Show Idle (>14 d.) Chans

← 2019-02-19 | 2019-02-21 →
feedbot: << Bimbo.Club -- Philosophical Transactions. For the months of July, August and September, 1716. - Part VII.
mircea_popescu: !#seen ben_vulpes
a111: 2018-12-10 <ben_vulpes> << i did not know that statically-compiled mysql was solved; is it?
mircea_popescu: trinque did you ever see the guy irl lately ? say this year or something ?
mircea_popescu: meanwhile in other obscura that apparently exists,
asciilifeform: guten morgen mircea_popescu
asciilifeform: << i suspect he was barfing on the 'interfaces' thing.
a111: Logged on 2019-02-20 04:54 mircea_popescu: "as a language, it requires you to specify in great detail what you do not know in order to obtain the experience necessary to learn it."
asciilifeform: << very good idea; sadly pertinent , however , is the abi breakage ( chronicled e.g. here : )
a111: Logged on 2019-02-20 01:00 trinque: relatedly, here's a worthy task for an idle hand : find the oldest kernel you can boot on hardware you're willing to use
asciilifeform: i.e. if digging up a vintage kernel, would also have to take up the gcc from the period, or backport the current one, as i understand
mircea_popescu: at the worst this'll get us a nibler notion of the gcc tree.
mircea_popescu: item is doomed to v-fication, seeing how
a111: Logged on 2019-02-17 23:24 mircea_popescu: actually ada not having a backend can (and probably will) hide all sorta surprises.
mircea_popescu: nimbler*
asciilifeform: vfication is a 'not whether, but when'(tm)(r)
mircea_popescu: something like that
asciilifeform: the most rampant inconsistency, btw, is not even in the userland abi, but in driver (module) end
asciilifeform: linus specifically opposed making a stable one
asciilifeform: ( he saw the notion as 'making life too easy' for nvidia & other blob pushers )
mircea_popescu: bad virus records!!!
mircea_popescu: the computing 90s are soon to be just a memory, as these artefacts of sheer wtf slowly disappear.
mircea_popescu: asciilifeform yes, but linus is also an idiot.
asciilifeform: see related thrd, , re similar in gcc
a111: Logged on 2015-01-10 06:25 asciilifeform: mats: gcc... a) i dont get it... << what's the hard concept here? rms wishes to avoid gcc turning into a sweet and defenceless piece of candy for the 'embrace & extend/extinguish' folks.
mircea_popescu: they should have built an army in the 90s, rather than go for weakass jwz-catering.
mircea_popescu: their window's closed, nobody cares what they think nao.
asciilifeform: afaik they no longer think anyffin
mircea_popescu: these two tend to go hand in hand yes.
asciilifeform: successfully turned vegetable
asciilifeform: ( well, fungus, in rms's case )
mircea_popescu: mushroom is veggie.
asciilifeform: it is interestingly imho reminiscent of plant toxin , d00d couldn't think of anyffin but 'let's make it poison' in re defense
mircea_popescu: yeah huh.
mircea_popescu: now we're left with all those bleaters with livers, that are also spurious. "but mp, why won't you go into the liver ?" "wut ?"
mircea_popescu: "it worked for free and open plant!!!" "yeah ?" "UNFAIR!"
asciilifeform sees linux kernel, gcc, ftr, as 'life support' item, rather like trb -- worth freezing and maintaining until proper replacement, but not really items with a serious future
mircea_popescu: it seems rather obvious that the 2020-2025 republican future will consist of a whole lotta etc as we unwind the "optimizing" part in and discover a whole lotta and
a111: Logged on 2019-02-17 15:22 mircea_popescu: holy shit, EVERY SINGLE THING. look here :
a111: Logged on 2019-02-17 14:49 mircea_popescu: see, engineers are worse than whores. a whore might pretend like she's not working, but an engineer does inept shit like "/* This should optimize out, but it is wise to make sure this assumption is correct. Should these have different sizes, we cannot cast between them and the overlaying onto ERRATIC will not work. */" so as to ~pretend~ like he doesn't see WHY exactly he wants to take that code out. seriously, ooga-booga-bu
a111: Logged on 2019-02-17 23:42 asciilifeform: ( for n00bz: ) writing a compiler back-end aint actually hard. asciilifeform & many many other folx, did it ~as homework~ , at school. the hard thing is writing a ~decent~ optimizing backend.
a111: Logged on 2019-02-17 16:05 asciilifeform: ( and when found that ~despite this~, , was pant-shittingly hilarious, how koch still managed to be the tortoise in the race )
a111: Logged on 2019-02-17 15:12 mircea_popescu: well, either that or they o(N) + n log n for some values of N, n.
asciilifeform: i fully expect ^ adds up to ~100% of the c opensores mass, yes
mircea_popescu: maybe 98. but in any case, the only discernible reason anyone uses it is that everyone is, individually, too lazy to write a proper one.
mircea_popescu: whether that can be remedied in a group or not remains, i suppose, to be seen.
asciilifeform: it's a standard 'spittoon in one strand' : for so long as you retain ~any of the c liquishit, yer stuck with gcc ( which is the only thing that builds it , the 'portability' of the lang is a fiction from day 1 )
mircea_popescu: certainly it will have to come before any kind of job, because very strictly speaking you can't write a working user enviroment on top of extant gcc.
a111: Logged on 2019-02-17 16:10 mircea_popescu: asciilifeform i very much don't expect you want the prb 12 that is emacs. rewrite yes ? :D
mircea_popescu: ie, is rather the result of "squeezing the most juice broken pliers permit"
a111: Logged on 2019-01-22 18:44 asciilifeform: nao tbf could even make same observation re emacs ( it's never the 1970s c core that bombs, always extension script that hangs )
asciilifeform: i also suspect that on modern irons, 'retargetable compiler' is plugging the wrong end of the funnel.
