log☇︎
35500+ entries in 0.008s
asciilifeform: simple physiological problem, really
asciilifeform: mircea_popescu: actually very similar to nao -- d00d who spend childhood cleaning chicken shit instead of on horseback, was not lord material
asciilifeform: ( mircea_popescu , recall garcia lorca's 'alma de charol' ? )
asciilifeform: mircea_popescu: the way i understand it, there's a set of folx with souls, and another, larger, set of meatpuppets with soul-emulator in ms 'excel'. and no, can't bridge gap, how would this even go.
asciilifeform: possibly suffers the delusion where 'they'll beg to have me back' or somesuch.
asciilifeform: d00d tired of cleaning monkey shit, handed temple over to the monkeys.
asciilifeform: hey, plenty of folx manage to 'miss engineering' (e.g. by playing with 'feelings' as child instead of junkyard)
asciilifeform: this -- is coming to an end.
asciilifeform: d00d successfully kept out the reaching tentacles (poettering's 'kernel dbus', etc)
asciilifeform: mircea_popescu: linus's kernel was the last remaining 'open sores' item that was 'usable from upstream' afaik. ☟︎
asciilifeform: the practical consequence will be 'aah you have iron that wasn't around in sept 2018? go and backport and come back to us with vpatch'
asciilifeform: i suspect the time to snapshot and pour cement, trb style, approaches.
asciilifeform: mircea_popescu: linux kernel as a going concern prolly has ~months of life in it, if not weeks
asciilifeform has not yet had occasion to try this
asciilifeform: for the rare exception, ada standard actually gives the tools to bake a custom heap that uses specified storage and allocation algos, bypassing the os derpery entirely
asciilifeform: ( and in fact this was sop in the aerospace/avionics world, until recently 'went microshit' )
asciilifeform: re heaps, it is asciilifeform's current pov that it is possible (contrary to popular delusion) to write virtually any nontrivial proggy, entirely heaplessly
asciilifeform: e.g. Foo(1 .. 3) := (others => 0) compiles into a memset(..., 0) ; Foo(1 .. 3) := Bar(5 .. 8) compiles into a memcpy . etc
asciilifeform: neither is connected in any way with heapology
asciilifeform: ( and memset is used to clear a range )
asciilifeform: mircea_popescu: memcpy is simply a multiword mov
asciilifeform: there's no way to get out of memset()/memcpy() or equiv, tho ( and yes a sane iron would give it as an iron primitive )
asciilifeform: you ~could~ call malloc() and free() a 'gc apparatus' (a braindamaged one, imho) but i specifically exclude it in my proggies, by banning heap
asciilifeform: there ain't a 'gc apparatus' in gnat
asciilifeform: gc ?!
asciilifeform: whole unix socket thing was ~always a massive ball of ???
asciilifeform: ( ~all i got is make_socket, get_packet, send_packet , that's it. )
asciilifeform: aa
asciilifeform: mircea_popescu: 'defined' is the item that differs in asciilifeform's, in fact. i dispensed with ~90% of the knobs simply by going 'sockets are udp, that's it'. no tcpism, i.e. no accept() and the associated streamolade, no unix sockets with their weird, etc.
asciilifeform: mircea_popescu: interesting, none of'em stay put ?
asciilifeform: plox to expand?
asciilifeform: not under this name
asciilifeform: ( asciilifeform at least posted snippets of various incomplete strange , over the yrs )
asciilifeform: but yes imho ave1 would've won from at least giving hint re what he's doing
asciilifeform: i'ma disagree that '~same' , but this will only be possible to explain once mine's posted. ☟︎
asciilifeform: i suspect that the designated replacement-torvalds, in all of its 'black chix coad' glory, is already being groomed
asciilifeform: the funny part is that it won't save him any moar than it saved the 1937 folx.
asciilifeform: i half-suspect that it was written 'for' him.
asciilifeform: mircea_popescu: torvalds's thing reads like the zinoviev&kamenev 'confessions' ☟︎
asciilifeform: eventually either he or asciilifeform will substitute in the necessary asms, and then can build proggy sans linux etc
asciilifeform: mircea_popescu: i must disagree that ave1 duplicated the effort -- his item is foundation for a fyootoor raw-irons gnat, imho
asciilifeform: mircea_popescu: torvalds danced in the noose longer than most, is all i can add
asciilifeform: ave1: http://btcbase.org/log/2018-09-17#1850727 << neat idea, and in keeping with the libc-removal . for some reason feels 'heavy', tho, but i am resisting the temptation to try an' figure out why atm, i am in the process of wrapping up my own ( conventional ) sockets thing for release today ☝︎
asciilifeform: sad, but i expected sooner rather than later. showed symptoms.
asciilifeform: oh hey loox like torvalds finally gone zombie ?
