log☇︎
43700+ entries in 0.234s
mircea_popescu: <asciilifeform> mircea_popescu: science died. << "and yet we can't conserve tech in amber".
asciilifeform: mircea_popescu: ah, this - no. but realize, all of these nice things (rebirth of ferrite core, nonvolatile ram) is wasted on unix.
asciilifeform: mircea_popescu: speaking of magnetoresistive ram (behaves like eeprom, but no ion migration wear and virtually no write latency)
asciilifeform: mircea_popescu: can buy it now. (see, e.g, 'digikey corp.') patent lockup, monopoly, ruinous cost.
asciilifeform: mircea_popescu: mram ?
asciilifeform: mircea_popescu: science died.
asciilifeform: mircea_popescu: but forget about ice cream men. ask your intelligence men if, say, LukOil produces own terahertz silicon. is there any question what the answer is?
asciilifeform: mircea_popescu: but ice cream monger wants to live too.
asciilifeform: mircea_popescu: if it does, i can't see why you wouldn't have your own, terrahertz bus << because ice cream truck does not bring in the income to convert it to galactic ice cream empire.
asciilifeform: mircea_popescu: and memory access is the bottleneck.
asciilifeform: mircea_popescu: anyway, bad analogy. i'm speaking of actual problems where higher clock on system bus equals more cash.
asciilifeform: mircea_popescu: multiplier.
asciilifeform: mircea_popescu: it isn't a broom to me, but the 'rifle' i carry every day.
asciilifeform: mircea_popescu: just explaining that, e.g., 2x ram bandwidth - closer to 'oxen' than otherwise.
asciilifeform: mircea_popescu: what did happen was - the 'strong oxen' died. simply died. and this made micros look good, because alive looks better than dead.
asciilifeform: mircea_popescu: kiler micros did better than the mainframe <<< noooo.... not for shared-memory problems! not then, not now. please, understand the basic issue.
asciilifeform: mircea_popescu: people didn't use to upgrade worth a shit << number crunching - existed. even if you're not involved with it, realize that it was original purpose of computer. it matters.
BingoBoingo: <mircea_popescu> people didn't use to upgrade worth a shit not even a decade ago, when all the incentives were alligned for them to. << What happened was Microsoft's "Let's mimic the Auto Industries idea of the Model Year" won even if MS could not execute on it
asciilifeform: mircea_popescu: 2008 processor << can get ~2x the clock now. i fully grasp that for your application - there is no reason whatsoever to do this.
asciilifeform: mircea_popescu: if it sucks, he'll do worse than you << find naggum's piece on how 'market can only tackle one dimension of multidimensional analysis.' yes, other fellow will fold the protein faster, while running poetteringisms.
asciilifeform: mircea_popescu: it's always been cheaper to buy last year's tech and buy a lot of it and network it << to some extent, the problem i spoke of is a 'two strong oxen' problem. need shared memory.
asciilifeform: mircea_popescu: if it sucks you don't update << and lose the contract to the fellow who isn't trying to fold protein on 486.
asciilifeform: mircea_popescu: i was trying to disagree, logically, with 'we can simply preserve in amber hardware from golden age and use that for everything'
asciilifeform: mircea_popescu: i'm more or less wasting life just by having the misfortune to be parked on planet 3 in 2014. this is quite clear.
asciilifeform: mircea_popescu: except the world broke since naggum wrote this. not even the oil prospectors write custom os now.
asciilifeform: mircea_popescu: http://www.xach.com/naggum/articles/3233532779857997@naggum.net.html
asciilifeform: mircea_popescu: if i were to rent hardware, would have to do most of the management work personally on account of massive existing toolchain crud.
asciilifeform: mircea_popescu: last i checked, rented gpu was not worth the cost. but say i did. then i get: the very same modern hardware... just - rented.
asciilifeform: mircea_popescu: from a 'business perspective' very few of the things folks do to survive make sense. ice cream truck will not make the owner rich. nor the knife sharpener in buenos aires.