asciilifeform: really oughta have ~1~ compiler, and for a sane arch; and if yet cannot afford to actually siliconize the sane arch, then the sad iron oughta be booting straight in bios to a compact (asmistic) emulator of same
asciilifeform: this'd cost, yes, some % of horsepower, but in exchange gets a 100klok toolchain instead of millions of ??? loc
asciilifeform: all the bounds checks, exception handling liquishit, etc. that gnat shits out, oughta be happening in either silicon ( which we aint baking yet ) or failing that, in the 'sane arch emulator'
asciilifeform: rather than bloating erry single binary on the box with identical chunks of sanitycheck
asciilifeform: i hold that the answr to the quandary depends on whether the right cut of that knot is made. and imho ^ is that cut.
a111: Logged on 2019-02-20 15:43 mircea_popescu: whether that can be remedied in a group or not remains, i suppose, to be seen.
a111: Logged on 2018-10-29 16:37 asciilifeform: 'notation is worth +80 iq' or how did it go.
mircea_popescu: i can see it.
mircea_popescu: imo is the perfect approach to "you wish to use insane hw, you pay cost for it".
asciilifeform: absolutely that.
mircea_popescu: that "% of horse power" is entirely not % of horse power ; the measure of horse power is at the torque end, not at horse's feelings end.
asciilifeform: it was this very front that asciilifeform published the sage thing for.
mircea_popescu: i don't give a shit some insane nonsense "feels like" it's doing "a lot of work". the work done is measured at the useful end.
mircea_popescu: otherwise what, every useless 30something female with crotch rot under fluorescent lighting, also "is doing a lot of work" because she "feels" so ?
asciilifeform: the Right Thing will look roughly like a 256kB chunk of asm that the box boots straight into, and afterwards forgets that it's an x86, arm, etc.
asciilifeform: yer still stuck with certain idjicies of iron (e.g. dma) but then when finally bake new iron, you have a ready compiler & userland to plant on it, that actually behaves.
asciilifeform: i strongly suspect that any other approach to the problem (i.e. 'we'll do exactly what the gnutards did, but they were mice and we are men' ) is certain to degenerate into ~same rubbish as what the gnutards ended up with.
asciilifeform: nao tbf, this approach was not available to the gnu people, on acct of sheer poverty of 1980s, when erry byte & cycle counted, and 'must not waste' , yer entire memory space in 1985 might be 256kB. but today plenty of 'what to waste', why not 'waste' for actual gain ( like bitcoin 'wasted' mains current ) , rather than microshit/gnu-style.
asciilifeform: various folx attempted something roughly like this, but succumbed to own idiocy -- e.g. the java people, who 1) wrapped their 'vm' around the nonsensical shape of their tardlang 2) stopped short of 'and no this doesn't run on your tard os, is IS the os, fuckyou' ;
asciilifeform: or gabriel_laddel, who thought that linux kernel is an acceptable item to build the '256kB' out of, through power of magical thinking
asciilifeform: or the deathbed bolix people, who did ~99% of the job but sank on the titanic of the dec alpha, the only arch they managed to massage into compliance before ran out of dough
asciilifeform brb, teatime
mircea_popescu: << absolutely not. it strictly depends on whether we can continue maintaining the productive atmosphere, in which etc continue to be pissed in the face in the most baffling and painful manner where they feel that "they have horsepower" and the measurement comes back "get the fuck lost, moron".
a111: Logged on 2019-02-20 15:54 asciilifeform: i hold that the answr to the quandary depends on whether the right cut of that knot is made. and imho ^ is that cut.
a111: Logged on 2019-02-05 07:29 mircea_popescu: !!rate verisimilitude -1 children, seen and not heard.
a111: Logged on 2018-11-30 04:53 mircea_popescu: !!rate zx2c4 -1 moron
mircea_popescu: THAT is what it hinges on.
mircea_popescu: it shouldn't be difficult to notice that the "gcc plugging wrong end of funnel" is not simple coincidence, but the necessary result of the feelings and optionalities these idiots perceive themselves to be having without actually having them.
mircea_popescu: as long as the tim swansons of the world remain an oppressed minority, the world can be sane ; and not futher.
mircea_popescu: << why. cuz republic without republic, how the fuck is it supposed to stand.
a111: Logged on 2019-02-20 16:08 asciilifeform: various folx attempted something roughly like this, but succumbed to own idiocy -- e.g. the java people, who 1) wrapped their 'vm' around the nonsensical shape of their tardlang 2) stopped short of 'and no this doesn't run on your tard os, is IS the os, fuckyou' ;
mircea_popescu: they failed to identify the victims, and failed to victimize them.
mircea_popescu: actually when Mocky_ is done with qatar history, maybe he regales us with a summarized java history ~from inside~ on his blog.
mircea_popescu studiously and quite deliberately ignored the whole thing as a wikitarida/reddit of its time.
asciilifeform: mircea_popescu: i also ignored. but imho it's an interesting entomological case.
mircea_popescu: could be, yea
asciilifeform: << i dun even disagree that nuffin's worth doing if you don't kill the locusts first. but also imho gotta other things , the result of 'pesticide uber alles' agriculture is monsantostan.
a111: Logged on 2019-02-20 16:26 mircea_popescu: THAT is what it hinges on.
asciilifeform: also matters to actually plow the field, and even matters with what (iron beats wooden stick) and so forth.
asciilifeform: << could be quite interesting read. didjaknow, for instance, that the javatards got showered with enuff printolade that they baked chips ? and that none of these were in any sense commercial success ?
a111: Logged on 2019-02-20 16:37 mircea_popescu: actually when Mocky_ is done with qatar history, maybe he regales us with a summarized java history ~from inside~ on his blog.
mircea_popescu: i'm not proposing a reign of terror, merely intolerance.
asciilifeform: funnily enuff asciilifeform was just recently describing to pet , how mircea_popescu's pesticide worx, and finally clicked when i 'recall cell culture ? when yer growing eukaryotes, you add antibiotic, or guess what dish will be full of next day'
asciilifeform: the benighted folx who 'wai is mircea_popescu so eager to hit us with spiked club', i can tell they never cell cultured.
asciilifeform maintains a kind of purgatory for these , in #a, and it is quite cheap on acct of the fact that usually only takes'em a coupla hrs to realize that no, this aint the fungal farm, fungal farm is three doors down, and then go to where their place is.
