Stallman was ALWAYS fucking right. Listen to the guy.
This immediately reminded me of the story about the equations for relativity working at speeds faster than the speed of light. Since behaviour at the speed of light is undefined, it effectively divides the universe into two speed regimes: the tardospace and the tackyspace.
Whichever one I'm in myself, I've always figured Stallman was in the other one. Ultimately, eCommunism is parasitic. Now perhaps that isn't actually a show stopper and Stallman is as brilliant as you say he is.
Dude, you're giving Godwin's law a bad case of blueballs. I hear you loud and clear. Genocide is good for business. Or they wouldn't be doing it. This view that rationality is whatever people do is tantamount to Godwin's Stargate.
Pretty much the whole of where we need to focus our attention in this messy world is the universal patina in human affairs of lapsed rationality.
The economic premise of "expressed preference" (that the discrepancy between stated goals and actions lies in the verbal blather) was a viable (if narrow and naive) hypothesis so long as the brain remained an inscrutable black box. We have fMRI now. The underpinnings of a consistency of expressed preference are barely detectable under the junk heap of cognitive artefact.
Yet your little sermon remains a popular sermon. I'd be very interested to know what mental glee button it activates in the sermonizer. I'd also be interested to know why my mental "sigh" reflex is more powerful than most. Long ago I was attracted to Brouwer's intuitionism, where reasoning from contradiction is prohibited (this makes the enterprise less brittle to error, but also greatly impedes progress).
Hilbert famously retorted "Taking the Principle of the Excluded Middle from the mathematician... is the same as... prohibiting the boxer the use of his fists."
A desktop is a large, high-bandwidth display position roughly where your hands would stitch together two animal hides, with an efficient method to make a long series of decisions to either knit or purl, possibly supplemental with a system of analog gestures, by means of which one selects an object of interest upon which to knit or purl.
The key ingredients are bandwidth, stamina, interactivity, and precision.
The human focal plane at fingertip distance isn't going anywhere, nor is the field of view or bandwidth of human vision. The demise of the human attention span, however, is an eternal news item, since way before we first supposed that fusion power would be perfected in a single congressional appropriation.
These rumours are much exaggerated. Even in a heroine flophouse back in the era of French Indochine, you'll find some deviant skulking in a dimly illuminated corner of some back room, puzzling out Ulysses. We'll suffer our passions in whatever format circumstance demands, whether it be a lowly tablet or a cell phone. But eventually--when the magnum opus rattles inside--a man wants a bigger canvas and fewer beanbag chairs.
If I met Linus in person, perhaps in the seat beside me on a long flight, there is a line of questioning I'd be interested to pursue, which I don't think comes around that often in the eternal next big thing techno-weeny grilling rotisserie.
We all know that the segment registers in x86 are a bit of an abomination. I say "a bit" because the Deep Blue toy Titanic didn't actually sink, even if we wish it had. Let's put aside the fantasy that we go "12 for with last year's Maxim cover models" and contemplate a more realistic, trivial, and yet potentially history shattering counterfactual: that the x86 segment overlap had been eight bits instead of four bits, and the original address space had been 16MB instead of 1MB. We will further posit that given the larger address space, a 64kB dedicated address range was supplied for each of the eight expansion slots, with no possibility of device conflict (at least until DMA enters the picture). Instead of asking "Address 0x3F8 come in please, is anyone on this channel?" BIOS would ask "slot based at address 0xF1:0000, are you populated?"
I know a fair amount about the early history of x86, but I never managed to figure out if there were any applications or compilers or OS back-flips that heavily relied on 16 byte segment alignment granularity that would have been much compromised had the alignment been a courser 256 bytes (implied by an 8 bit segment overlap).
In this scenario, the average amount of installed memory would have been much greater, much sooner. Presumably, a larger share of PC purchase dollars would have flowed to installed memory. This might have dented the CPU revenue, or it might not and merely have expanded the pie faster and sooner.
DOS extenders wouldn't have been needed. Multitasking alternatives to DOS would have met with more fertile ground. Design decisions in Windows 3.x might have been a little different had the average machine already had twice as much installed memory. You wouldn't have needed EMS to install a simple disk cache. Direct mapped video cards would have been thicker on the ground sooner (maybe a few could be configured to grab multiple slots worth of address space).
But most of all, the transition from single-tasking Wintel crapbox to a fully pre-emptive multitasking nix clone wouldn't have had to walk over burning coals in the late 1980s and early 1990s. One day your DOS development machine would have had 4MB of main memory, 4MB of Hyperdisk cache (my personal favorite) and the next day it would have been running Linux 0.1.1 without gobs of buffer tweaks.
One consequence would have been more difficulty with the idea of a virtual DOS box. I remember people laughed in the OS/2 era when they learned that every virtual DOS program gobbled up a full megabyte of memory to ensure compatibility (think hundreds of dollars per megabyte). But if the DOS programs were further along into cooperative multitasking by then, they might have been enough better behaved to find a better solution.
Here's the question for Linus, with his unique perspective on the evolution of the x86 architecture. How much difference would it have made if the original 8086 design had struck the pinky fingernail with the peening hammer instead of the thumbnail? Was it actually the segments that caused so much grief or the fact that the segments were jammed into such a small address space that set Linux on its particular launch trajectory?
Don't get me wrong: segments blow goats. But every technology suffers for decisions made in the paranoid beginning of no-one gives a damn.
If we had a time machine and went back in time to the senior architect of the 8086 and told him (let's guess it was a guy) "eventually, something on the order of 10 billion chips of this design and its successors will be sold" what do you think he would say? He'd certainly say: "In that case, I'm not changing a damn thing, even the most grotesque detail, as I must have done something right. Whatever it is beats me. Segment registers blow goats."
Alternate version: if you could send one tweet back in time to the original x86 design team, what would it say? "You guys blow goats" or something more constructive?