asciilifeform: ty mod6
asciilifeform: mod6: np
asciilifeform: !!v FC2D6272E9AF4411FE0E8ABE50E84DA17246A8042AC4D92F4EFD5E76CE380E04
asciilifeform: !!invoice mod6 0.04570496 8 new-gen rk disks + 20 adhesive-backed hedgehogs + s&h
asciilifeform: 1s, will cough up invoice
asciilifeform: mod6: worx
asciilifeform: aa ok
asciilifeform had nfi their face were this level of... round
asciilifeform: mircea_popescu: a++++ ocelot
asciilifeform: mircea_popescu: plox to clarify, is http://btcbase.org/log/2018-09-16#1850668 an arg in favour or against posting $subj ☝︎
asciilifeform: ok i'ma bbl
asciilifeform: simple enuff item, has small example harness, and eggog handling
asciilifeform: then diana_coman may find it useful, i'ma hurry up & genesis ( realistically tomorrow posted. )
asciilifeform: it only does udp tho
asciilifeform: diana_coman and mircea_popescu can read and use if they like it, i'ma genesis
asciilifeform: http://btcbase.org/log/2018-09-16#1850630 << i spend past day cleaning up mine, but not quite done yet ☝︎
asciilifeform falls down
asciilifeform: http://btcbase.org/log/2018-09-15#1850588 << bahahaha ☝︎
asciilifeform: right
asciilifeform has a sad half-finished one
asciilifeform: prolly will need asm keccaks, given how heavy thing seems to be
asciilifeform: this makes for 2 working keccaks nao, unless i'm forgetting somebody's ☟︎
asciilifeform: ( i suspect tho, that this kind of thing is curable with time. e.g. asciilifeform was taught brit eng, not ameri-, but i doubt that anybody can tell this today )
asciilifeform: *riyadh
asciilifeform: Mocky: asciilifeform at one time worked for a d00d who spoke the palestinian dialect. he griped that on trips to e.g. riyard he was scoffed at 'like if spoke alabama english in nyc'
asciilifeform doesn't doubt that it worx
asciilifeform: mircea_popescu's algo, aha
asciilifeform: Mocky: very difficult ( at least i found ) to grasp the phonology without a live speaker.
asciilifeform: Mocky: how's your studies going ?
asciilifeform: salam Mocky
asciilifeform: i'ma stfu at this point, as i promised to mircea_popescu not to interfere in euloric matters. will answer q strictly re gnatology, if diana_coman runs into any puzzlers.
asciilifeform: diana_coman: found it, http://ossasepia.com/2018/04/17/rfc-euloras-communication-protocol-eucomms/ , aite ☟︎
asciilifeform: ( item was copied from asciilifeform's older and not specifically related notes re 'g' , wanted to put in l0gz. )
asciilifeform: diana_coman: asciilifeform forgot that you already had the protocol. then you do not need any of this, indeed.
asciilifeform: anyway there is prolly >1 sane way to do this.
asciilifeform: the salt, goes in the cipherola payload, so gets switched out in erry round. ( given as you have a trng, this is not an exorbitant entropy expense )
asciilifeform: ( anybody lacking the secretsalt and currentkey, sees simply 512byte of rng )
asciilifeform: ideally these will not be distinguishable by any third party from any other type of packet.
asciilifeform: )
asciilifeform: ( a reply here, in turn, is not the idjit tcp 'ack', but a packet containing hash(currentsecretsalt + prevpacket) + cipherola-to-current-key , i.e. can only have been generated by the box on the other end, and can only be authenticated by yours
asciilifeform: the only adjustable knob becomes , how long to wait for an expected reply before pronouncing the connection 'dead', and that's it
asciilifeform: this way you also dun have to concern with 'packet congestion', a perennial plague of tcpdom
asciilifeform: diana_coman: if one receives an old (i.e. not currently expected ) packet ( e.g. straggler from a previous round ) it gets rejected in the same way as any other rubbish.
asciilifeform: ty BingoBoingo !
asciilifeform: diana_coman: the simplest way i thought of to finesse the reordering thing is simply to run synchronously ( i.e. at no point does sender send a new packet unless old one replied; if there's no reply, sender retries a certain # of times, and if still silence, declares the connection broken )
asciilifeform: wb danielpbarron
asciilifeform: ( if one can be found that fits 'in pocket' )
asciilifeform: at some point i'd like to get a bigendian box of some description into pizarro , for such work
asciilifeform: ( good chance to test just how portable is the latter )
asciilifeform: but since phf has a bigendian gnat, could theoretically build ave1's
asciilifeform: aaa
asciilifeform: ( imho what we need is the canonical ave1 gnat, on bigendian )
asciilifeform: phf: which gnat didja use
asciilifeform: this being said, still gotta come up with bigendian box to test with at some point ( iirc trinque had one somewhere, but afaik no gnat for it yet )
asciilifeform: ( why? i'ma leave as exercise for reader ! )