BingoBoingo: <asciilifeform> it is interestingly imho reminiscent of plant toxin , d00d couldn't think of anyffin but 'let's make it poison' in re defense << Datura
asciilifeform: BingoBoingo: virtually errything that grows on satan's green earth
asciilifeform: fulla poison. animal kingdom dun sit still tho, which is why we think of e.g. belladonna as poisonous but not cabbage
asciilifeform: ( cabbage -- much moar toxin by mass, but we have the enzymes )
asciilifeform: i suspect also that erry sargeant in erry unit on planet, also could summarize compactly.
mircea_popescu: mataperros actually grows wild here.
mircea_popescu: apparently the dogs can learn to give it a berth. great perfume too
BingoBoingo: Se extraño a toxicodendron
BingoBoingo: Gotta get some here before considering ground level properties
feedbot: << Qntra -- Ohio Teen Sues Washington Post For 250 Million USD In Damages Over Harassment Campaign Spearheaded By Organization
mircea_popescu: watch them lose, get gutted, etc. anyone recall ALL THE OTHER pantsuit vehicles, meanwhile gutted in exact same manner ?
mircea_popescu: for that matter -- lulz of all time here, oscar arias, very much item, now #metoo'd.
mircea_popescu: apparently if you sign up to star in one production, you're stuck doing all the sequels also.
mircea_popescu: dood thought he can get just the nobel prize.
asciilifeform: mircea_popescu: thiel killing gawker, sure ; rando kid vs volkischer beobachter ? will be odd if it dun get 'appealed' for 25 yrs, until $kid dies of hunger
BingoBoingo: Recently (within the past two months) WaPo's slate magazine hired one of the more flamboyant Gawker alumni to do a "sex column" with some burnt out porn chick turned old woman
mircea_popescu: see, the old british ban on champerty & maintenance meanwhile dead. the fact that it happened to rando kid dun matter, trial can be bought.
BingoBoingo: asciilifeform: If anything the tort in this case is clearer
asciilifeform: mircea_popescu: wouldn't this require a thiel to show up to bankroll $kid's suit ?
mircea_popescu: the problem with common law is that it dun work.
mircea_popescu: asciilifeform plenty of them. it's profitable, see.
asciilifeform: i suppose if suit already brought, likely there's already a thiel.
asciilifeform: at some pt before upstack thrd gets lost in the sands of time -- would like to see mircea_popescu's ex cathedra pov re , when he has the time
a111: Logged on 2019-02-20 15:57 asciilifeform: the Right Thing will look roughly like a 256kB chunk of asm that the box boots straight into, and afterwards forgets that it's an x86, arm, etc.
asciilifeform: ( specifically -- is Right Thing ? and if yes, then how much sweat is worth to put into keeping ye olde gcc alive, and how much sweat instead for same ? and what oughta be in it ? )
mircea_popescu: asciilifeform precisely.
mircea_popescu: asciilifeform re bios emu -- i am certainly not against trying this. it's not possible to say much more than that as things stand now.
asciilifeform put ~decade into various attempts at $subj, but ~0 forward motion until made the sage breakthrough and gained ability to debug bios. but has been on back burner in preference to moar urgent tmsrisms ever since
mircea_popescu: understand tho, it has a very visible facet of wishful thinking. i mean yes, obviously, way the fuck better to have all the needful stuff in one place than added to each binary. this much is certain. nevertheless, the notion that you can stuff a converter from insanity to sanity "in the bios" requires just as much a magical stone as any other "universal sanity-insanity bridge".
asciilifeform: there is no magical stone.
asciilifeform: it's a quite serious set of open problems, many documented by asciilifeform in the log (e.g. how to drive nic?)
mircea_popescu: nevertheless, the attempt will certainly inform.
asciilifeform: i suspect that it's a ~smaller~ problem than 'bake 17 sane gcc-like back-ends'
mircea_popescu: like all attempts to date have, here or anywhere else.
mircea_popescu: well that for sure.
mircea_popescu: the obvious counter there will be "but not cheaper than baking ONE sane gcc". which is true, but nevertheless not as useful -- sometimes having an interface, even if "spurious" is better than not having it, which is why parents don't customarily discuss family finances with their 12yos.
a111: Logged on 2019-02-17 23:49 asciilifeform: re bolix back end, i suspect it aint very useful as starting point, because was far ~too easy~ item , in that the iron per se was sane (i.e. performed bounds and type checks, so much of what gcc is stuck doing in soft, was unnecessary )
asciilifeform: i suspect cheaper than 1 sane gcc likewise.
mircea_popescu: other than this, the "we don't want to fuck our brains with $bad-arch" is a dead end -- you will, whether to write gcc for it or to write bios for it.
asciilifeform: rright but you want to put the sweat into item that gets leveraged, rather than duplicated 9000 times, is the notion.
mircea_popescu: asciilifeform it's never cheaper to have a coupling than a straight link. nevertheless, couplings are used.
mircea_popescu: asciilifeform large part of people's complaints re leverage are insane. take the recent naggum piece discussed, he complains, literally, that "he wants to lie once rather than everywhere". duh, no sed where this man lives ?
asciilifeform: consider from pov of 'what is total mass of binary that has to be audited'. in which case is the mass smaller : where there are 9000 bins on the machine, each consisting 90% of a coupla MB of 'bounds check this array' and 'propagate this exception' ? or if the 1 runtime does these, and the bins -- invoke.
mircea_popescu: from the pov of auditing it is certainly cheaper to move it into the "emulator"
mircea_popescu: however, consider situation 1, where "it is in emulator" and situation 2, where "it is in this one tmsr standards library".
mircea_popescu: the 2nd gets linked in however many binaries you wish, still is a single item.
asciilifeform: i suspect not only cheaper in manpower, but faster in wall clock. cuz can hand-massage emulator, whereas compiler barf , even with peepholing etc, is invariably bloated.
asciilifeform: mircea_popescu: in practice, given static link, it's duplicated 9000 times, rather than single.