There's absolutely no reason why you should first look for things in Google instead of asking them in a forum, other than your personal opinion that it's the right thing to do.
Uh, you didn't get the point at all. It drags the whole discussion down and consumes a finite resource of patient responses long before any intelligent questions are asked. Is that what you most enjoy? Next you'll be telling us there's no reason not to piss in a public swimming pool other than the opinions of others who swim in the same water.
It's not nearly so much of an opinion as you make it out: it's a personal value concerning how the world works and effective use of our limited resource to combat the seemingly infinite stupidity of humans in groups. Belief in human freedom is not an opinion, it's a value, and values are often quite substantive, well considered, and integrated with a larger world view. Opinions can be any old bullshit.
But I agree, he continued to harangue pedantically until he sounded a bit robotic. Nevertheless, his value has social merit and yours is slack-ass lazy.
Would we confine our vision to the Milky Way or snub the 1373 Cincinnati because Hubble smoked his? Would we shun relativity, or shelve the works of Tolkien because he and C. S. Lewis had done the same? If so, then where will it stop?
This is beneath bike shed, where the dog used to poop.
There's no immediate end to the bandwidth explosion. This would all sort itself out given a bit more time and space. Yet not only do the consumers need to watch Batman XIII no less than five minutes from now, they also want to watch it Holographic HD. Stupid fucks. What else can you say? Another decade, the network will hardly even notice the imposition of people born yesterday zooming in expound upon the blade orientation that deposited a snick of razor burn astride the groomable peach-fuzz.
Meanwhile, the networks have engineered this stampede of stupid fucks to justify tilting the economic and political landscape on a semi-permanent basis.
Stupid fucks over the glossy cliff. Happens every time.
A password prompt is as clear as an "authorized personnel only" sign. Do you go around checking if those doors are locked?
I can tell you're one of the people who simple don't get the IE/Apache "do not track" square dance.
If the client has no ability to suppress the password screen, it's not much different than Microsoft setting a global "do not track" attribute that was intended to reflect an explicitly activated user preference, which renders it meaningless.
The closest you can come with many software packages to explicitly leave the door ajar (since you can't disable the password screen completely) is to set the password to 123456 or ftp. The later is considered obscure.
Among those with strong presumptions of security competence, typing 123456 is the moral equivalent to checking whether This Door Is Intentionally Left Ajar
Among those with no presumptions of security competence, no signal exists which reflects end-user discretion. This of course soon degenerates to the tyrany of the social machine. Check out the Barry Schwartz TED talk if you don't believe me for the episode on Mike's Hard Lemonade. Social services terrorized the child and they all knew (or strongly suspected) that it was all a big mistake.
I'd guess Canadian Mennonites are the big consumers of white vinegar (for pickling--and they have the bowel disease to prove it), with homesick Brits as the big consumers of cider vinegar. Otherwise, outside of fish and chip shops, you rarely see the stuff around.
Among older English Canadians, you'd be closer to the mark with inferior/unidentifiable bubbly orange melted cheese (curds are the superior Quebecois surrogate), or for pancakes and pastries, maple syrup.
For pasta, we make an exception to the bubble orange cheese and shake Kraft brand pre-grated Parmesan out of a green plastic cylinder. You could make a similar food stuff by fabricating a solid soft-plastic cylinder, then using a mixture of table salt and MSG as grit, boring out the center until it's off-white and fluffy.
Just in the past two months cooking at home I've done Mexican, Italian, French, Moroccan, Greek, Persian, northern Indian, Chinese and Thai--not counting a leg of chicken with roast vegetables (who owns that?)
Among younger Canadians, you'd be a lot closer to the truth listing our favorite condiment as "ethnic foil pouch".
Wikipedia lists 25 different ethnic groups each accounting for at least 1% of the Canadian population. The Japanese article lacks a comparable table, because there isn't much need. A different source lists Japanese 98.5%, Koreans 0.5%, Chinese 0.4%, other 0.6% (although there has lately been a large influx of foreigners by Japanese norms). In any of the three big Canadian cities (and several smaller ones), you could without too much effort go two full months without duplicating the ethnicity of your lunch or dinner (supposing you don't count India, China, and Africa each as a single country). And if one night you're stuck for inspiration, skip a meal and tick off North Korean.
I don't eat a lot of seaweed myself, as I'm not fully equipped.
At any company that's ever designed or manufactured a cell phone, the modern doctrine is "patent first, ask questions later".
The idea that any such corporation would ratify their patent application stream against their patent-pending portfolio under any metric of superficial common sense (common sense is always superficial) is beneath the dignity of nerds anywhere, except on a slow news day, or at a once-proud page view hamster wheel and troll feeder.
Women want to know that men aren't just in it for the action. Cash would make a nice signal of male intent, but from the male side, it's too darn fungible. The man wants to make sure the women isn't in it just to reap the reliable signal.
Enter the diamond: expensive to buy (reliable signal to women), lousy to resell (putting the men more at ease), and a nice racket for the middlemen, who supply the tsunami of romantic propaganda to wrap the whole ridiculous spectacle in broad public approval.
I liked your first post, and I like your rebuttal as well.
Most people don't realize that despite all the hype about "Arab Oil", most of our oil imports come from Canada or Central or South America.
A few years ago, I saw a map of American oil imports that divided it into five regions, all substantial, only one of which was the Middle East. Here's more recent information cribbed from the FT:
Canada is the largest source of US oil imports, accounting for 25 per cent of the total in 2011. Members of Opec, the oil producers(Slashcode fuckup of trailing apostrophe) cartel, supplied 47 per cent of US imports, with Saudi Arabia, Venezuela and Nigeria the largest contributors.
Skill testing question: which of Saudi Arabia, Venezuela and Nigeria are in the Middle East? Mexico was also a large (non-Opec) source when last I looked. This is partly political, but also in large measure simple geography. It's 8500 miles from Mexico to Hong Kong, but only 5700 miles from Juneau to Hong Kong. The view is different from Houston.