asciilifeform: mircea_popescu: no dc. we're talking about a handful of boxes in a cheapo office. if any more, must have dc cage, then nothing left to eat.
asciilifeform: mircea_popescu: visit more threadbare pauper's shops made of shoelace and bamboo
asciilifeform: mircea_popescu: mircea_popescu: two 0 old cpus are faster than a 0 modern << not for molecular dynamics sim. need fast shared memory. ☟︎
asciilifeform: mircea_popescu: can only afford a few dozen chickens before 'bracket' rises (cost of next tier of floor space, cooling, begin to dwarf the machines)
asciilifeform: mircea_popescu: you haven't even heard the rest of the fun. at my other job, also stuck with maximally modern hw. because how many floating point ops per second determines, more or less directly, how we're paid.
asciilifeform: mircea_popescu: the single biggest hog in my butcher shop.
asciilifeform: mircea_popescu: not only winblows, though. but it's the 'heaviest' victim.
asciilifeform: mircea_popescu: l0l!
asciilifeform: mircea_popescu: in car trunk, and it has to have room for six fire extinguishers.
asciilifeform: mircea_popescu: aha. laptop.
asciilifeform: mircea_popescu: as i tried to explain, recent hardware means, inevitably, either recent kernel, or laborious pick-and-choose driver vivisection.
asciilifeform: mircea_popescu: was explaining why doomed to recent hardware!
asciilifeform: mircea_popescu: cannot, much as i'd love to, use 486, or bk-0011, or handmade pdp-11 in fpga.
asciilifeform: mircea_popescu: need max cpu, max ram, or will hate life.
asciilifeform: mircea_popescu: on account of my profession, i need max disk, cpu, ram, attainable. in portable box. still say 486?
asciilifeform: mircea_popescu: remember, this is a box i have to actually live with.
asciilifeform: mircea_popescu: stupid << answer then. i requisitioned portable comp, could have asked for anything. should have asked for 486? with 640x480 lcd ?
asciilifeform: mircea_popescu: who exactly is doomed to grunt on recent hardware << just about anyone using issued portables. anyone with rented servers. most folks.
adlai: mircea_popescu: omg thank you tho mutth!
nubbins`: mircea_popescu cold storage for the discerning consumer, nothing more.
BingoBoingo: <mircea_popescu> i want prison-made physical btc. toothpaste. << Oh, if I go in I'll have an arts and crafts project.
artifexd: mircea_popescu: Nope. No ideas. Just asking.
leen_: need to speak to mircea_popescu
BingoBoingo: mircea_popescu: i'd pay fiddy bucks for one, and that's generous. << Nobel prize is gold so you can sell it at one of those cash for gold store fronts in the ghetto
jurov: mircea_popescu: EVERY BALANCE CHANGE ON MPEX SHOULD HAVE A REASON NAMED IN JSON/JSONSTAT OUTPUT. FULL STOP.
assbot: Logged on 27-07-2014 19:56:00; mircea_popescu: kuzetsa ah i think you gotta pay kako for access to that ?
assbot: Logged on 07-09-2014 17:02:37; mircea_popescu: something like that. so one night while in here, i see movement in the corner of my eye
assbot: Logged on 13-10-2014 03:29:22; mircea_popescu: asciilifeform nah, just watch people respond to stimuli.
adlai is wondering how mircea_popescu judges intelligence based on eye movement... it seems very dependant upon the setting
asciilifeform: mircea_popescu: recall my brick lisp machine?
asciilifeform: mircea_popescu: http://log.bitcoin-assets.com/?date=13-10-2014#870269 ☝︎
asciilifeform: mircea_popescu: this is the deal breaker for me. i actually run the monitors at about 50% gamma. << i did get it down to 30% or so. would've liked even less - while retaining contrast. simply to avoid... baking.