mircea_popescu: even so. as far as reading goes, you still read it the same number of times, whether it's linked or bios'd.
mircea_popescu: then again this whiole discussion is moot, because step 1 towards that magical bios asm blob is a tmsr standards lib to replace glibc anyway.
asciilifeform: the appeal of 'emulate sane iron' is partly 'now can has readable disasms for errything', and partly 'can then bake actual iron around same'
mircea_popescu: certainly.
asciilifeform: mircea_popescu: it isn't entirely same problem
mircea_popescu: not entirely, no.
mircea_popescu: but the overlap is large.
asciilifeform: ( glibc & co. do not handle devices, impose memory organization scheme , etc )
asciilifeform: i'm not currently certain that there is any meaningful overlap.
mircea_popescu: certainly. but once you start talking about unified abi, you're very much starting there.
mircea_popescu: ie, the fact that glibc doesn't come with sane memory allocator ~is a failure of glibc~.
asciilifeform: it's a forced failure, in the sense that it is imposed by trying to be a layer on top of os, rather than os per se
mircea_popescu: or in other words, the overlap is large even if the barbarians cut it wrongly and on the basis of their cuts it seems to not be\
mircea_popescu: but srsly now, the "standards lib" lets me print to screen but not print to ram ? wut.
asciilifeform: moar concretely, e.g. memory allocator -- attempts to put sane one on top of linux's, all caught fire in very similar way, i.e. impedance mismatch ( whether libc's or java's or commonlisp's )
asciilifeform: asciilifeform's hypothesis, is that it is only worth to add a layer if you can remove dozen by doing so
asciilifeform: rather than '1 moar layer'
mircea_popescu: surely. i wasn't proposing adding layers.
mircea_popescu: terminology fails, mostly because terminology was made by morons and we're trying to discuss analysis in roman numerals here, but consider "glibc" would be a... well i guess a kernel mod the program links against as a library ?
asciilifeform: in asciilifefor's cosmography, sane machine has neither 'linker' nor 'compiler' in the customary sense
asciilifeform: ( has : translator, from various preferred lang syntaxes to the actual machine coad, but without massive expansion of the ast )
mircea_popescu: to follow the logic : the notion that you'd have a kernel mod to interface with peripherals, but not with code, is somewhat bizarre.
mircea_popescu: asciilifeform the large problem here is that the ast will have to be ultimately a homomorphism of machine language. which is the dubious part in the "emulator" pov.
asciilifeform: mircea_popescu: in pctardation land, it's a consequence of the (massively failed, historically) attempt to hack around the fact that erry peripheral has free run of entire bus
asciilifeform: the central imho open q is, how much of the iron's retardation can be kept out of such a system.
asciilifeform: ( can has boat hull that dun leak ? )
mircea_popescu: asciilifeform i'm aware, evolved not designed. nevertheless, the fundamental breakage here is that glibc is proposed as a "library" rather than a "kernel mod". not that these terms make any fucking sense anyway, but what the fuck am i gonna do.
asciilifeform: the unix 'tower of layers', nominally separable, is spurious ( but mircea_popescu knew this ).
mircea_popescu: anyway, i think i fully understand what you mean re "no compiler, no linker" : it is evidently a broken situation when you have TWO patsh from "what master said" to "what machine does".
asciilifeform: so happens that i have a page re subj : .
mircea_popescu: which is the point, it's a waste of time to consider "how linus separated" or "how rms thought should be separated". whoile thing's built on magic musherooms, "sky quadrants" etc.
mircea_popescu: right, "compilation, you mean caching ?".
mircea_popescu: fucking bs "science" where ALL the notiuons are broken.
asciilifeform: aha, exactly so
mircea_popescu: very early 1500s medicino-chemistry-philosophical-barbershop flavour to the whole field.
asciilifeform has entire site about this..
mircea_popescu: but back to it, let's try and use different terms : i deem your "whole thing is an emu, in machine lang, bios boots into it" to be a "approach from machine end".
mircea_popescu: nothing wrong with this. and will have to be at least attempted.
mircea_popescu: nevertheless, the obvious alternative, factually there as holes follow fills and so on, is "approach from code end". and this approach'd be something like a ~proper~ standards lib. ie, both with proper access and proper primitives.
mircea_popescu: and i'm saying that a) both approaches share a lot of overlap (you say no, but you agree next line that conceptually, field is broken -- that is overlap!!! you just agreed!)
mircea_popescu: and that b) it is likely to be easier to start picking low hanging fruit from right rather than left, simply because cheaper fruit lowerly hanging there.
asciilifeform: 'proper primitives' is not independent of a) what backing iron b) what operator expects to ride on top (langs, memory management style) tho.
mircea_popescu: it's independent of a, because if it turns out all iron fails math we'll dump all iron, not re-write math.
mircea_popescu: though if memory serves intel mkt dept had internal proposal to "usg pronounces pentium right, everyone changes bookeeping to suit it"
asciilifeform: currently could almost say 'iron fails math'
mircea_popescu: there is that.
asciilifeform: in that e.g. it imposes ridiculously small bus widths, gimping rsa;
asciilifeform: and fails to provide locking in iron primitive, which gives the fungal bloom of software threading deadlock etc
mircea_popescu: moreover -- b is actually a major spot of research, because we don't even exactly know WHAT we will set down as "must be / must not be".
asciilifeform: mircea_popescu: asciilifeform's observation is that all historic attempts at 'pure (b)' were gimped by low expectations from iron, on acct of author's lifelong crippling experience with shit irons
mircea_popescu: this much is certain on strength of extant record.
asciilifeform: for instance -- the expectation that ram contents vanish when plug is pulled ( entirely avoidable since, what, late '90s or so ) ; or that machine is permitted to crash ( subj discussed in detail in log, crash oughta be seen as same calamity as catching fire , and this pov is entirely practical )
asciilifeform: or the 'bus width is handful of bytes' nonsense
asciilifeform: mircea_popescu: seaforth dead ??
asciilifeform: loox like moved to shithub
asciilifeform: luckily asciilifeform has last-known copy. will mirror if anyone wants.