For all the "eggs in one basket" people I'm reminded of the expression: The first rats off a sinking ship are the best swimmers. It might even be these same rats who shamelessly gnawed through an essential rope (who feels shame with one eye on the exit?)
There's nothing like a giant Mil-Space program lofting ugly bags of mostly water high into the celestial canopy to fluff a fevered planet's thermal blanket.
Amen, brother, about the importance of knowing when the documentation for the underlying layers is lying to you about important edge cases. This is not a typical skill out there.
A related skill is being able to observe a system behaving incorrectly and reasoning where the defect might have been introduced. You might be dealing with a program with 50 million lines of source code. Even with smaller programs, you need to have a good mind for boxes.
Ken Thompson said that the real reason Belle won so many chess tournaments was less about the custom hardware and more about having fewer bugs (i.e. programmers who really could). Bugs in a chess program are especially pernicious. Did it really evaluate all the lines of play to the desired depth? You can hardly crawl over the listing by hand to discover if Deep Blue is skipping 0.0001% of the lines of play it's supposed to investigate (perhaps because of some exotic edge case in multi-processor locking on your hash transposition table).
Knuth said that literate programming never really caught on because the group of people with excellent analytic skills *and* excellent verbal presentation skills was just too small a group to reach critical mass.
In some programming domains where there's a fairly direct and intuitive cause-and-effect model, a fairly large subset of the population could accomplish useful work.
How many people could find Intel's FDIV bug in the Pentium hardware design? The one I'm thinking of is where a tiny piece of the hardware look-up table was improperly truncated in the final mask.
Some people are good at making lists of "everything that can go wrong" and some people aren't. Even geniuses make lists too short.
We believe that tailoring your web experience â" for example by showing you more relevant, interest-based ads, or making it easy to recommend stuff you like to friends â" is a good thing.
This is "your" as the anonymous plural. I'm an individual, not an aggregate, thank you very much. I mainly have any friends left at all for not presuming that my tastes are theirs. Strangely, I surround myself with people who have strong minds and distinct tastes. My social circle is not an echo-chamber of group think.
These Big Data issues are important, but there are bigger things afoot. As you move into a society driven by Big Data most of the ways we think about the world change in a rather dramatic way. For instance, Adam Smith and Karl Marx were wrong, or at least had only half the answers. Why? Because they talked about markets and classes, but those are aggregates. They're averages.
While it may be useful to reason about the averages, social phenomena are really made up of millions of small transactions between individuals. There are patterns in those individual transactions that are not just averages, they're the things that are responsible for the flash crash and the Arab spring. You need to get down into these new patterns, these micro-patterns, because they don't just average out to the classical way of understanding society. We're entering a new era of social physics, where it's the details of all the particlesâ"the you and meâ"that actually determine the outcome.
Reasoning about markets and classes may get you half of the way there, but it's this new capability of looking at the details, which is only possible through Big Data, that will give us the other 50 percent of the story. We can potentially design companies, organizations, and societies that are more fair, stable and efficient as we get to really understand human physics at this fine-grain scale. This new computational social science offers incredible possibilities.
I don't mind my search results personalized, but my preference here is to have specific crud removed, not favoured results promoted. Alibaba and scribd and certain content mills would be early casualties, and no link to Elsevier in the top ten, ever. Mostly I can skim a list of 50 search results in the blink of an eye, thank you very much (and I don't find the skim gestalt useless, either).
Here's the thing, Google, you don't have to guess. Just give me a place to dial in my personal preferences, and then you'll know for certain: I don't want those stinking suggestions. My one burning desire in the user interface for the last decade is more capacity to disaggregate myself from faddish workflows. Ubuntu 10.10, that's how I like it, uh huh uh huh.
(*) I use a FF extension Make-Link to copy and paste links. Sometimes when you copy an all-caps link it comes out properly, if the all-caps was coded as a presentation style. I used to have an extension decaps to deal with this, but it broke in some FF upgrade. Over my dead body I'm retyping the title by hand to change the case, and neither am I leaving it there to scream at people.
Appalling coverage that we couldn't even put the word "faience" in the Slashdot preamble. What is this, MSM pity day? I still enjoy Slashdot, but all too frequently these days I loath the story submission.
Chose another profession. As someone who was a graduate research assistant, we all knew grant writing was part of the job. You want to keep doing research then you need to apply for grants.
What a horrible, defeatist attitude. I can't stand bugs in software (the vast majority exist because low standards are cheap). So I should chose a different profession?
As someone who was a graduate research assistant, we all knew grant writing was part of the job. You want to keep doing research then you need to apply for grants.
I've known since 1978 that "bugs were part of the job" and yet I persist.
Yours is an interesting perspective. The optimal solution to the marriage problem of jobs to talent is the assignment of least remorse: scientists who research on animals should have no feeling for animals, computer programmers should feel no embarrassment over bugs, politicians should enjoy lying, racers in the Tour should be human pincushions, etc.
To some extent, the world does work this way, but it's a strangely sociopathic step to actively endorse this.
Amazing what a wanger a certain kind of conservative gets over the thought of a deviant drooling. It's almost as if they think these materials are wasted on the average folk, and by outlawing its possession, seek to confine its possession among the cognoscenti.
[There follows a conversation between Hannibal and Clarice with some bearing on the title of the work, which this volume tastefully elides.]
And that day was when Intel followed the 486 with the Pentium after a fairly modest architectural change (limited dual issue) and then followed that with the Pentium Pro, which I regard as the largest architectural change (at the execution layer) in the history of x86.
The Pentium went pretty fast on selected benchmarks whereas the Pentium Pro went pretty fast on everything but a Microsoft OS != Windows NT (it also went pretty fast on everything at once--the server workloads). It just happens that the code generation strategy used for Windows 3.1 and Windows 95 were particularly loathsome to whole-register dataflow.