Naphex: mircea_popescu: fundamental value
Naphex: mircea_popescu: well you can't use fiat
badon: mircea_popescu: You're also too sophisticated for that kind of thinking (you're in bitcoin chans, afterall...)
badon: mircea_popescu: Beware big gold, you can lose a lot of money on it.
badon: mircea_popescu: Not true.
badon: mircea_popescu: It's difficult to predict which coins will be valuable, but I'm literally writing the book on the subject.
nubbins`: mircea_popescu look at the royal canadian mint for a perfect example. dozens of new designs per year, almost all of them would be "tales from the crypt volume 34" if they were books
nubbins`: mircea_popescu i have more than once kicked a violent drunk out of a bar by suggesting that he leave 8)
badon: nubbins`: I'll repaste them, one sec. I started keeping a list of examples based on mircea_popescu questions.
badon: Like mircea_popescu said, "you're getting some traction"
jurov: mircea_popescu: how is the foundation supposed to be zero asset corp. when it maintains treasury?
badon: mircea_popescu: Just how to register with it.
badon: mircea_popescu: Can you tell me about this "wot" thing that does voicing in here?
BingoBoingo: mircea_popescu: point
badon: mircea_popescu: It's happened before.
badon: mircea_popescu: I have a lot of enemies at the china-mint.info forum. They have tried to find fault, so ask them if I delete my own posts.
badon: mircea_popescu: No, what is that?
badon: mircea_popescu: That's when the discussion group was founded. I'm not sure how far back my articles go. At least to 2009, for articles that are still online on the SMF forum site.
badon: mircea_popescu: The silver chart is eluding me. You can see a review of my predictions for 2014, dating back to 2012, here: https://www.livebusinesschat.com/smf/index.php?topic=5632.msg37706#msg37706
badon: mircea_popescu: I call my shots publicly. Check my track record at LBC. No financial analyst in any market ever in the history of ever has been as accurate as I am.
badon: mircea_popescu: The lists are sortable by land area and population.
badon: mircea_popescu: No, I'm talking about in general, historically, not only now.
badon: mircea_popescu: That's an excellent example of why age and rarity alone are not good determinations of value, nor investment potential.
badon: mircea_popescu: You are correct in your impression, to a degree.
badon: mircea_popescu: Mintages of 500'000 are considered unusually small.
badon: mircea_popescu: 100 coins produced.
badon: mircea_popescu: Coins from China with a mintage of 100 can be bought for under $200 if you're lucky. The same coin in a mature numismatic market could sell for millions of dollars.
badon: mircea_popescu: Yes, those would be trackable.
davout: mircea_popescu: a way to speed the rollover up would be to not wait for each other to be around to perform our respective steps in the process
badon: mircea_popescu: That's also why the market is so reliably profitable.
badon: mircea_popescu: It is a challenge, truth be told, that's why the CC is so handy.
davout: mircea_popescu: that's if you're around too
badon: mircea_popescu, punkman: Have you heard of "the badon effect"?
davout: mircea_popescu: btw i'd be interested on feedback wrt http://fr.anco.is/2014/11/21/x-eur-november-20th-2014-statement/
gribble: Currently authenticated from hostmask davout!~davout@unaffiliated/davout. Trust relationship from user mircea_popescu to user davout: Level 1: 3, Level 2: 24 via 14 connections. Graph: http://b-otc.com/stg?source=mircea_popescu&dest=davout | WoT data: http://b-otc.com/vrd?nick=davout | Rated since: Thu Jan 27 16:22:24 2011
davout: mircea_popescu: successfully leading an it project: hard
badon: mircea_popescu: Anyway, did you see the simple research example? It's basically checking past sales prices, but the power of the CC allows you to count the number of specimen appearances on the market to determine rarity. Ordinary price guides would just show a bunch of sales, with no epiphany that all those sales are actually only the same 2 coin specimens over and over. The number of sales alone would mislead you into thinking the
punkman: mircea_popescu: no they are still working on that
badon: mircea_popescu: You haven't seen anything yet. We use SMF at the CC forum as a bug tracking system.