mircea_popescu: update your link too, and there we go
asciilifeform: will surely update.
asciilifeform: ( btw if anyone else sees a dead link, plox to write in , i do fix )
mircea_popescu: anyway, to land this far going baloon : currently, the large chunk of sanification work will likely still be ; there's no way to have a world without a backend, there's no way to learn how to build one without unpacking the only one that exists, and so on.
a111: Logged on 2019-02-20 15:41 mircea_popescu: it seems rather obvious that the 2020-2025 republican future will consist of a whole lotta etc as we unwind the "optimizing" part in and discover a whole lotta and
mircea_popescu: ada-musl will have to get its own backend, even if it's mpi-style confiscation.
mircea_popescu: then we see, esp on the basis of discoveries during.
asciilifeform: it's at the mpi confiscation stage atm.
mircea_popescu: just about.
asciilifeform: q is, do you actually ~want~ a thing that shits out 200kB of peepholed x86 ??? for 'hello world' .
mircea_popescu: this isn't the q. there's exactly one thing that shits bytecode. much like in 2011 there was exactly 1 thing that extended the blockchain.
mircea_popescu: it's what you start with.
asciilifeform: granted. sorta why i said 'it's a trb'
asciilifeform: nursing old gnat aint optional, currently, just like nursing trb aint optional
mircea_popescu: moreover, the recent exception handling discussions & digs exposed most eminently the hole the shit's in. you could improve on gcc right now, just by fixing its idiotic tokenizers, "optimized" log n-onsense ans so forth.
asciilifeform: aand , i suspect, by cutting coupla MB of entirely dead mass
asciilifeform: a la mpi etc
mircea_popescu: shit's exactly in the position of mod6's famous house,
a111: Logged on 2019-01-16 02:24 mod6: This all started because we need a new door, it's old as shit, and all the weather stripping is bad, etc. So of course, this isn't std door size. So I paid some good money to have a custom one made to size. When the carp came out to install it, the first thing he did was pull off one piece of molding, and stuck his file down in the bottom area where that joist is located, and it pushed right through.
asciilifeform: btw do you recall fabrice bellard's compiler ?
asciilifeform: !#s bellard
a111: Logged on 2017-03-23 16:39 asciilifeform: trinque: fabrice bellard had an interesting experiment, where he stuffed tinycc into kernel, and had it self-build on boot
mircea_popescu: ah the tcc thing ?
asciilifeform: he had a kernel-building compiler in coupla 1000 loc .
mircea_popescu: yeah, ok, this very much germane to discussion.
asciilifeform: afaik then he sank into the sea, never continued.
mircea_popescu: sounds like he did a lot of the same stuff here contemplated. shoot fellow an arrow maybe ? is he old ?
asciilifeform: iirc old
asciilifeform: but worth a shot ( tho asciilifeform's track record for 'write 'im a mail' is currently abysmal )
mircea_popescu: << also. did the guy disappear ?
mircea_popescu: (note: "" moved to ""). and other keks.
asciilifeform: seems to have disappeared. and tcc was 'picked up' by fungi.
asciilifeform: nao it resembles minix, 100x the mass of the original, and ~less~ function
asciilifeform: bellard himself abandoned it, not the least reason , near as i could tell, was that he did not relish idea of maintaining 9000 backends
asciilifeform: ( or, alternatively, marrying intel 4evah )
asciilifeform: it's a sisyphian proposition.
mircea_popescu: yeah, but it's in ruby, or w/e the fuck unity
mircea_popescu: i mean, "<asciilifeform> nao it resembles minix, 100x the mass of the original, and ~less~ function" <mircea_popescu> yeah, but it's in ruby, or w/e the fuck unity.
asciilifeform: tcc ? was in c
asciilifeform: ( tho i last looked 5 or so yr ago. thing was self-building , and could build 2.x kernels in coupla sec )
mircea_popescu: jesus my poor point.
mircea_popescu: i was trying a tongue in cheek, "yes, now it has more code and less functionality, but it follows whatever trend of insanity".
mircea_popescu: then it was too late, and then it just became incomprehensible.
asciilifeform: i confess, did not follow the thing's trajectory into the ground, do not know which fashionable insanity in particular it followed as it died of immunocompromise
asciilifeform: but i think i get mircea_popescu's pt
asciilifeform: currently i suspect that the corpse of tcc is 1 of the moar fuckable corpses in the unixtardation graveyard, tho.
asciilifeform: iirc orig ver had a working ( x86, no x64 ) backend, and weighted 200kB or so in src .
asciilifeform: *weighed
mircea_popescu: << this, incidentally, is a fine example. who THE FUCK came up with the notion of making "registers" ~without~ making a) as many of them as the bus width and b) a special register which keeps a mask of registers, so that PUSH only actually occurs if needed ?
mircea_popescu: let me guess, the same fucking people who made mul without carry.
mircea_popescu: as a general design principle! you may NOT have "multiple" on a machine. it's 0, 1, buswidth-count. THAT IS IT.
asciilifeform: and , mircea_popescu , this is just the needle point on the tip of retardation iceberg.
mircea_popescu: and whenever you go with option 3, guess what. 1 IS MASK!
asciilifeform: the tard arch people passed the 'register killing' ball to compiler, as result bloating ~all~ compiler with massive gnarl, ~just for reg killing logic~
asciilifeform: this is quite typical example.
asciilifeform: writing sane compiler targeting e.g. x86 is not entirely unlike dancing ballet while carrying 50kg weights
asciilifeform: !#s bergeron
asciilifeform: ^ see also..
mircea_popescu: i suppose the pov where design is ~constrained~ activity, and the concept of "design competition" must strictly describe a situation where machine a with 64 registers of which one is mask and machine b with 64 registers + ONE separate mask register compete.
mircea_popescu: ie, whenever you encounter "competitions" of malformed nonsense, you're exactly in the position of the ~spectator~ to special olympics.