Aside from trademark issues with 586, Intel faced the problem of convincing people that a 60MHz 586 was faster than a 90MHz 486 (this was only true on selected benchmarks and at significantly higher power consumption levels, but who's counting?) When the Pro came along, on consumer Windows there was very little ILP performance gain to upsell with a disjoint name. If you were buying NT 4.x, you could do the math yourself.
Microsoft was coming up with weird names for a different reason: to distance themselves from their own product history. And also to keep the rubes away from the hardship of running an OS that actually worked (and therefore necessarily broke all your crapware device drivers).
I suppose this is why Henry didn't just number his wives, or keep the first one around for a long time. Or maybe it was just the wife who got the shivers when beckoned as "seven of nine" (evidently he died before his time).
Lost in the noise floor here is that Opus pretty much destroys the premise that good audio compression requires a psycho-acoustic model. These are hard to develop and limit participation.
In my own tests with Opus with my own voice dictation, I was pretty happy with Opus at its highest compression ratio.
It's pretty ridiculous to think that Opus should also have tackled lossless and extreme voice compression (codec2 territory). You people, please turn in your bignum libraries on the way out (or would you rather ditch your IEEE 754?)
Lossless is a different game and extreme voice compression is still a work in progress (I predict this can go a lot further yet, ending up as almost a MIDI file for the human larynx with a compression layer on top that models typical speech production--this may or may not be language neutral).
I do find it interesting that we are talking about statistics courses, and nevertheless this fellows seems to be taking the conclusions he drew from one course and extending them to the entire genre.
If he was saying "massive online education" can't possibly work your comment would have some teeth. He's not saying this at all. His subtext is that an educational site determined to remain free is unlikely to invest in enough content refinement to achieve excellence, and the leadership example set by Thrun himself does not inspire faith. For leadership critique, N=1 is often all you've got.
If an introductory statistics course by the main cheese happens to be the worst course offered (I somehow doubt this), that's a fluffy dark cloud over the Udacity horizon.
The corollary to your point seems to be that only a person with enough free time and energy to take a large representative sample of all courses offered is competent to make cutting criticism. You'd rather than a dull knife pouring over a large sample, than a sharp knife pouring over a single data point that--by all that's right and sane in this world--ought to cast the Udacity promise in a flattering light.
The real problem with the AngryMath blogger is his insistence on the One True Feedback Loop. Normally with academics the OTFL is peer review (Russ Roberts and many of his recent guests are sniggering and snorting and spewing their coffee over the presumption of that-which-is-published having a necessarily strong relationship to that-which-is-true--but an oasis does not a rainforest make).
Now, I've become fairly âoereligiousâ about the text of mathematics â" reading the details correctly, and writing with precision, being absolutely paramount. (And I've found that for my remedial students, this fairly simple-sounding skill is a nearly insurmountable stumbling block.)
That nails the problem dead on: the educational system has expanded so greatly as to now include a majority of students who just don't get rigour. If we made gymnastics mandatory for all nine year olds, would they all demonstrate a natural gift? Hardly. Many of the kids who barely survive the experience are precisely those who have precision of thought bred into their bones.
Could Nadia ComÄfneci cancel x from both sides of an equation? Hard to say. Would she remember to annotate for x not equal to zero? Any guesses? This would come naturally from better instruction? No, most likely it would pose a nearly "insurmountable bar".
Education is not wasted on the blunt majority. There are many other cognitive modes that are constructive and successful in life. Just don't ask them to cancel x or recall the difference between a sample and a population more than two drinks after the final exam.
It's the same reason that Wikipedia is largely good enough. Whatever errors the blunt majority finds in Wikipedia is no greater than the errors they make in their own thinking on a daily basis. Yet somehow they survive, or even thrive. And those of us with sharp edges are supposed to be astute enough to quickly recognize the difference. The first pass through Wikipedia is good enough to satisfice. When you later roll up your sleeves to make a plan better than satisficing, a sharp knife verifies his or her sources, whether Wikipedia or five grades up (I do this even with Knuth).
So Udacity (or a more astute competitor) will swell to fill the educational void where students who don't belong in the top drawer are spending a small fortune to obtain an education they largely fail to absorb. The real benefit to society is beating into these people (on the anvil of oppressive student loans) that there is such a thing as a sharp knife, and they are not it. Moral of the story: try very hard to arrange your ambitions so as not to require a sharp knife, but when this can't be managed, hire one, and don't shed
If the edit gets accepted, be grateful, otherwise stop trying and write it off as an occasionally useful resource but a failed social experiment.
Wikipedia is not a "failed social experiment". It generated content several orders of magnitude faster than many of its starry-eyed supporters predicted, and an order of magnitude more reliable than the median projection of the poo-poo brigade. Then it matured and succumbed to human nature.
What else shall we throw on the junk heap of failed social experiment? Government? Religion? Law? Agriculture? Peer review? Fermentation?
Every one of these human institutions contains horrific problems. Of these, I'd say the law is in worse shape than Wikipedia. A bad sentence in Wikipedia can be beaten down or at least successfully harassed (the people who keep restoring the bad version run the same risk of getting shut down for edit wars--though with determination and practice they might run this risk more astutely than Johnny-come-lately). A bad law just sits there and festers, never entirely off the books.
So your syllogism is that Wikipedia is a failed experiment because you can try to fight against the crap, but law is a miracle of human civilization because you can't even try to do anything about the woefully incompetent and ugly bits?
A patent itself is fairly neutral. It all depends on how it's used. The end game for this one, used offensively, is another layer of middle men.
Now we just need the xmas shopping ap. If you can see what prices your kid brother is getting and they are a lot higher than the prices you are getting for the same goods, that would make a fine gift.
its pretty clear that Samsung would not have been selling anything even close to the Galaxy phones if they had not studied the iPhone
Never heard of enabling technology (such as high-quality low power displays)? Never heard of convergent evolution? Never noticed that many technical product categories come into existence almost overnight due to economy of scale effects?
