log☇︎
579600+ entries in 0.356s
mircea_popescu: what i don't understand is how people like linus don't just return the "award"
mircea_popescu: this is the sum total of a life wasted. what's to prize in there ?
asciilifeform: at any rate, i'm unable to say when exactly we reached the point where 'the hunchback can only straighten in his grave' - but it was some years ago.
mircea_popescu: guy has literally not done anything in his entire life. intern in a meanwhile bankrupt, never relevant trade mag.
asciilifeform: lol that was actually my first thought when saw.
mircea_popescu: they're really labouring hard to make sure they actualy get boiled in oil, these people.
asciilifeform: notice paul graham, elon musk, buncha minor catamites, in there.
mircea_popescu: yeah, makes total sense. buterin, the new torvalds
mircea_popescu: 2002 : Linus Torvalds, Creator, Linux
gribble: The World Technology Network: <http://www.wtn.net/>; 2014 World Technology Awards Finalists | The World Technology ...: <http://www.wtn.net/summit-2014/world-technology-award-finalists>; [About The WTN] | The World Technology Network: <http://www.wtn.net/about>
mircea_popescu: ;;google World Technology Network
mircea_popescu: o wait, too late for that.
asciilifeform: check out the other nominees.
mircea_popescu: let's pick a derp with absolutely no merits to give the nobel prize to.
mircea_popescu: fucking evil, this shit.
asciilifeform: sadly, nothing there re: the 'yet to exist.'
mircea_popescu: you have got to be kidding me ?!
mircea_popescu: "Vitalik Buterin beat Mark Zuckerberg for the 2014 World Technology Network award in IT software. The award is presented for Buterin's role as co-creator and inventor of Ethereum which has yet to exist and co-founding Bitcoin Magazine. "
mircea_popescu: who the fuck heard of this.
mircea_popescu: why DOES IT TAKE SO LONG
asciilifeform: but far cry from this marvel.
mircea_popescu: https://www.whatsmydns.net/#A/trilema.com << heh! wtf is this bizarro bs.
asciilifeform: plus elaborations of how the actual deed was done in various time & places (crushing, pulling, cutting, burning, etc)
asciilifeform: mircea_popescu: the piece where explains three basic types of eunuch op
mircea_popescu: asciilifeform: i recounted various passages to pet, who cringed << specifically ?
decimation: yeah I remember the old AT chassis, built like a peice of HP test equipment
asciilifeform: (if you were good with, among other things, a jigsaw)
asciilifeform: (genuinely own, as in, ttl logic)
asciilifeform: 488 was among the last of the standard buses intended to humour folks who want to build own hardware
decimation: yeah I used to call it 'centronics' connectors, didn't realize their heritage
assbot: IEEE-488 - Wikipedia, the free encyclopedia ... ( http://bit.ly/14P1gwE )
asciilifeform: https://en.wikipedia.org/wiki/IEEE-488#mediaviewer/File:HP7935_rear_HP-IB_port.jpg << i remember when a typical pc had this kind of beautifully painted steel chassis.
decimation: the problem with 'taxing the wave of massive adoption' is that eventually when the wave passes on, one needs to ride the next wave to sustain all the fleas that have attached to the revenue cow
decimation: http://www.forbes.com/sites/greatspeculations/2014/06/10/why-qualcomms-royalty-revenue-will-continue-to-decline/ << " Qualcomm’s royalty rate for LTE-only is about 125 basis points lower than for 3G. As the industry transitions from the 3G to 4G technology, Qualcomm’s average royalty rate will come under pressure."
decimation: to complete the circle
decimation: probably many of these 'black boxes' were developed by 'startups' funded by usg tax dollars through universities
asciilifeform: as in, circuit that vendor did not look into and is in fact contractually obligated to at least pretend not to.
asciilifeform: but i would be greatly surprised if modern x86 dies did not have 'licensed' black boxes in them.
asciilifeform: i am told that each has a massive proprietary one
decimation: maybe intel or amd has an actual toolchain?
asciilifeform: in this sense.
asciilifeform: i'm not entirely certain that any piece of silicon that hit the market in the past half decade or so, is a 'real asic.'
decimation: they all want to be the next qualcomm
decimation: this is pretty much the business model for every 'fabless chip corp'
decimation: my understanding is tha altera is generally easier to deal with here, because they actually do their own r&d to produce these 'ip cores', whereas xilinx tends to contract it out - involving third parties in your product
asciilifeform: incidentally, the last time i gave a damn (> 1yr ago) i learned that all the 'miner asic' products had been produced this way.
asciilifeform: e.g., your ethernet card, ddr controller, whatever libs you had the misfortune to rely on
asciilifeform: naturally you cough up royalties for their 'property'
decimation: including all the fpga bugs
asciilifeform: e.g., xilinx wants you to prototype on their fpga, and then offers massively-discounted asic process which simply consists of their fpga die plus custom metallization layer.
asciilifeform: decimation: asic << and here's where the turdmeisters win. many 'asic' products on the market presently are actually 'hardcopy fpga'
mircea_popescu: for the sort of people who are genuinely surprised that their thing even works.
mircea_popescu: well its supposed to be a "prototype"
decimation: and if you are going to run something in huge quantities, might as well make asic
decimation: it's cheaper to run a full server than to pay for the engineering time to dick with the fpga
asciilifeform: generally used when there is absolutely no alternative.
asciilifeform: generally, you'll have it figured out by the time... the chip goes out of print.
decimation: which results in shitty RTL code to program the thing
asciilifeform: if you grind away at it long enough, you can sometimes figure out the actual physical locations of things on the die
asciilifeform: because it uses the general-purpose routing fabric, and propagation delays eat you alive.
decimation: I know folks who have done this, and suceeded, but it took many months of debugging and swearing
asciilifeform: if you go and actually try to implement, e.g., ddr2 controller, using generic logic (yes, i did, and it fit) you quickly discover that it is useless.
decimation: 'you can do anything that we allow in our walled garden, shitty tools and shitty errata!!"
asciilifeform: which makes the entire proposition of fpga a thinly-veiled scam
decimation: but yeah the number of onboard 'serdes' modules is becoming a big deal
decimation: asciilifeform: yeah or you can buy the ip core
asciilifeform: decimation: what joy. more proprietary crud that has to be on the die for fpga to talk to anything external.
assbot: What is JESD204 and why should we pay attention to it? | EE Times ... ( http://bit.ly/14OXcg3 )
decimation: the latest hotness in fpga is to use a serial bus, after folks have lived in the ddr hell
asciilifeform: the basic signalling scheme, not the voltage levels necessarily
decimation: maybe rs-482 too
asciilifeform: decimation: what of it. rs-232 is a 'time warp back to' 1962.
asciilifeform: decimation: but i escaped having to program them.
decimation: it's like entering a time warp back to 1979
asciilifeform: decimation: although did once work in a lab where quite a few gadgets were using this classic link
decimation: asciilifeform: have you ever had occasion to program test equipment that speaks gpib?
decimation: asciilifeform: I suspect the same pattern one sees with bitcoind is common in free software, 'the ancients' (typically one elite guy) writes something useful, then shitgnomes dangle turds on it for decades
mircea_popescu: the hallucination that "people getting togheter" "solve problems" is also a recurrent mode of psychotic social behaviour.
mircea_popescu: but this hallucination was alive at some point.
asciilifeform: one could dispute particulars, but the popular image of 'open source' where 'thousands of people' have meaningful contributions to a project - was always a hallucination.
decimation: I suspect most folks thought they were getting 'free warez'
decimation: or at least, there wasn't much will to oppose
decimation: everyone in the 'free software movement' was happy to allow redhat to take over everything important in userland
mircea_popescu: is bitcoin going to be a "clean room riff" on core-derpage ?
asciilifeform: not to mention the userland being there for the taking
asciilifeform: to the extent 'linux' was a clean-room riff on 'minix' - same deal
decimation: didn't linus say he probably wouldn't have done his thing if bsd was available?
mircea_popescu: asciilifeform it was alive at the time it made linux overnight.
asciilifeform: mircea_popescu: aside from examples like 'emacs', most of what is actually useful in 'open source' was written in other times, other walls, pre-dating it
decimation: really, the 'free software' movement was a kind of reaction to microsoft overreach, which has largely splintered into a hundred different corporate factions
mircea_popescu: so the remaining pretenders don't manage to cling on to the shoes of their forebears. this should be shocking or what ?
mircea_popescu: well, in truth the entire "free software" thing was dead a while back.
asciilifeform: or something like this.
asciilifeform: lest anyone think tortoise gets nothing from this deal - think again. he gets to derp in front of the other cartoon animals, that - even while his shell is bursting from megacock thrust - that nothing at all is happening
decimation: I don't blame redhat et.al., like larry ellison, they are just exploiting what they can exploit
asciilifeform: and then again halfway.
asciilifeform: achilles said to tortoise, let me stick the cock in halfway. ☟︎
asciilifeform: because that is what 'meeting halfway' is.
mircea_popescu: i was merely following the hints.
asciilifeform: decimation: they actually voted for surrender to enemy