mircea_popescu: there MAY NOT exist "conflict" on baseless distinctions, which goes right the fuck into the whole
a111: Logged on 2017-04-09 23:14 mircea_popescu: "who do you think will win, mr 13, the team with two guys and monkeys or the team with three guys and monkeys ?"
a111: Logged on 2016-06-14 14:12 asciilifeform: did we ever do a screw threads thread ?
asciilifeform: the worst thing, imho, is that the 'baseless distinction' is quite often 'in one strand' with actual distinction -- e.g. the 'speshul olympics' of x86 vs arm : both are shit archs, but 1 has 35 yrs of compat-crud, and eats 100x the wattage vs other
a111: Logged on 2016-01-19 21:55 ascii_butugychag: the whole point of engineering is to cut apart the happenstance-linked crud from what you actually want.
asciilifeform: you can't, for instance, buy a x64 that ~only runs in 64 mode~ : no, you gotta have the (broken) msdos compat, inner 386, etc. cuz winblowz.
asciilifeform: likewise you can't buy a southbridge that dun do the acpi liquishit, cuz again microshit decree.
asciilifeform: ( current pc arch 'standard', to the extent it exists even, is a microshit authorship )
mircea_popescu: kinda makes it a particularly delicious spot to rape the femstate.
asciilifeform: iron ? the most delish, imho, spot conceivable. but -- expensive.
mircea_popescu: conquering china, also expensive. who knows, maybe it can be massaged to pay for itself.
asciilifeform: imho if anyone can conceive of just how, it's mircea_popescu
asciilifeform: we dun lack the 'how' ; if mircea_popescu commissions an iron, i'ma draw up an iron. q is how to massage to pay itself.
asciilifeform: (e.g. FG, for instance, was drawn up from the start to be siliconizable w/out any major change, if/when time comes)
asciilifeform: could fit a million or 2 of'em on 1 die, with modest density.
asciilifeform: ( would want a windowed / microscopable die, naturally, if actually did this, or how wouldja audit )
asciilifeform: i find it hard to picture how a sane cpu, where optimizing compiler is 10k loc and 'fits in head', could have no market. but then again the bolix people proved that it ~is~ possible to go broke with one.
hanbot: in bootstrapping adventures, it looks like the flow for a machine that knows to v to get vtools going is like so: grab some ancient v, ie mod6's, use it to press phf's vtools vpatch, then eat phf's "updated for vtools", then press vtools to keccak head. anyone feel like spotting this for me?
hanbot: where "knows to v" = "knows no v". heh.
asciilifeform: hanbot: it is also possible to bootstrap any vtron using naked gpg, ancient gnupatch, and bare teeth ( manually check sig and patch -p0 < foo.patch, for ea. )
feedbot: << bvt's backtrace -- Cuntoo Installation Report
mircea_popescu: hanbot write down the whole story as you go, but basically yes.
hanbot: aite. gracias.
asciilifeform: hanbot: other nitpick -- is asciilifeform's orig. demo; is mod6's vtron
asciilifeform: hanbot: actually exists in modern ( keccakistic ) incarnation -- there's 2 variants, largely identical , diana_coman's (the 1 i use now) and phf's (he had some incompatibility in his python iirc which required own variant)
asciilifeform: both of'em use phf's keccakistic vdiff.
asciilifeform: !#seen esthlos
a111: 2018-10-23 <esthlos> << apologies alf, I'm running behind! trying to gather time to get caught up in the next week or two
asciilifeform: ^ also had a vtron. but then sank to the bottom of the sea, so unmaintained afaik.
hanbot: right right, i mixed 'em up. and yeah, i'm planning on grabbing phf's keccak in step 3, if only because i've seen diana_coman's pop up in cuntoo tests so i'd like to test the ver less traveled.
BingoBoingo: !#born esthlos
a111: 2017-12-16 <esthlos> hello >>
asciilifeform: didn't even live a whole year, huh.
asciilifeform: it's a shame, imho; i quite liked
mircea_popescu: esthlos had a pretty cool vtron if i recall.
BingoBoingo: From the mines: "Viví 6 meses en Madrid y varias veces se me rieron en la cara (la mayoría empleados públicos, y una vez en un banco) por mi manera de hablar, pero mi peor experiencia fue en entrevistas de trabajo donde me dijeron que el problema era que no hablaba español." - Some Uruguayo
BingoBoingo: A bit deeper "Uruguayos no somos Latinos, somos Europeos": "Mi amiga tuvo un hijo en EEUU hace un tiempito y cuando lo van a anotar, dentro de las cosas que ponen en la partida es raza (WTF EEUU??!), la administrativa no quería ponerle "caucásico" así que le puso Latino white o una onda así."
BingoBoingo: And in still other lols: "Policía fue encontrado muerto, atado de pies y manos; la Fiscalía investiga: Murió ahogado en una cantera de Las Piedras. Sin embargo la principal hipótesis es la de un suicidio." >>
BingoBoingo: Dead cop further down page than dead graffiti communist punk
feedbot: << Loper OS -- PCB Radiography Kindergarten, Continued.
asciilifeform: wb Mocky
Mocky: thx asciilifeform
Mocky: I know of a couple attempts at java os, one I installed on my Compaq iPaq in 2002. but they were shit
a111: Logged on 2019-02-20 16:08 asciilifeform: various folx attempted something roughly like this, but succumbed to own idiocy -- e.g. the java people, who 1) wrapped their 'vm' around the nonsensical shape of their tardlang 2) stopped short of 'and no this doesn't run on your tard os, is IS the os, fuckyou' ;
Mocky: I was hopeful back in the day at java chip and java os, nothing worthwhile surfaced
asciilifeform: a good thing, too
asciilifeform: ( or we could've ended up with a 'crapple future' 20y ago, instead of nao )
Mocky: i mean, entire jvm is c/c++ and filled with c/c++isms so not exactly in the direction of , more in the direction of 'just want to' cross platform
a111: Logged on 2019-02-20 15:50 asciilifeform: really oughta have ~1~ compiler, and for a sane arch; and if yet cannot afford to actually siliconize the sane arch, then the sad iron oughta be booting straight in bios to a compact (asmistic) emulator of same
asciilifeform: Mocky: imho all of javaism was a massive exercise in 'just want to'ism
Mocky: I mostly agree. They did manage a coherent threading model / memory model which turns out to be the thing I miss when I'm writing in something else
asciilifeform: Mocky: coherent how ? iirc thing still relied on explicit semaphorism
mircea_popescu: yeah, wut ?
