asciilifeform: no magical feature of ada
asciilifeform: tho it does helpfully give pascal-style pass by ref, to avoid pointerisms
zx2c4: explicit carries everywhere, cool
zx2c4: what's W_Carry?
asciilifeform: it is result of the specific mathematical approach taken. where code must remain SMALL and NO branching on cryptobits, other than death on div0, is permitted. no memory indexing on cryptobits, either.
zx2c4: or, rather, where can i see the implementation of W_Carry
asciilifeform: zx2c4: see ch1
zx2c4: Shift_Right((A and B) or ((A or B) and (not S)),
zx2c4: Bitness - 1)
zx2c4: aahh that trick
asciilifeform: you can browse whole ch13 at http://www.loper-os.org/pub/ffa/hypertext/ch13/ffa_calc__adb.htm , if lazy
asciilifeform: html concordance.
asciilifeform: iirc you recced it last time
zx2c4: O_o really?
zx2c4: that would be very surprising
asciilifeform: possibly was sumbodyelse
zx2c4: !s fxt
zx2c4: !#s fxt
asciilifeform: it's sitting in a 3ring here in my tortureroom for whatever reason.
asciilifeform: http://btcbase.org/log/2018-11-30#1876333 << dun seem to be in the logs, possibly was in a private letter from sumbody..
a111: Logged on 2018-11-30 03:31 asciilifeform: possibly was sumbodyelse
asciilifeform: ( it's a not-terrible numerics schoolb00k , ftr )
asciilifeform: let's bounce the rubble, also.. http://btcbase.org/log/2018-11-30#1876304 >> ftr i have iron right here where this will bomb, cuz byte not 8bit there.
a111: Logged on 2018-11-30 03:22 asciilifeform: ... or that the #define ULT(a, b) ((a ^ ((a ^ b) | ((a - b) ^ b))) >> (sizeof(a) * 8 - 1)) macro dun turn to barf in the preprocessor on acct of some esoteric beard shaving from dennis richie ?
asciilifeform: ( longtime tuned-in folx, will know what irons )
asciilifeform: i'ma bet was not acctd for in the 'proof'isms..
asciilifeform: btw , q for ffa readership, can anybody think of a way to make the digit slider routine non8bitbyte-clean ? ( beyond having ffa eat its pistol on boot if it finds itself on such machine, lol )
asciilifeform: http://www.loper-os.org/pub/ffa/hypertext/ch13/fz_io__adb.htm#29_14 << subj. presumes existence of 'nibbles'
asciilifeform: afaik all of ffz outside of fz_io , is arbitrary bytebitness and wordbyteness -clean..
asciilifeform bbl,meat,and will come back on console with actual kbd lol
mircea_popescu: http://btcbase.org/log/2018-11-30#1876164 << dude, get the fuck off the planet, and take your pompously vacuous self-important idiocy with you.
a111: Logged on 2018-11-30 02:36 zx2c4: nice post
mircea_popescu: what the everloving fuck.
mircea_popescu: next shit out of your dumb mouth fails to string match apologizing for being quite so fucking stupid, i'ma fix the negligence whereby you can still speak here.
zx2c4: On my phone right now, woke up from a dream to see this, not sure if im still dreaming. Care to explain what's provoked your ire?
mircea_popescu: !!rate zx2c4 -1 moron
zx2c4: My "nice post" remark wasnt sarcastic, if thats what youre responding to