Apple clearly benefited from work at Xerox. Without Xerox, we'd still be using text consoles. I'm not so sure Samsung benefited from work done at Apple. What Apple established was credibility of consumer demand for a new class of expensive toy.
Do they own that? Or is it just the nature of business that everyone piles on to a hot new product category?
And since this is an average it means they have to sell several 75-80mpg cars to offset the low 35-40mpg SUVs.
No, it doesn't mean this at all. It just means they need to exclude every vehicle making less than 50 mpg from the efficiency designation. This is how a previous California fleet efficiency requirement pushed the majority of new car sales into the SUV and light truck category greatly enhancing profits at (mainly) American auto companies while actually decreasing the average rolling stock fuel efficiency on California highways.
Win, win, win. We don't want to be independent of the Middle East. We just want to burn their oil for nothing to maintain our single-occupancy autobahn.
Chances are, you're firewall or proxy, or clicked the Back button accidentally to reuse form. Please again. If problem persists, and other options have tried, contact site administrator.
Lameness filter encountered. Post aborted!
Filter error: You can type more than that for your comment. Slashdot requires you to wait between each successful posting of a comment to allow everyone a fair chance at posting a comment.
It's been 1 minute since you last successfully posted a comment
Slashdot Mirror
This immediately reminded me of the story about the equations for relativity working at speeds faster than the speed of light. Since behaviour at the speed of light is undefined, it effectively divides the universe into two speed regimes: the tardospace and the tackyspace.
Whichever one I'm in myself, I've always figured Stallman was in the other one. Ultimately, eCommunism is parasitic. Now perhaps that isn't actually a show stopper and Stallman is as brilliant as you say he is.
Dude, you're giving Godwin's law a bad case of blueballs. I hear you loud and clear. Genocide is good for business. Or they wouldn't be doing it. This view that rationality is whatever people do is tantamount to Godwin's Stargate.
Pretty much the whole of where we need to focus our attention in this messy world is the universal patina in human affairs of lapsed rationality.
The economic premise of "expressed preference" (that the discrepancy between stated goals and actions lies in the verbal blather) was a viable (if narrow and naive) hypothesis so long as the brain remained an inscrutable black box. We have fMRI now. The underpinnings of a consistency of expressed preference are barely detectable under the junk heap of cognitive artefact.
Yet your little sermon remains a popular sermon. I'd be very interested to know what mental glee button it activates in the sermonizer. I'd also be interested to know why my mental "sigh" reflex is more powerful than most. Long ago I was attracted to Brouwer's intuitionism, where reasoning from contradiction is prohibited (this makes the enterprise less brittle to error, but also greatly impedes progress).
Hilbert famously retorted "Taking the Principle of the Excluded Middle from the mathematician ... is the same as ... prohibiting the boxer the use of his fists."
I guess my fists don't give me a hard-on.
A desktop is a large, high-bandwidth display position roughly where your hands would stitch together two animal hides, with an efficient method to make a long series of decisions to either knit or purl, possibly supplemental with a system of analog gestures, by means of which one selects an object of interest upon which to knit or purl.
The key ingredients are bandwidth, stamina, interactivity, and precision.
The human focal plane at fingertip distance isn't going anywhere, nor is the field of view or bandwidth of human vision. The demise of the human attention span, however, is an eternal news item, since way before we first supposed that fusion power would be perfected in a single congressional appropriation.
These rumours are much exaggerated. Even in a heroine flophouse back in the era of French Indochine, you'll find some deviant skulking in a dimly illuminated corner of some back room, puzzling out Ulysses. We'll suffer our passions in whatever format circumstance demands, whether it be a lowly tablet or a cell phone. But eventually--when the magnum opus rattles inside--a man wants a bigger canvas and fewer beanbag chairs.
If I met Linus in person, perhaps in the seat beside me on a long flight, there is a line of questioning I'd be interested to pursue, which I don't think comes around that often in the eternal next big thing techno-weeny grilling rotisserie.
We all know that the segment registers in x86 are a bit of an abomination. I say "a bit" because the Deep Blue toy Titanic didn't actually sink, even if we wish it had. Let's put aside the fantasy that we go "12 for with last year's Maxim cover models" and contemplate a more realistic, trivial, and yet potentially history shattering counterfactual: that the x86 segment overlap had been eight bits instead of four bits, and the original address space had been 16MB instead of 1MB. We will further posit that given the larger address space, a 64kB dedicated address range was supplied for each of the eight expansion slots, with no possibility of device conflict (at least until DMA enters the picture). Instead of asking "Address 0x3F8 come in please, is anyone on this channel?" BIOS would ask "slot based at address 0xF1:0000, are you populated?"
I know a fair amount about the early history of x86, but I never managed to figure out if there were any applications or compilers or OS back-flips that heavily relied on 16 byte segment alignment granularity that would have been much compromised had the alignment been a courser 256 bytes (implied by an 8 bit segment overlap).
In this scenario, the average amount of installed memory would have been much greater, much sooner. Presumably, a larger share of PC purchase dollars would have flowed to installed memory. This might have dented the CPU revenue, or it might not and merely have expanded the pie faster and sooner.
DOS extenders wouldn't have been needed. Multitasking alternatives to DOS would have met with more fertile ground. Design decisions in Windows 3.x might have been a little different had the average machine already had twice as much installed memory. You wouldn't have needed EMS to install a simple disk cache. Direct mapped video cards would have been thicker on the ground sooner (maybe a few could be configured to grab multiple slots worth of address space).
But most of all, the transition from single-tasking Wintel crapbox to a fully pre-emptive multitasking nix clone wouldn't have had to walk over burning coals in the late 1980s and early 1990s. One day your DOS development machine would have had 4MB of main memory, 4MB of Hyperdisk cache (my personal favorite) and the next day it would have been running Linux 0.1.1 without gobs of buffer tweaks.
One consequence would have been more difficulty with the idea of a virtual DOS box. I remember people laughed in the OS/2 era when they learned that every virtual DOS program gobbled up a full megabyte of memory to ensure compatibility (think hundreds of dollars per megabyte). But if the DOS programs were further along into cooperative multitasking by then, they might have been enough better behaved to find a better solution.
Here's the question for Linus, with his unique perspective on the evolution of the x86 architecture. How much difference would it have made if the original 8086 design had struck the pinky fingernail with the peening hammer instead of the thumbnail? Was it actually the segments that caused so much grief or the fact that the segments were jammed into such a small address space that set Linux on its particular launch trajectory?
Don't get me wrong: segments blow goats. But every technology suffers for decisions made in the paranoid beginning of no-one gives a damn.
If we had a time machine and went back in time to the senior architect of the 8086 and told him (let's guess it was a guy) "eventually, something on the order of 10 billion chips of this design and its successors will be sold" what do you think he would say? He'd certainly say: "In that case, I'm not changing a damn thing, even the most grotesque detail, as I must have done something right. Whatever it is beats me. Segment registers blow goats."
Alternate version: if you could send one tweet back in time to the original x86 design team, what would it say? "You guys blow goats" or something more constructive?
Uh, you didn't get the point at all. It drags the whole discussion down and consumes a finite resource of patient responses long before any intelligent questions are asked. Is that what you most enjoy? Next you'll be telling us there's no reason not to piss in a public swimming pool other than the opinions of others who swim in the same water.
It's not nearly so much of an opinion as you make it out: it's a personal value concerning how the world works and effective use of our limited resource to combat the seemingly infinite stupidity of humans in groups. Belief in human freedom is not an opinion, it's a value, and values are often quite substantive, well considered, and integrated with a larger world view. Opinions can be any old bullshit.
But I agree, he continued to harangue pedantically until he sounded a bit robotic. Nevertheless, his value has social merit and yours is slack-ass lazy.
This is beneath bike shed, where the dog used to poop.
There's no immediate end to the bandwidth explosion. This would all sort itself out given a bit more time and space. Yet not only do the consumers need to watch Batman XIII no less than five minutes from now, they also want to watch it Holographic HD. Stupid fucks. What else can you say? Another decade, the network will hardly even notice the imposition of people born yesterday zooming in expound upon the blade orientation that deposited a snick of razor burn astride the groomable peach-fuzz.
Meanwhile, the networks have engineered this stampede of stupid fucks to justify tilting the economic and political landscape on a semi-permanent basis.
Stupid fucks over the glossy cliff. Happens every time.
I can tell you're one of the people who simple don't get the IE/Apache "do not track" square dance.
If the client has no ability to suppress the password screen, it's not much different than Microsoft setting a global "do not track" attribute that was intended to reflect an explicitly activated user preference, which renders it meaningless.
The closest you can come with many software packages to explicitly leave the door ajar (since you can't disable the password screen completely) is to set the password to 123456 or ftp. The later is considered obscure.
Among those with strong presumptions of security competence, typing 123456 is the moral equivalent to checking whether This Door Is Intentionally Left Ajar
Among those with no presumptions of security competence, no signal exists which reflects end-user discretion. This of course soon degenerates to the tyrany of the social machine. Check out the Barry Schwartz TED talk if you don't believe me for the episode on Mike's Hard Lemonade. Social services terrorized the child and they all knew (or strongly suspected) that it was all a big mistake.
I'd guess Canadian Mennonites are the big consumers of white vinegar (for pickling--and they have the bowel disease to prove it), with homesick Brits as the big consumers of cider vinegar. Otherwise, outside of fish and chip shops, you rarely see the stuff around.
Among older English Canadians, you'd be closer to the mark with inferior/unidentifiable bubbly orange melted cheese (curds are the superior Quebecois surrogate), or for pancakes and pastries, maple syrup.
For pasta, we make an exception to the bubble orange cheese and shake Kraft brand pre-grated Parmesan out of a green plastic cylinder. You could make a similar food stuff by fabricating a solid soft-plastic cylinder, then using a mixture of table salt and MSG as grit, boring out the center until it's off-white and fluffy.
Just in the past two months cooking at home I've done Mexican, Italian, French, Moroccan, Greek, Persian, northern Indian, Chinese and Thai--not counting a leg of chicken with roast vegetables (who owns that?)
Among younger Canadians, you'd be a lot closer to the truth listing our favorite condiment as "ethnic foil pouch".
Wikipedia lists 25 different ethnic groups each accounting for at least 1% of the Canadian population. The Japanese article lacks a comparable table, because there isn't much need. A different source lists Japanese 98.5%, Koreans 0.5%, Chinese 0.4%, other 0.6% (although there has lately been a large influx of foreigners by Japanese norms). In any of the three big Canadian cities (and several smaller ones), you could without too much effort go two full months without duplicating the ethnicity of your lunch or dinner (supposing you don't count India, China, and Africa each as a single country). And if one night you're stuck for inspiration, skip a meal and tick off North Korean.
I don't eat a lot of seaweed myself, as I'm not fully equipped.
Gut bacteria in Japanese people borrowed sushi-digesting genes from ocean bacteria
At any company that's ever designed or manufactured a cell phone, the modern doctrine is "patent first, ask questions later".
The idea that any such corporation would ratify their patent application stream against their patent-pending portfolio under any metric of superficial common sense (common sense is always superficial) is beneath the dignity of nerds anywhere, except on a slow news day, or at a once-proud page view hamster wheel and troll feeder.
Women want to know that men aren't just in it for the action. Cash would make a nice signal of male intent, but from the male side, it's too darn fungible. The man wants to make sure the women isn't in it just to reap the reliable signal.
Enter the diamond: expensive to buy (reliable signal to women), lousy to resell (putting the men more at ease), and a nice racket for the middlemen, who supply the tsunami of romantic propaganda to wrap the whole ridiculous spectacle in broad public approval.
I liked your first post, and I like your rebuttal as well.
A few years ago, I saw a map of American oil imports that divided it into five regions, all substantial, only one of which was the Middle East. Here's more recent information cribbed from the FT:
Skill testing question: which of Saudi Arabia, Venezuela and Nigeria are in the Middle East? Mexico was also a large (non-Opec) source when last I looked. This is partly political, but also in large measure simple geography. It's 8500 miles from Mexico to Hong Kong, but only 5700 miles from Juneau to Hong Kong. The view is different from Houston.
For all the "eggs in one basket" people I'm reminded of the expression: The first rats off a sinking ship are the best swimmers. It might even be these same rats who shamelessly gnawed through an essential rope (who feels shame with one eye on the exit?)
There's nothing like a giant Mil-Space program lofting ugly bags of mostly water high into the celestial canopy to fluff a fevered planet's thermal blanket.
Amen, brother, about the importance of knowing when the documentation for the underlying layers is lying to you about important edge cases. This is not a typical skill out there.
A related skill is being able to observe a system behaving incorrectly and reasoning where the defect might have been introduced. You might be dealing with a program with 50 million lines of source code. Even with smaller programs, you need to have a good mind for boxes.
Ken Thompson said that the real reason Belle won so many chess tournaments was less about the custom hardware and more about having fewer bugs (i.e. programmers who really could). Bugs in a chess program are especially pernicious. Did it really evaluate all the lines of play to the desired depth? You can hardly crawl over the listing by hand to discover if Deep Blue is skipping 0.0001% of the lines of play it's supposed to investigate (perhaps because of some exotic edge case in multi-processor locking on your hash transposition table).
Knuth said that literate programming never really caught on because the group of people with excellent analytic skills *and* excellent verbal presentation skills was just too small a group to reach critical mass.
In some programming domains where there's a fairly direct and intuitive cause-and-effect model, a fairly large subset of the population could accomplish useful work.
How many people could find Intel's FDIV bug in the Pentium hardware design? The one I'm thinking of is where a tiny piece of the hardware look-up table was improperly truncated in the final mask.
Some people are good at making lists of "everything that can go wrong" and some people aren't. Even geniuses make lists too short.
This is "your" as the anonymous plural. I'm an individual, not an aggregate, thank you very much. I mainly have any friends left at all for not presuming that my tastes are theirs. Strangely, I surround myself with people who have strong minds and distinct tastes. My social circle is not an echo-chamber of group think.
From [all-caps title suppressed]
I don't mind my search results personalized, but my preference here is to have specific crud removed, not favoured results promoted. Alibaba and scribd and certain content mills would be early casualties, and no link to Elsevier in the top ten, ever. Mostly I can skim a list of 50 search results in the blink of an eye, thank you very much (and I don't find the skim gestalt useless, either).
Here's the thing, Google, you don't have to guess. Just give me a place to dial in my personal preferences, and then you'll know for certain: I don't want those stinking suggestions. My one burning desire in the user interface for the last decade is more capacity to disaggregate myself from faddish workflows. Ubuntu 10.10, that's how I like it, uh huh uh huh.
(*) I use a FF extension Make-Link to copy and paste links. Sometimes when you copy an all-caps link it comes out properly, if the all-caps was coded as a presentation style. I used to have an extension decaps to deal with this, but it broke in some FF upgrade. Over my dead body I'm retyping the title by hand to change the case, and neither am I leaving it there to scream at people.
Appalling coverage that we couldn't even put the word "faience" in the Slashdot preamble. What is this, MSM pity day? I still enjoy Slashdot, but all too frequently these days I loath the story submission.
What a horrible, defeatist attitude. I can't stand bugs in software (the vast majority exist because low standards are cheap). So I should chose a different profession?
I've known since 1978 that "bugs were part of the job" and yet I persist.
Yours is an interesting perspective. The optimal solution to the marriage problem of jobs to talent is the assignment of least remorse: scientists who research on animals should have no feeling for animals, computer programmers should feel no embarrassment over bugs, politicians should enjoy lying, racers in the Tour should be human pincushions, etc.
To some extent, the world does work this way, but it's a strangely sociopathic step to actively endorse this.
Amazing what a wanger a certain kind of conservative gets over the thought of a deviant drooling. It's almost as if they think these materials are wasted on the average folk, and by outlawing its possession, seek to confine its possession among the cognoscenti.
[There follows a conversation between Hannibal and Clarice with some bearing on the title of the work, which this volume tastefully elides.]
And that day was when Intel followed the 486 with the Pentium after a fairly modest architectural change (limited dual issue) and then followed that with the Pentium Pro, which I regard as the largest architectural change (at the execution layer) in the history of x86.
The Pentium went pretty fast on selected benchmarks whereas the Pentium Pro went pretty fast on everything but a Microsoft OS != Windows NT (it also went pretty fast on everything at once--the server workloads). It just happens that the code generation strategy used for Windows 3.1 and Windows 95 were particularly loathsome to whole-register dataflow.
Aside from trademark issues with 586, Intel faced the problem of convincing people that a 60MHz 586 was faster than a 90MHz 486 (this was only true on selected benchmarks and at significantly higher power consumption levels, but who's counting?) When the Pro came along, on consumer Windows there was very little ILP performance gain to upsell with a disjoint name. If you were buying NT 4.x, you could do the math yourself.
Microsoft was coming up with weird names for a different reason: to distance themselves from their own product history. And also to keep the rubes away from the hardship of running an OS that actually worked (and therefore necessarily broke all your crapware device drivers).
I suppose this is why Henry didn't just number his wives, or keep the first one around for a long time. Or maybe it was just the wife who got the shivers when beckoned as "seven of nine" (evidently he died before his time).
Lost in the noise floor here is that Opus pretty much destroys the premise that good audio compression requires a psycho-acoustic model. These are hard to develop and limit participation.
In my own tests with Opus with my own voice dictation, I was pretty happy with Opus at its highest compression ratio.
It's pretty ridiculous to think that Opus should also have tackled lossless and extreme voice compression (codec2 territory). You people, please turn in your bignum libraries on the way out (or would you rather ditch your IEEE 754?)
Lossless is a different game and extreme voice compression is still a work in progress (I predict this can go a lot further yet, ending up as almost a MIDI file for the human larynx with a compression layer on top that models typical speech production--this may or may not be language neutral).
If he was saying "massive online education" can't possibly work your comment would have some teeth. He's not saying this at all. His subtext is that an educational site determined to remain free is unlikely to invest in enough content refinement to achieve excellence, and the leadership example set by Thrun himself does not inspire faith. For leadership critique, N=1 is often all you've got.
If an introductory statistics course by the main cheese happens to be the worst course offered (I somehow doubt this), that's a fluffy dark cloud over the Udacity horizon.
The corollary to your point seems to be that only a person with enough free time and energy to take a large representative sample of all courses offered is competent to make cutting criticism. You'd rather than a dull knife pouring over a large sample, than a sharp knife pouring over a single data point that--by all that's right and sane in this world--ought to cast the Udacity promise in a flattering light.
The real problem with the AngryMath blogger is his insistence on the One True Feedback Loop. Normally with academics the OTFL is peer review (Russ Roberts and many of his recent guests are sniggering and snorting and spewing their coffee over the presumption of that-which-is-published having a necessarily strong relationship to that-which-is-true--but an oasis does not a rainforest make).
That nails the problem dead on: the educational system has expanded so greatly as to now include a majority of students who just don't get rigour. If we made gymnastics mandatory for all nine year olds, would they all demonstrate a natural gift? Hardly. Many of the kids who barely survive the experience are precisely those who have precision of thought bred into their bones.
Could Nadia ComÄfneci cancel x from both sides of an equation? Hard to say. Would she remember to annotate for x not equal to zero? Any guesses? This would come naturally from better instruction? No, most likely it would pose a nearly "insurmountable bar".
Education is not wasted on the blunt majority. There are many other cognitive modes that are constructive and successful in life. Just don't ask them to cancel x or recall the difference between a sample and a population more than two drinks after the final exam.
It's the same reason that Wikipedia is largely good enough. Whatever errors the blunt majority finds in Wikipedia is no greater than the errors they make in their own thinking on a daily basis. Yet somehow they survive, or even thrive. And those of us with sharp edges are supposed to be astute enough to quickly recognize the difference. The first pass through Wikipedia is good enough to satisfice. When you later roll up your sleeves to make a plan better than satisficing, a sharp knife verifies his or her sources, whether Wikipedia or five grades up (I do this even with Knuth).
So Udacity (or a more astute competitor) will swell to fill the educational void where students who don't belong in the top drawer are spending a small fortune to obtain an education they largely fail to absorb. The real benefit to society is beating into these people (on the anvil of oppressive student loans) that there is such a thing as a sharp knife, and they are not it. Moral of the story: try very hard to arrange your ambitions so as not to require a sharp knife, but when this can't be managed, hire one, and don't shed
Wikipedia is not a "failed social experiment". It generated content several orders of magnitude faster than many of its starry-eyed supporters predicted, and an order of magnitude more reliable than the median projection of the poo-poo brigade. Then it matured and succumbed to human nature.
What else shall we throw on the junk heap of failed social experiment? Government? Religion? Law? Agriculture? Peer review? Fermentation?
Every one of these human institutions contains horrific problems. Of these, I'd say the law is in worse shape than Wikipedia. A bad sentence in Wikipedia can be beaten down or at least successfully harassed (the people who keep restoring the bad version run the same risk of getting shut down for edit wars--though with determination and practice they might run this risk more astutely than Johnny-come-lately). A bad law just sits there and festers, never entirely off the books.
So your syllogism is that Wikipedia is a failed experiment because you can try to fight against the crap, but law is a miracle of human civilization because you can't even try to do anything about the woefully incompetent and ugly bits?
A patent itself is fairly neutral. It all depends on how it's used. The end game for this one, used offensively, is another layer of middle men.
Now we just need the xmas shopping ap. If you can see what prices your kid brother is getting and they are a lot higher than the prices you are getting for the same goods, that would make a fine gift.
Never heard of enabling technology (such as high-quality low power displays)? Never heard of convergent evolution? Never noticed that many technical product categories come into existence almost overnight due to economy of scale effects?
Apple clearly benefited from work at Xerox. Without Xerox, we'd still be using text consoles. I'm not so sure Samsung benefited from work done at Apple. What Apple established was credibility of consumer demand for a new class of expensive toy.
Do they own that? Or is it just the nature of business that everyone piles on to a hot new product category?
No, it doesn't mean this at all. It just means they need to exclude every vehicle making less than 50 mpg from the efficiency designation. This is how a previous California fleet efficiency requirement pushed the majority of new car sales into the SUV and light truck category greatly enhancing profits at (mainly) American auto companies while actually decreasing the average rolling stock fuel efficiency on California highways.
Win, win, win. We don't want to be independent of the Middle East. We just want to burn their oil for nothing to maintain our single-occupancy autobahn.
Chances are, you're firewall or proxy, or clicked the Back button accidentally to reuse form. Please again. If problem persists, and other options have tried, contact site administrator.
Lameness filter encountered. Post aborted!
Filter error: You can type more than that for your comment.
Slashdot requires you to wait between each successful posting of a comment to allow everyone a fair chance at posting a comment.
It's been 1 minute since you last successfully posted a comment