Mocky: there are explicit and implicit semantics, but if you read the spec, you know what you'll get. multiple threads setting a value on the same variable never create garbage
asciilifeform: eh setting. how'bout getting.
asciilifeform: easily create deadlock.
mircea_popescu: yeah, how about multithreaded i++ ?
Mocky: setting and getting
asciilifeform: dunno, the multithread java they had us homework at uni, deadlocked as easily as farting.
mircea_popescu: Mocky so if i declare a procedure which increments a global and a local counter, and then call it in a loop from multiple threads, at the end the globa lflag will contain the sum of the local flags ?
asciilifeform: and i wrote 9000 proggies that produced diff answer erry time they ran depending on what the thread gods felt like that day
Mocky: can do multithreaded i++ and never get garbage, if you mark it volatile forces read and write to main memory
mircea_popescu: but the ensemble will not move at the speed of a single core will it ?
mircea_popescu: asciilifeform rng!!!
Mocky: i'm not saying they solved threading once and for all, or that eliminated deadlock. any system that allows you to acquire locks in different order can be deadlocked. i'm just saying that there are concurrent primitives that can be understood and which have guaranttes that hold
asciilifeform: mircea_popescu: hey it's approx how microshit's 'rng' worx
mircea_popescu: Mocky just trying to evaluate if this thing worth digging up
Mocky: and yes can increment global and local counters in threads and have it add up, but requires locking or cas
asciilifeform: ahahaha with locking lolk
Mocky: laugh if you will, yet still can't do it even *with* locking in standard c++, as far as I can tell
asciilifeform: i dunthink ~standard~ cpp exposes threads ( maybe cpp14 ?? )
Mocky: so not optimal, maybe even laughable, but yet has a model and as spec that is not self contradictory
Mocky: it is on c machine after all
asciilifeform: eh it's the lang that made e.g. integers a 'non-class' object and thereby non-storable in lists etc
asciilifeform: arguably managed to ~add~ retardation on top of ordinary cmachine, no mean feat
Mocky: yeah, early java lists were shit in that way but no more. now shit in diff ways
asciilifeform: ... and where they added iterators 15 yrs after lang released, to great fanfare
asciilifeform: could go on, but wai.
asciilifeform: ( or where promised 'portable' but this this day i've never seen a gui proggy in java that runs on >1 os w/out modification. and etc )
asciilifeform: *to this day
Mocky: I'm not defending java, I'm stating that in addition to pile of shit, theres a coherent memory model, thread model that is not agony to work with.
Mocky: i've seen dozens of those. look equally shit on every platform
asciilifeform: i found the thread model just as ridiculous as that of pthreads.
asciilifeform: explicit threading is ipso facto sad.
asciilifeform: srsly the ~fortran~ people understood this. in 1990.
asciilifeform: ( ever wonder wai fortran still in use at '9000 cpus' massively parallel comps ? this is why )
Mocky: i wouldn't argue that explicit threading is good. but instead of concurrent i++, consider concurrent assignment to heap references, even without locking always have a valid reference, never a garbage pointer, no matter how many threads or cores
asciilifeform: sounds like reads locked by default then
Mocky: anyway coherent != good
Mocky: also threads in infinite loops can always be stopped on every platform
asciilifeform: cuz they dun map to native threads
asciilifeform: thing's got own scheduler
Mocky: zackly
asciilifeform: ... which means they're liable to all end up on 1 physical cpu, lol
Mocky: nope
asciilifeform: on linux, defo did, when i last gave a shit
Mocky: serious question asciilifeform, where do I look to see better threading that java. I honestly dont' know
asciilifeform: e.g. fortran-90
asciilifeform: or if you gotta explicit thread, ada's tasks in sjlj mode ( or even zcx , if you dun care re terminations )
asciilifeform: or even erlang's thread machine
Mocky: iirc that was after backus turned to shit in committee
asciilifeform: ( 100% lock-free )
Mocky: fortran threading I never looked at, but will. pretty sure threw away all my punch cards tho
Mocky: erlang I did look at briefly, but not the concurrency, seemed to me like a puzzle lang
asciilifeform: ( re f90 )
asciilifeform: Mocky: not a 'puzzle lang', tho i can see how you might've come to the conclusion ( it's been suffering from heavy case of )
a111: Logged on 2015-06-22 21:22 mircea_popescu: ascii_field this goes right into our discussion about how the only reason lisp is ok where c is shot is that the hordes haven't sat on lisp but on c, would have been the other way around if they had sat the other way.
a111: Logged on 2017-03-30 14:38 asciilifeform: mircea_popescu: i will also nitpick : 'erlang' does not belong in the list, it was a 1980s product that worked quite well in its industrial niche (large telco switches) but was later stolen and used as a totem by the folx from yesterday's thread ( )
a111: Logged on 2017-03-30 14:43 asciilifeform: anyway erlang is imho only worth discussing as part of a larger pattern -- industry after industry independently discovered that c -- and its entire approach to logic -- is poison
a111: Logged on 2017-03-30 14:50 asciilifeform: trinque: erlang wasn't simply about 'uptime', or even 'no pointer arithmetic', it also was the only case i know of where process migration actually worked
asciilifeform: the threads (they were 'lightweight', i.e. zcx-like, things, but with termination (and restart!) mechanism) were snapshottable objects, to the extent that the runtime knew how to move'em between not only cpus but machines, and in such a way that the proggy continued w/out interruption (if e.g. machine catches fire and its threads gotta move before fire reaches cpu etc)
asciilifeform: i mention this strictly cuz we're apparently doing a 'dead langs' thrd
asciilifeform: there ~are~ dead langs that got something right.
asciilifeform: afaik java simply aint 1 of'em tho.
asciilifeform: there, 'what is original, aint good, and what was good, aint original'
Mocky: sounds accurate enuf
asciilifeform: Mocky: i sympathize re that you toiled in it for 20y; for my part , i work with... microshit kernel diddles etc, for bread. but i'll be the last to say 'here's what we can learn from microshit' (afaik there is ~0 to be learned from microshit, other than how to run the hell away)
asciilifeform: some things are simply steaming piles of shit, from which there is nuffin to say, and the only use of which is that one can make a few bux cleaning'em up
asciilifeform hopes to live long enuff to stop cleaning up piles o'shit for a living, but not holding breath
Mocky: indeed. I've had my fill of java, but get paid for it so it's still in my immediate future
Mocky: but i don't have asciilifeform's mental model of a much better lang with which to compare. I have c c++ python javascript that I also get paid to work with occasionally, which ~entirely punt on threading
asciilifeform: Mocky: they forgo threading cuz on c machine you have exactly 3 options : a) forgo threading (i'ma not detail python's 'forgo threading and then lie about it' as separate variant, it is beneath contempt) b) sit down on unix's threads (pthread) c) implement own scheduler
asciilifeform: (c) is generally not an option in lightweight ('scripting') langs, so these generally end up (a)
trinque: << he's alive and well, sinfully busy atm helping me chase down some cash. sends his regards to the republic
a111: Logged on 2019-02-20 15:02 mircea_popescu: trinque did you ever see the guy irl lately ? say this year or something ?
asciilifeform: (c) also dun play very well with the underlying (b) that one is forced to contend with if you actually want physical parallelism, at least not w/out extraordinary care
asciilifeform: trinque: aah so mired in saecular matters, like phf
trinque: quite so
a111: Logged on 2019-02-21 02:06 asciilifeform: ... which means they're liable to all end up on 1 physical cpu, lol
a111: Logged on 2019-02-21 01:45 mircea_popescu: but the ensemble will not move at the speed of a single core will it ?
asciilifeform: i suppose this is good news, trinque ( i feared that he had gone to the bottom )
asciilifeform: mircea_popescu: lang that quietly locks ~errything behind yer back, whether you wanted or not, does tend to run 'at speed of 1 core', a la python
asciilifeform: ( not mega-surprise )
mircea_popescu: "This used to be the page for my fork of Fabrice Bellard's Tiny C Compiler, but I got sick of competing with a mostly dead CVS archive that nevertheless remained "official". Every time I worked on my fork it inspired new work in the old CVS archive, and every time I set my fork aside the old project ground to a halt. Even though the old tcc project repeatedly stagnated whenever I stopped working on my fork for a few months, n
mircea_popescu: obody wanted to work with me on my version because it wasn't "official"."
mircea_popescu: this is actually more substantially gnu than any tech/engineering/cs/anything. invidious wikitards rather than anything else.
asciilifeform: lolwat, what means 'mostly dead' if 'new work' ?!
asciilifeform: 'mostly pregnant'
asciilifeform: !#s plowing flies
a111: 12 results for "plowing flies",
mircea_popescu: "(People sent me bug reports about the 0.9.24 release. Yeah, that release contained a lot of code copied out of my tree into CVS, but the release was based on CVS, not on my tree.) By 2008 as CVS sank into obsolecense, TCC had clearly decided to go down with the ship. No matter how much work I put into my fork it would never eclipse the "official" tcc project (which could of course read my code to advance their tree)."
mircea_popescu: basically, the cunts were just taking dood's patches and dropping them in their toilet.
mircea_popescu: "hen I stopped work on my fork, the other project stopped, but when I started up again, so did they. All I was doing was keeping the CVS tree just active enough to exclipse mine, and I was tired of it."
asciilifeform: not even clear, from this telling, which the cunts and which the dood with patches
asciilifeform: cuz, see, no v.
asciilifeform: it resembles that story with rms & lmi
asciilifeform: mircea_popescu: is the quoted item bellard ? or sumbodyelse
mircea_popescu: i believe rob landley.
asciilifeform has nfi who is landley
mircea_popescu: ers side of the great bazaar brawl.
asciilifeform: << apparently 'My objection all along has been that CVS is not a real source control system, and no modern project should be burdened with it' etc
mircea_popescu: anyway, guy did busybox, basically.
asciilifeform: i dunget why he gives a fuck re 'eclipsed' tho
asciilifeform: wainot run along with his own patch chain, not giving fuck re other end
mircea_popescu: because he's a man alone ?
feedbot: << Trilema -- Uccellacci e uccellini
mircea_popescu: anyway, has a shitty html blog, all the symptoms people go through in that quarter or two between when they get the republic illumination and when they become actually useful/productive.
mircea_popescu: except in his case it's been ongoing for decade+
mircea_popescu: but, if looking for more targets to practice your email curse...
asciilifeform: i'ma roust him when we get own tccism going
mircea_popescu: There were functions named g() and o() <<< bwahahaha.
asciilifeform: 'My wife's name is Fade. We got married at Penguicon 2007. Fade's boss Steve Jackson officiated, and Eric Raymond was best man. (The wedding fit into a 1 hour panel slot and was moved once so as not to conflict with an Elizabeth Bear panel Fade wanted to attend, and a Charlie Stross panel Steve wanted to attend. Yes, we're all that geeky.)'
asciilifeform does not even know what to add to ~that~..
mircea_popescu: Someday if I get back to this topic, I want to glue either sparse or the tcc front end to qemu's tcg back end and produce a new compiler that A)
mircea_popescu: supports all the targets qemu does, B) can build linux and busybox and uClibc and itself (thus providing a self-bootstrapping system; I'd upgrade busybox to have missing bits like "make"). << maybe HE could be doing the tcc work, at any rate.
asciilifeform: iirc classic tcc built self + kernel
asciilifeform: mircea_popescu: if you can muster the juice, wainot try an' wake him . iirc mircea_popescu has a ~100% 1shot 1kill record for these.
asciilifeform brb,meatsystems
← 2019-02-19 | 2019-02-21 →