asciilifeform: mircea_popescu: why would you phrase it that stupidly << the answer is, because i haven't the titanic brain that would be needed to come up with a 'smart' way to phrase it. so generally i don't bring the subject up, unless asked first. 'basic research' is a fundamentally... unreasonable proposition from the ROI standpoint. i'm not particularly good at justifying unreasonable things.
asciilifeform: see also the famous tacoma narrows bridge.
asciilifeform: mircea_popescu: since mr. m started sailing under the flag of the sv/vc circus, the thing's been well-hyped and spammed to virtually every forum where technowankers while away their days.
asciilifeform: mircea_popescu: random person i ran into today had actually heard of bitcoin, AND of urbit << random, as in, on the streets of ar ?
asciilifeform: mircea_popescu: the proverb was attributed to mathematician alan perlis. i wasn't there when he uttered it. here's the context where i last mentioned it: http://www.loper-os.org/?p=1390
asciilifeform: Naphex: that's what i did last time i had an entirely unconstrained design (a lab automation system, controlling robot & various scientific instruments attached to it)
asciilifeform: Naphex: assuming an actual choice of instruments (rather major assumption) - sbcl.
asciilifeform: fluffypony: one hint, in the past, was that apple clearly indicated a desire to see 'gcc' die in a fire
asciilifeform: artifexd: if headline is genuine, the only real surprise is that apple had not attempted this earlier.
asciilifeform: from apple's point of view - it isn't quite enough to have an effective vendor lock-in through obj. c and the mac api - it is necessary to breed programmers who can't even -think- on a competing platform.
asciilifeform: we're more or less doomed to an eternal proliferation of proprietary piss-ant languages. so colour me unsurprised.
asciilifeform: i did get a number of people playing with fpga (the closest thing to 'freedom' in this context, that i know of) who perhaps wouldn't have otherwise.
asciilifeform: (summary of some of the obstacles in the way of any serious attempt to ditch legacy crud while remaining even a little bit commercially viable)
asciilifeform: now you have games (and even virii!) with built-in 'lua'
asciilifeform: only recently we ended up with x86 boxes fast enough to disregard the performance hit from high-level languages on a cpu without hardware type-tagging
asciilifeform: jurov: 'abuse' (fairly well known game)
asciilifeform: diametric: this is sort of why i bring up my long-term research only when asked about it. and at other times, work with whatever materials are available in the jungle i inhabit, in a way that mostly makes sense to people.
asciilifeform: i've long ago given up on convincing any serious number of people of this point through rhetoric - it'll have to be by physical demonstration. but first i gotta live long enough.
asciilifeform: jurov: the way i understand it, you see 'architectural purity' as an optional frill, like the 'non-conflict zone minerals' folks, or enviro-whiners. i do not.
asciilifeform: mircea_popescu: where's that piece of yours re: 'offensively stoopid poor' and 'cattle management technology' ?
asciilifeform: 'The Ogre does what ogres can, Deeds quite impossible for Man,...'
asciilifeform: jurov: microshaft can do many things that i cannot.
asciilifeform: get the african kids started on winblows.
asciilifeform: jurov: olpc was a microshaft brainwashing project (admitted as much! hence why it shipped with an x86 turd instead of a cheaper risc cpu)
asciilifeform: kickstarter works - when it works - for 'i want to make a bicycle drive shaft from hastalloy.' and other well-defined engineering problems.
asciilifeform: re: kickstarter: lol. suggested experiment: open a kickstart, with the headline 'pay my living expense for 20 years. in the end, you might get a useful gizmo if i live that long.'
asciilifeform: i've had this discussion with more people than i can count. and the only ones who 'grokked' the basic idea, were the ones who woke up to the realization that much of what we consider to be the 'inevitable' brokenness of computing infrastructure (ubiquitous bugs, e.g. buffer overflows, irreducible complexity of os, compilers) is anything but.
asciilifeform: jurov: my half-baked fpga prototype cost me about 500 usd. but picture, in my shoes, proposing that it be manufactured en masse, to an industrial magnate.
asciilifeform: jurov: see the dc current historical example.
asciilifeform: jurov: gotta distinguish market failure from technological failure.
asciilifeform: here's an essay describing my hypothesis regarding why:
asciilifeform: jurov: we've been witnessing the failure of this approach for two decades now.
asciilifeform: jurov: you need entirely new silicon (type-aware cpu, hardware garbage collector, etc)
asciilifeform: jurov: can you change a trabant into a toyota?
asciilifeform: jurov: python, ruby, etc. are, arguably, cheap 'soviet style' copies of common lisp. as admitted by, for example, matsumoto (author of ruby)
asciilifeform: net earth will be programming these idiotic boxes pretty soon if managers don't wise up to the fact that the equivalent of automatic switches already exist and have done so for at least 20 years.'
asciilifeform: 'I remember someone saying that if it hadn't been for automatic switches in the telephone network, the entire population of planet earth would have had to be telephone operators to handle the load of telephone usage in 1993 or thereabout. I get the eerie feeling that because modern computer systems are so incredibly braindamaged in their design and in the tools used to program them, the entire population of pla
asciilifeform: jurov: smbx had perverse incentives (usg funding that appeared bottomless - until it died suddenly. reagan's 'star wars.')
asciilifeform: if westinghouse did not have similar strength and motivation to demolish the crap - we'd have been stuck with it.
asciilifeform: situation was entirely analogous to the current one re: computing - in that the inferior technical solution would have been embraced readily by equipment makers, for whom the waste is a bonanza
asciilifeform: the distinction between source and object code is an evil thing, rather like leaded gasoline - ought to have gone away a generation ago.
asciilifeform: even on a very minimal machine, with a few kb of memory, it is possible to use a language like Forth - which preserves the semantics of the code, and has genuine repl.
asciilifeform: it is possible - and has been done - to build cpu which executes high-level language more or less directly. but, given as i have an entire blog about this, i will omit the details here.
asciilifeform: people who design baroque contraptions (ml, haskell, ocaml, go - tradition) for 'smarter compiler' - that ends up grinding the original semantics of a program into incomprehensible soup - as if god came and said to mortals to use pdp-11 forever - are solving the wrong problem.
asciilifeform: you ought to be able to do 'who-calls' and other useful operations that need the machine to retain semantics of the original code
asciilifeform: it is important to understand what a proper repl consists of. i.e. a bash loop that repeatedly calls a compiler is not a genuine repl.
asciilifeform: yeah, the folks responsible for 35 years of static language well past sell by date
asciilifeform: 'go' is funded by the same bozos who spend $billion on a company making GUI thermostats.
asciilifeform: diametric: every so often, folks invent another 'ocaml' - compiled language with 'strong typing.' you lose the REPL - you lose me.
asciilifeform: benkay: and this reply to a reader asking about urbit (author of urbit was labouring under the same ill-conceived notions that produced haskell) - http://www.loper-os.org/?p=1390
asciilifeform: jurov: plus, ferroresonant monster gets 25+ years of lifetime, vs. the laughable 3 or 4 modern semiconductor-based ups is good for.
asciilifeform: jurov: 'online' ups was costly last i priced it (for 1kW+. the tiny things are worthless in my setup) and rather inefficient (~60%)
asciilifeform: jurov: the 200ms figure was taken from actual measurement of an ancient ups i owned several years ago. i can easily believe 4ms with solid-state relays, in modern unit. still does nothing against small transients.
