...is it s diffuse and decentralized nature, a network of networks, not a single network. An organization or individual with the power to "fix" the internet would have the power to destroy it or lock it down.
Ahem, buddy. To what extent is "fixing" the internet going to amount to replacing cash with plastic? Cash is bad. It contains traces of cocaine, and a guy can get mugged for carrying it around.
Plastic is good. Corporations keep records of your transactions, which they provide free to any capable hacker. You can earn credit with plastic, and one day become entitled to upkeep a mortgage.
Even if a shirt and tie internet is someday invented and deployed to the masses, I'm hoping a portal remains to the bad side of town for those of us still clinging to coin.
With email, we want some semblance of anonymity, the ability to cold-call (write to someone you've never written to before, who hasn't written to you, either), yet no ability to churn poo in mass quantities.
This is surprisingly difficult to engineer. With voting systems, first past the post is known to have more flaws than average, yet we persist with it on the grounds, I suppose, that people deserve the fruits of their inability to emotionally comprehend a system that works.
For simplicity, every alternative system is lumped under the heading "the Italian model". This scares most people more than the mafia.
I was explaining to my sweetie the other day that math is all about spending hours to crack tiny grains of rice. Many of the people who struggle with math get caught up in the manipulations. The big ideas are tiny: positional number system, the digit zero, and challenge-response proof structure (aka calculus).
Let me explain that last one. Continuity was a tough nut to crack. All that infinity, how do you stop? It turns out, you don't actually show that the slope equals a value (that would be stopping, and stopping is verboten), you instead show that error bound can be made arbitrarily small (for any epsilon challenge, a delta response exists). It's a small idea, but essential, and rest of calculus follows directly.
Fibonacci, a mathematician born in the Republic of Pisa who had studied in Bejaia (Bougie), Algeria, promoted the Indian numeral system in Europe with his book Liber Abaci, which was written in 1202
This late date never fails to stun me. So much for the obvious being obvious.
My own proposal, which I contemplated in idle moments some years back but never fully fleshed out, is that we add computational cost to the email syn packet (aka the "cold call"). email messages part of a back and forth exchange could linked cryptographically by any of the methods that prevent hostile packet insertion in e.g. ssh sessions. The details are difficult and exceed my attention span, but it has obviously been done.
The receiving mail host could inspect the incoming message, determine that the packet is a syn packet (not linked within an established exchange) and then decide to impose a computational cost on the sending machine: please factor this product of two large primes, then I will trust you enough to relay this message.
The essential feature is that this functions as challenge response: the imposed cost (product length) can simply scale as a function of how bad the spam problem becomes. If the amortized computational cost imposed exceeds the expected return, the economic incentive to push spam will vanish. It's far easier for the receiving host to generate the prime product than the sending machine to perform the factorization.
There are other asymmetrical math problems if this has some defect. It could equally be solving SETI frames or protein folding, if those have no fatal flaws. (A determination which is best left to specialists.) The size of the prime product challenge could rise and fall in a manner similar to TCP/IP congestion control: if more spam gets through, cost escalates, until the spamholes declare a loss and bugger off again.
The game theoretic proposition from the spammers perspective is this: the receiving hosts can band together to make pushing spam arbitrarily expensive.
Some legit mail (cold call subset) might entail a mail host devoting hours to a factorization challenge. Ideally this computation would be delegated back to the email origin. I'd happily let my system grind for day to authenticate one outb
The fundamental aspect of the Wikipedia concept was the fact that there wasn't a bureaucratic layer between your information and the world.
If you believed that, you believed that the purpose of Wikipedia was to provide a permanent sandbox to adolescence. Grow a pair, dude, a pair of lobes.
Perhaps you've noticed that the legal professional has this slander and liable thing. Now that the English Wikipedia is out of the boost phase, it only makes sense to put some checks and balances in place so that some juvenile pimple hiding behind a NAT the size of Singapore can't write harmful material about living persons on the basis of shoot first, ask questions later. Have you ever noticed how many times "George is a faggot" has been added to Wikipedia articles?
Banking grew up (and then regressed again), the wild west grew up, why shouldn't Wikipedia? If you ask me, it's pretty amazing they stuck to their wild west guns for as long as they did.
From Wikipedia:
Within fifty years of the Great Chicago Fire of 1871, the population had tripled to over 3 million.
Three million smells like change. Around about 1920, America experimented with prohibition.
OK, so Wikipedia won't get it right the first time. But shouldn't they try? Or should it be doomed to live in its mother's basement for all eternity so that you can personally identity with Wales, for the rest of his life, never admitting "our first conceit is no longer right".
Yeah that last one is a real warning message, pick a jar of peanuts and read it for your self. Anyone ignorant enough to not know that a jar of peanuts contains peanuts needs a lifetime treatment at the local electrical shock therapy center.
There are so many things wrong with this sentiment, from an engineering perspective, I hardly know where to begin. This flies in the face of almost everything we've learned about software development in over the last thirty years. But I'm posting belatedly, so I won't belabour the point.
1. common sense is not common 2. use case blindness: not everyone is standing in a brightly lit retail store, carefully contemplating the interior contents of each individual bottle, with a full slate of ordinary human senses 3. deniability in the edge case (we thought it was just as obvious with these sugar frosted chocolate coated peanut clusters as with the plain roast peanuts, so we didn't put the label on the bottle) 4. adding complexity to little advantage
I just love reading code with lots of condition logic to handle errors were no useful recovery is possible.
Simple: if it has peanuts, label it as containing peanuts
Complex: if it has peanuts, label it as containing peanuts, unless the product is obviously peanuts, and in case the interpretation of obvious ever needs to be litigated, here's fifty pages of criteria to legally define "obvious peanuthood".
Yeah, sure, let's reformulate the entire legal system so we can indulge in brief orgasms of self-satisfaction when the lazy idiot next to us does himself avoidable harm and wails in primal dismay. Let me tell you, I enjoy watching lazy idiots get their comeuppance. There's no greater joy in life than watching some stupid fool attempt to weave his way through heavy congestion with a series of phone-booth lane changes, on reflex and attitude and no anticipation, only to find himself mired at the center of dead stop while everyone he nipped around crawls slowly past in the next lane.
But I don't go around advocating catastrophically stupid engineering practise to enshrine comeuppance as a constitutional virtue.
After a long career in the tourism and non-for-profit sector, my brother became fond of the statement "the first rats off a sinking ship are the best swimmers". When this happens, the contrast ratio tends to increase, leaving a dysfunctional organization ever more dysfunctional. If the best and brightest of rural America are heading into the cities, the same applies. Averaged across America as a whole, nothing has changed, but you do have a slightly smaller, more woeful bucket.
Arthur Benjamin suggests we have the wrong focus in our math education. We should be teaching statistics, not calculus.
A typical person, after learning some calculus in high school, applies this skill precisely zero times in the rest of their adult life.
Statistics, however, is something we encounter on a daily basis, such as this article, with its potentially bamboozled statistical claim (did it properly account for a selective migration effect? Impossible to say from the story summary.)
People tend to have a relatively poor intuitive grasp on statistics, yet it impacts many of our daily decisions. Worse, even among those who have a reasonable grasp of statistics, few have a solid handle on robust statistics, which can be surprisingly subtle.
Does everyone know the old joke that you can take the dumbest guy from a room of 50 pound foreheads and move him into a room of evolution deniers, and the average IQ in both rooms increases (really). This is just to point out that it matters how you draw the lines, as every corrupt politician knows instinctively. It doesn't mean that a single additional person voted in favour of the corrupt politico, yet moving the line can still result in victory.
On another front, urban migration is a fact of the modern world.
I have a friend who paddles at an elite level. As the club where she presently rows, the coach recently decided to split the top six athletes three each in an A and B boat. Two things happened: A) the race time averaged across the two crews improved, B) neither boat medalled. The six elite athletes were not impressed.
In Canada, we're inclined toward this kind of social experimentation. I deliberately live on the edge of a slightly seedy area of town, because I oppose further polarization (seedy by Canadian standards is no great hazard to life and limb).
In America, the balance is tipping so that one more year of life for some rich old white fart is procured at great expense, while a far cheaper intervention for an inner city black kid, who might live another twenty years with the benefit of treatment, is often neglected.
Here's an interesting question for debate: how does our widespread statistical ignorance bias social policy? If schools taught statistics instead of calculus, would the coefficient on power-law wealth distribution change one way or the other?
Yesterday, cleaning out the back room, I stumbled across an old photo album my partner had never seen. I told her how much I hated my grade 4 year. I didn't know it at the time, but I was forced to write with a pen for the first time during the Arab Oil Embargo. For me, it was the elementary school classroom pencil and eraser embargo.
The attractive Ms Pinder also wanted me to adopt a cursive script. I had a form of written dyslexia: letters from any word that might complete my sentence would jump the queue in the middle of whatever word I was laboriously spooling out. I couldn't slow myself down enough to use either a pencil or a pen, but at least with a pencil I had a fighting chance.
It happens I lived in a small town just a country road away from where a classic computer nerd had grown up, long before this meme was established. He was already off in the big world helping to invent APL, but I would visit his parents and play in his old bedroom with his amazing robot cars and stuff. His father used to tell me the story about the first day he left the house after purchasing an early edition black and white television. When he came home, his son had every piece of it, down to the last tube, laid out on the living room carpet. His father described him as having the messiest room he'd ever seen (I felt I had a shot to compete with that one) but that when it came to his wires, they were laid out like he was taking dictation from God. My Dad had hung out with him leading an after school church group in his high school graduation year and he had shown my dad, who also a bit of engineering school, some mod stuff about computers.
Since I was an easily bored child, one night when I was making trouble as an eight year old, my dad randomly started to show me stuff he had learned from D. He illustrated the binary number system with an egg carton and some black marbles. I got it right away. Afterwards, whenever I got a boring arithmetic problem in school, I would first change it to some other random base system, solve it, then change it back to decimal. OK, you wanted me to show my working, there it is. I was sending out major distress signals (hey, I'm a little bored here) but the stun wardens of the 1970s were unable to clue in.
Ink and cursive writing and obsession with spelling drove me to new heights of frustration. For a ten year old in 1973, I had a pretty forward view of computers. I knew the spell checker was coming, I just didn't know when, or in exactly what form. The book about the nature of algorithms my dad had checked out of the university library made this clear to me: if you could define a mechanical procedure, a computer would certainly do it. The only apparent road block was actually getting my hands on such a machine. Three years later I got my hand on a TI-30, it was the best I could manage, though I did also manage to get the 8008 data sheets from a military surplus mail order outfit. I just didn't want to mow the entire subdivision all summer in order to own one, and even if I did own one, it wasn't going to spell check my essays.
So there I am, surrounded my cultural artifacts from the future, with not much hands-on opportunity, speculating wistfully about exactly what I could get away with in school, given the future existence of these machines.
My attitude was this: if I've managed to get enough of the right letters out of my trick fingers that the teacher unambiguously knows what word I've intended, then I've done enough. What's she ragging on me about? I was ripe for a copy of Shannon's 1948 monograph "A Mathematical Theory of Communication", which I would have devoured as a young child. I was already thinking hard about numeric representation and English words as code points in a larger representation space. The fly in the ointment, I realized, is that the homonyms would continue to be a problem long after the computers arrived. So I worked hard to spell the homonyms correctly (and the plural
The truth about people's food opinions are like people's opinions on global warming - there really isn't enough empirical evidence to prove anything beyond any doubt.
I don't find that insightful.
There's no controversy that human activity is adding CO2 to the atmosphere, and no controversy I've ever run into that a sustained increase in CO2 level has a positive impact on terrestrial heat capture due to insolation. There's perhaps minor controversy over how much of the added CO2 remains in the atmosphere (and for how long), and a fair amount of controversy over how quickly the additional heating resulting from this leads to macroscopic climate change.
There's enormous controversy over whether the macroscopic changes will prove cataclysmic. Few scientists are formally trained in cataclysm, so I fail to understand why so many insist on weighing into this side of the debate (but it does seem to factor into granting application approvals). I do wish that any scientist who referenced Chernobyl in cataclysmic terms would recuse themselves from the debate over the likely consequences of global warming. Nor do I completely grasp how human removal of a lake in one place, and the human addition of a lake in another place are both automatically tagged as harm to the environment. In any case, there's relatively little controversy about the fact or mechanism of global warming, and a great deal of controversy over its rate, interpretation and human consequence.
Likewise, there's no controversy that human metabolism contains homeostatic feedback pathways, though there is a great raging controversy over the situations in which (and degree to which) the homeostatic mechanisms prevail over gluttony, fate, and life style.
That said, I've never seen it claimed that the elimination of soft drinks from a diet made anyone fatter. (Even artificial sweeteners have been linked to adverse changes in caloric balance.) Neither have I seen it claimed that a moderately brisk 40 minute walk three times a week dealt any healthy person's metabolism a dire consequence (fibromyalgia is no-one's idea of good health). I suspect there's a large group of people who if three times a week substituted a 30 minute walk in place of consuming a half litre of cola, would shed ten pounds rather casually.
Most other interventions seem to have a plenitude of YMMV asterisks. Metabolism is a complex system.
You can't afford to have a rational conversation with the public about the merits and demerits of your product, because your job is on the line if the sales curve slacks, so you've got to do your damnedest to sell the thing, no matter how awful it is.
Yes, but you sacrifice the goose (credibility with the public) as a byproduct of living to cuss another day. For me this is the central subject matter of this thread: that the race to the bottom has exceeded itself to such a degree that the press release has become just a fancy spam packet in the mind of the thinking public.
My definition of hell is entering a profession where my victories and my defeats verge on indistinguishable. Monetize any lingering shred of credibility, and once the credibility is gone, move on, like a prostitute whose desirability is inversely proportional to her career earnings.
From a game theory perspective, in a game of competent rivals this outcome should not exist: the cost of manufacturing the PR exceeds the value imparted to those who read it (at arms length with nose between fingers).
In the real world, there are sheep to be fleeced. A relatively small proportion of the public is essentially suckered by PR verbiage into providing just enough marginal profit to employ the hopeless schmuck who churned the thing out, so despite having reached bottom, it doesn't mercifully fade away.
I say this knowing that the conversion rate on penis enlargement is shockingly large. I'm just not capable of wrapping my mind around the vast septic system of human gullibility. We're a depressingly dumb animal once the primal pathways light up and the PR now functions as an unpleasant nasty-gram of human nature.
I'm stunned, I don't know why, to see people debating this as if this is the first time the issue has crossed their consciousness. News flash: this has been in the public water supply for at least two decades now. It's important, and if you haven't given it some thought long ago, you're not taking life seriously, you're just a woodpusher in the game theory of human realpolitik.
It boils down to a very simple premise: that entropy is a munition.
If you have some large chunk (say 100MB) of random bits in a file on your computer, there is no way to prove that there isn't some password that will decrypt this block of bits into meaningful information. Any chunk of information content which looks like pure entropy can be accused of harboring munitions, if you're trying to hit the preservation of society nerve, or child pornography, if you're trying to hit the righteousness of the flesh nerve (we all care about flesh). Steganography is the art of boiling a thin soup: very small amount of pure entropy hidden in a huge amount of tedious backdrop (say 200GB of licit pink matter).
If you have a large quantity of real physical entropy, there is of course no way to produce a password, and neither is there any way to prove that the entropy is real.
The authorities find this unbearable, so we are now deep into guilt by association. Caught hanging out with random bits, go directly to jail.
Any public discussion of the matter would conclude that our social concept of judicial fairness is incompatible with this new guilt by association model. What kind of society would declare entropy a munition? How would we all go about scrubbing anything that looks like entropy from our electronic records? It's not clear it is possible to comply with the implications of this, even if greater society drank the Orwellian Spook-Aid.
Hence the secrecy. If the spooks destroy 1000 innocent lives in the course of protecting society as we know it, it appears to be a cost we're going to have to bear.
The easy way to cease to think seriously about this is to invoke Stalinist escalation: that 1000 lives is soon 30 million lives.
Don't be so hasty. Sun Tsu beheaded one giggling princess to make every other princess march with the discipline of soldiers. For his needs, one was enough.
The credit industry doesn't work on principles much better than our agents of darkness. The suits have succeeded in labeling credential fraud as identity theft. Note the slight shift in blame here: it's not the design of VISA at fault (which could hardly be worse), it's your fault for offering up your digits in the first place (well, you can't use your VISA card without doing so, but why niggle?)
I hand pieces of information about myself to thousands of institutions. If the information is gathered and used against me, somehow I'm to blame, not the thousands of institutions who regard protecting the sensitive information they demanded from me as a cost center to be outsourced to India.
The great line in Brazil is "Confess quickly, or you'll jeopardize your credit rating."
Our credit system is nearly as arbitrary and secretive as this business of guilt by entropy. Innocent before proven guilty. The credit system is exempt from our normal social protections against slander. Any merchant can file a damaging untruth about me with little basis in fact, few avenues of complaint, and no ultimate liability whatsoever. The rating agencies will then spread this slander around and I can't prosecute them for spreading damaging falsehoods about me, even if I finally prove that the original merchant lied, and no sensible agency would persist in believing the original claim.
If we're not up in arms about the violation of our social norms concerning slander implicit to the credit industry, I don't harbour much hope that cottage outrage in this forum over incrimination by entropy is going to make any dent in the real world.
Stay tuned for the next exciting chapter, where encryption keys are extracte
The level of skill it implies, the time and the money, is out of reach of any ordinary user.
This would be an extraordinarily hard sentiment to formally define. How far back in history does one need to go to say the same about literacy? How far would one need to travel in the present world? Long before your definition reaches bedrock, it all becomes relative to a social construct.
Two of the great innovations of our number system (positional representation, and the digit zero) were incredible aids to making numeracy less "out of reach" for ever larger segments of the population.
The upward swing of the innovation cycle is making the dog walk in the first place, however badly. If the technology becomes pervasive, this is followed by the outward swing, making the technology ordinary. This point was also neglected by the post who suggested that open source needs to be better than closed source to overcome the adoption hurdle. But that only applies in the boost phase of the innovation cycle, not to ball point pens.
The desktop OS is halfway through the commoditization cycle already. For a broad audience, the browser already matters more. Decisions are increasingly driven not by what is best, but by what is most hassle free. Cost is not driving the bus. There are some pieces of software offered with no monetary cost I won't install because the software doesn't do enough for me to justify reading the license agreement, and the organization hasn't maintained a reputation where I'm willing to install the software on trust. (Ubuntu seems to be actively scouting the boundary with this latest move.)
Speaking of licenses, IANAL => literacy ain't worth much. The governing rules of society are out of reach to those governed. So that's what "out of reach" gets you as a debating tactic: absolutely nothing. Out of reach is not merely taken for granted, it's a governing principle.
The hassle factor is asserting control over the behaviour of our installed software has less to do with the learning the C language and more to do with byproducts of the software engineering life cycle. We're at the point in the innovation cycle where invention of the digit zero would be incredibly welcome.
The problem is that our software has an emotional IQ which in the animal kingdom would be dead square in the quadrant "too dumb to live".
A mother bear gets a bit testy about the space between herself and her bear cubs. I get a mite testy about a software installer shuffling around a system configuration that was carefully tweaked. To cite the most extreme example, I once lost nearly a month in a software development process because some stupid Microsoft JET accessory (don't ask) swapped a defective DLL in place of a DLL I had carefully chosen to be compatible with some other quirky POS (thanks, Microsoft).
In the animal kingdom, you take one look at momma bear, then you consider your survival odds if she gets testy about your next foot step. This the emotional IQ our software needs to develop. This is difficult at the present moment, because my software is blind: it doesn't know who the fuck I am. It doesn't know I have "bad mother fucker" tattooed on my wallet. (Note to the Sun Java installer: the next time you install the Yahoo toolbar because I forget to click off the Yahoo button--after clicking it off 13 out of 15 times already--I'm going to rip out your giblets and engage in a pagan ritual.)
This blindness is a byproduct of an inferior technology: our software engineering and release life cycle. As software engineers, we haven't yet figured out how to function effectively in a world where _every_ end user software install takes into account a personality profile of the software victim.
Testiness factors
1. maybe it's cool 2. what the heck is a DLL and why should I care?... 5. don't mess with it if I don't understand it... 10. mother bear 11. Godzilla
This makes a lot of sense, but given that Slashdot functions as a write-only medium, I doubt your advice will be widely heeded.
I used to think that literature was the human invention which freed us from our smallness of mind. Now that drama has been replaced by reality, and literature has been replaced by pop psych, could it be that we're better off turning the page on turning the page? The distilled essence of 10 billion lives lived is heavy going. Who has the time?
Half the commentary track on Criterion's Solaris (Tarkovsky) was about the female matrix's success at "becoming more human" and her human companion's inability/reluctance to love. What the hell does "more human" actually mean I wondered to myself. I guess that's what the movie was trying to tell me.
Then I watched The Last King of Scotland. Idi Amin struck me as frightfully human (though the Garrigan simulacrum did not).
I find this notion that there is some intrinsic state of "more human" emotional activation too Zen like for my taste. The presumption that the average person has a reliable in-built compass for pursuing this state strikes me as akin to "just keep driving around until you've memorized all the streets, then you'll know where you're going". There's no substitute for self-reliance--if you haven't lost the ball (or run out of gas) before enlightenment dawns.
A map of the human spirit might be a better starting point for those of us whose days are numbered. Some maps are better than others. Pop psych is the literature of the drowned.
Well, stop drowning and start swimming. I agree completely.
Sorry to sound like a Microsoft fanboi or whatever, but Word is a more powerful tool than most give it credit for or bother to figure out, since a lot of its capability is kind of "hidden" to make it user friendly out of the box.
The number one skill for getting along with Word "out of the box" is learned helplessness. Let it randomly move everything around whenever it gets the whim to do so. If your fonts mysteriously change, just call it art.
I often help other proficient computer users unmangle their Word documents when they get a corpse bride formatting artifact. Sometimes I just have to use the mouse to delete something that the keyboard delete fails to fully remove, or use a left delete instead of a right delete. Sometimes I boil the corpse in Open Office, then return it to the scene of the crime, where it behaves fine... until the next corpse bride resurrection.
Why should a simple word processor have a steep learning curve before one achieves any semblance of predictability? AmiPro made it possible to write a quick ten page document without encountering a single undead artifact. WordStar made that possible. Back in 1985. Why not Word?
The commentary track for Last King of Scotland notes that foisting your failures onto the shoulders of others is a symptom of megalomania.
Idi Amin: Yes, you are my advisor. You are the only one I can trust in here. You should have told me not to throw the Asians out, in the first place. Nicholas Garrigan: I DID! Idi Amin: But you did not persuade me, Nicholas. You did not persuade me!
Amin was a difficult man to work with. Word is a difficult program to work with. They both have a fondness for dancing bullets.
I'll happily add Word to my toolkit the day Microsoft publishes a suite of 10,000 regression tests enforcing predictability, not for the benefit of some obscure legacy macro package, but from the perspective of the end user writing simple documents.
Article contains the terms "ATM Machine" and "PIN Number". Read at your own risk.
Languages are shaped by cognitive cost. This is what Steven Pinker seems not to get. There _is_ an innate language instinct, it's just not what he thinks it is. What we all share is the ability to introspect the cognitive cost of figuring out "WTH is this dude trying to convey?"
One of the key insights on language is that Lempel-Ziv compression never transmits the compression dictionary. The dictionary is implied because the compression program and the decompression program share the same dictionary construction heuristic. This is a trick you can pull off only if the two sides of the channel share the same cognitive architecture. There are no shortage of examples out there of how fast communication breaks down when the parties begin with fundamentally different premises on how to structure the categories of thought.
Here's another fundamental question: what portion of the brain's cognitive activity is devoted to power management? For one thing, glucose is precious resource, and the brain is a chug-a-lug organ where it comes to glucose consumption. For another, the brain is costly to cool. From the real-time perspective (which governed 5.999 million years of human evolution), there's not much use firing up the abstract-noun chocolate factory when you need a survival response in under 100ms.
There's another truism here: fool me once, shame on you, fool me twice, shame on me. (Or, if you've spent forty years fouling your spark plugs, "fool me once, shame on -- shame on you. Fool me -- you can't get fooled again.")
When you get surprised by a lion, first you need to act, secondly, you need to record, to avert recurrence, after deferred reflection.
However, the brain does not record broad-spectrum. There's just too much. It's easy to build a PVR these days with 1TB of storage. I still haven't seen one where the tuner is replaced by a DC-to-daylight recording mode.
You can't defer deciding what to record for very long. So this is an obligatory cognitive function when your brain is already heavily loaded. At high enough stress levels, the recording function does shut down. Assessing and responding to cognitive burden is a mission-critical survival function. This is a key foundation for language learning.
A child doesn't need a special gene to discover the linguistic consequences of garden path sentence structures. "Oh damn, my mind when the wrong direction, and I wasted cognitive effort". Thus a child can self-infer a constraint on viable grammatical form, even if, in the manner of an LZW dictionary, the constraint is never explicitly conveyed from the language proficient to the language learner. The underlying assumption that makes this work in practise is that the architectural model of the child's brain resembles that of the rest of the population. This is 99% satisfied by being a member of the same species, without any weird genetic Pinkerisms.
As the language convention becomes more sophisticated, some parameters in the ambiguity resolution process become social constructs. Given a conflict between two heuristics, which takes priority? The important thing to realize about socially determined linguistic parameters is that they tend to vary across discourse settings. Experts have slightly different rules among themselves than apply in heterogeneous settings, where, e.g. half the people involved are ESL.
There was a thread here the other day on the consequences of a non-specialist treating guilt and liability as vaguely synonymous in exactly the wrong forum (wrists cuffed to ankles by the minions of RIAA).
A person incapable of pedanticism is not likely to succeed with either law or software. (This is one of the reasons why the IANAL meme on slashdot annoys the hell out of me: if the law is too complex to be successfully interpreted by a concentrated group of the weediest pedants on planet earth, just maybe perhaps the root c
That they are scrambling to censor is proof this is absolutly 100% legit.
Played much at the $1000 no-limit Hold'em table? Drowned any witches lately? Three weeks overdue on your first assignment from the first week of Introduction to Counter-Intelligence 101? Completely missed the confirm/deny coin-flip that has become the sole entertainment value of MLB in the juiced-ball era?
Obama has publicly opposed waterboading, but the fervently uptight spooks would still like know what the baddies are thinking. Why don't you apply for the job, you seem to possess special powers of seeing right through any kind of mask a person wraps around his/her motives.
But hey, run a spell checker on your application form, you don't want to be excluded for the wrong reason.
I had good initial success with ltspice under Wine. Not perfect: some dialog boxes don't focus input fields normally (but if you're persistent, you can get it to work).
No problem with the simulations. The command line simulation is more powerful than I at first suspected. I especially appreciated the ability to do multiple plots in parallel. Some basic logic elements were missing (multiple input muxes, IIRC). So there was a startup curve learning how to make my own.
Once annoyance was the plotting setup. The default background is black, which makes screen plots great, but sucks for printed output. You can change the background to white (you also have to make all the colours darker for contrast), then it works fine for printing, but your monitor display is nowhere near as good, and I never found a way to automate the switch. It needs a way to set up the monitor and printer palettes independently.
There is an RLGC simulator for coaxial transmission lines, but it's not fully general. According to my notes: "At least two of RLGC must be non-zero. If G is non-zero, L and C must be zero." I understand there is a way to transform coaxial parameters so you don't need G non-zero, but I haven't yet learned how to do this, so this was a bit annoying.
At once point an obscure error message from ltspice exactly matched the string in source code I found online for Spice 3f4. I believe the underlying simulation code is forked largely unchanged from 3f4.
I've done a lot of embedded programming, mostly in C/restricted C++. I often contribute to the schematic design at a block level, or for a key circuit element critical to the technology, but I'm far from an everyday EE guy. So you can either conclude that ltspice was easily adopted by a non-specialist (well, an elite non-specialist) or, if you have less success than I did, you can conclude that I just didn't push it very hard.
Ousted Chairwoman Patricia Dunn and former ethics chief Kevin Hunsaker surrendered, were booked and released Thursday, a day after being charged -- along with three private investigators -- with felonies for their roles in HP's spying scandal.
Ethics officer dragged off in handcuffs, did it really hurt HP's business? What was it all about? It was about keeping their dirty laundry behind closed doors, no matter how appalling or borderline illegal.
I'm more of a KTB than a CYA. KTB = kill the bastards
CentOS is an example of "life happens". Interesting how many uninformed people who just caught their first whiff of this immediately chime in to explain that they should have tried harder to keep a lid on this, without even checking that maybe this box already had a tick mark.
The underlying assumption is that making an effort to keep a scandal behind closed doors will always work if the people involved warrant respect. Similar to the belief system of the HP executives. In their quest not to be damned by their knee-jerk shareholders for not trying hard enough to bung the leak, they went all the way to extra-legal.
I'm tremendously unimpressed by people who maintain their social standing by undermining the credibility of our public institutions.
Fundamentally, most support contracts are a tax on social insecurity. Am I the only person on slashdot with a vastly better track record at debugging failed software on my own steam than getting assistance over a telephone support hot-line?
I've had a few excellent external support experiences. On the other hand, in the time it takes to fight through the telephone support system of a typical company to a person who actually knows something I could have reversed engineered the antikythera.
Or in the case of HP, when one of their printer drivers made a friend's Windows 2000 machine unbootable, and their driver uninstaller refused to run under safe mode because it demanded a higher screen resolution that safe-mode VGA, and then my note about this on their support website (which took me 15 minutes to compose) returned "404 not found" after I pressed submit. So much for big, redundant iron. The dudes can't even keep their customer-support web server running (for customer_type==peon). Nothing screams "we care" like "404 not found".
That incident with HP cost me half a day of my life. End result of my support escalation: "404 corporation doesn't give a damn".
I once spent half a day debugging an obscure failure of EAccelerator in which tried dozens of Apache settings, ended up hand compiling, and then ended up running the whole thing under strace, finally filing a bug report which was incorporated upstream.
Which of those two experiences do you think I'm willing to repeat? Which of those experiences made me feel lower than an earth-worm? Which of those experiences made me feel like a useful member of the human race?
The most interesting property of the CYA reflex is how quickly and thoroughly it vaporises irrelevant considerations such as life quality.
The human brain was designed with a kill-switch on life quality. That's an amazing artifact from six million years of evolution under the parameters of nasty, brutish, and short. And still we marvel at the human capacity for genocide. Our CYA reflex is not the main player in this, but if you follow the wiring, I'm sure it's highly interconnected with the culpable wetware.
Paul Collier argues that most nations which experience a natural resource windfall experience a few great years, then end up worse off than before.
The countries where this doesn't happen are countries like Canada, which has a complex system of checks and balances. The "instant democracies" which have the vote, but none of the other trappings of effective government, aren't so lucky.
One of the reasons this works in Canada is that we accept what the government is there to accomplish, even if we grumble as much as any other nation about the obvious inefficiency in how this transpires.
The growing problems in rich nations with health care delivery runs much deeper than government. Our medical technology has reached the state where the last five years of our life expectancy (in failing health) is capable of consuming most of the wealth generated in our working years.
Most people wish to live as long as possible. Without checks and balances, the logical outcome is that 100% of the GDP is ultimately devoted to life expectancy and medical intervention. We're far from running out of ways to make health care more expensive for incrementally less return.
The way it works in America, as I understand it, is that a lower-middle income wage-slave works their ass off for 40 years, saves up enough money for a modest retirement, soon breaks a hip or experiences some other common medical ailment of the golden years, and is then hustled back to work at Wal-Mart for another decade to pay it off.
Politics in America has always cultivated a large pool of docile and desperate workers. I don't know if this has its root in the slave trade or not. I do know that it remains easy in America to end up out in the cold.
How many of the workers spoke up about the clearly ludicrous lending practises? At risk of losing their health coverage? Or some form of power against peon litigation? Not likely.
Historically, a productive economy was seen as a fragile thing, as if the economic miracle might just as easily evaporate. