asciilifeform: that's the problem tho, if you could articulate a useful model of the c shitsoup, it would not be shitsoup, but merely another civilization with which one could potentially make commerce, like with japan say
asciilifeform: ( and, importantly, must be able to substitute without rebuilding the proggy )
asciilifeform: ave1: the reason i wrote my mechanism as seen in ch8, is not only that it was the simplest physically possible that i could think of , but that i regard any invocation of randomola where i cannot substitute a known value for testing, as a serious problem -- how is one to know that the proggy actually does what is claimed with the randomola ? rather than, e.g., 'lose' some (or all of it) along the way and take a constant as the 'rando
asciilifeform: ( phf , i suspect, knows more re: exactly how this was built, than i , if yer curious re detail )
asciilifeform: was actually quite a feat, all of the various compilers were able to freely exchange data structures☟︎
asciilifeform: trinque: the interesting bit is that the bolix folx did 'make gut bacteria', they had a c and even fortran compiler shipped with the box
asciilifeform: ( on my systems i init the FG's tty at boot, so i did not bake it into ffacalc )
asciilifeform: iirc diana_coman's also inits the tty, in a FG-specific way
asciilifeform: ave1: there you are. did you ever read my ch8 ? is the item you are trying to make, identical to the one seen there ?
asciilifeform: to learn per se is harmless, supposing you're decently good at mental compartmentalization. asciilifeform for instance worx with high-pressure liquishit from undocumented guts of winblowz kernel, ~daily. perhaps it is lethal eventually, but observe, not yet dead. ( would you say it has visible effect on style of asciilifeform's publications to date ? would you know, re the winblowz, if asciilifeform did not confess it ? )☟︎
asciilifeform: just how near to the level of monkey, are men to drop, to 'engage' with monkeys ?
asciilifeform: trinque : out of curiosity : do you see, e.g., asciilifeform's amputation of all microshit #ifdef... crapola from trb, as mistake ? 'fails to strategically engage the world-as-it-is' ?
asciilifeform: (neither easy, not quick, nor cheap, but whoever lives to the end of this, will not ask after 'ffi' etc)
asciilifeform: the correct algo, i suspect, is not to lament an abstract 'lisp' that 'became' wrong thing, but to simply confiscate whole stack from the heathens.
asciilifeform: the only 'common language' one can find to talk to bacteria -- is bleach.☟︎
asciilifeform: (nobody's getting actual separation without separate machines, no.)
asciilifeform: re the process separation -- it is moar of a cardboard wall than actual wall, yes. i for one still prefer cardboard to no wall at all between toilet and kitchen.
asciilifeform: trinque: lisps on cmachine may or may not have ever 'lived'. rather than 'killed'. extremely hostile environment , where the conventions of civilized life are unavailable , os api passes unbacked promises around, barfs mid-op in midst of arbitrary ops without any error handling, etc☟︎
asciilifeform: c,cpp, not only do not have 'standard ffi' ( try an' link libs made by microshit, say, to gnu's, or intel's ) but not even has standard... integer handling ( e.g. overflow is ~undefined~ condition ), didn't seem to hurt its, if one dare use the word, 'successes'☟︎
asciilifeform: mircea_popescu: you can 'emulate ada' in c, sure. but 'array's length is kept ALWAYS right next to the array' is not part of the language's fundamental abstractions. and os api (any and all of'em) don't follow it; and libc; and so on.
asciilifeform: there are several interacting problems. one of which is that the fundamental mechanism baked into c for dealing with arrays (where it is not possible to mechanically calculate the length of an array ~at all times~ because it is not in fact stored anywhere ) is broken in the most fundamental way possible
asciilifeform: mircea_popescu: the directly analogous to 'hong kong' algo would be (brace yerself) separate processes for ada and c-crapolade, connected via ipc (under a unixlike, prolly 'domainsocket'). because otherwise, they live in same process, and if c-crapolade is entrusted with making e.g. a valid adastring, it can lie about the length and hose the ada routine, or simply fandango over address space as c-crapolade is wont to, and so forth.☟︎
asciilifeform: mircea_popescu: legendarily, in '90s ru baseball bats were sold errywhere, but no one had ever seen the ball or the glove...
asciilifeform: pretty sure there was at least 1 day with even lighter one
asciilifeform read the log just nao, and can't quite figure out whether ave1 was attempting to implement a mechanism similar to asciilifeform's ch8 ( http://www.loper-os.org/?p=2175 ) or something entirely else
asciilifeform: not substantially different from other american 'incomes', e.g. lockheed's , ( i.e. usg prints printolade, sends straight to $designateds )
asciilifeform: ( in lattice's 'ice', the elements are homogeneous, so it was a nonproblem )
asciilifeform: the seekrit tricky bit , is that if you can't compute an ~exact~ delay path, in nsec, between any point 'a' and 'b' in the fabric, yer tool is ~useless
asciilifeform: while i have a few min on the clock : attn folx : there are 7 unsold FG units in my inventory; of which 6 are currently installed in snsa boxen waiting for their day in the rack; this leaves 1 'bachelor' FG. if someone wishes to buy this one, talk to mircea_popescu for where to pay.
asciilifeform: it so happens that i've designed exactly such a device. but it will be filled with ffaware, not koch. and all things in their proper time.
asciilifeform: let's posit that asciilifeform were willing to put his signature ever again on kochware. ( he ain't. ) how would the key get on the device ? ( i.e. is this exactly the orig cardano ? )
asciilifeform: going ~out~, that is. ( they will go '2 day fedex' unless ben_vulpes absolutely must have faster )
asciilifeform: ben_vulpes: thursday 1700 (new york time)
asciilifeform: mircea_popescu: deffect -> defect ( ? or is this deliberate )
asciilifeform: in either case you gotta pick a max length for the thing, and copy ~up to~ that many bytes. but this is prolly obvious.
asciilifeform: ... or use the secondary-stack-using standard mechanism for this ( or ave1's replacement, potentially ) .
asciilifeform: ada folx: re making ada strings out of the c variety : strlen(char *) is a potentially lethal op ( suppose the nullterminator is missing ) so it will never be called implicitly by ada. you gotta either call strlen deliberately on c side, and then form ( can be on stack , declare ... Foo : String(1 .. Length) ... , say, a la http://btcbase.org/patches/ffa_ch4_ffacalc#L53 ) a proper ada string and copy the cstring into it.
asciilifeform: because whoknows, what if reader has a ega ?
asciilifeform was just about to retort to 'insurance on fg, lol' with 'fg is eternal iron' but then remembered that cpld's eeprom is rated for somewhere like 20yrs. naturally it can be refreshed via the jtag connector and my signed payload, as many times as one likes.
asciilifeform: ok worx. all goes out by thurs at the very worst.
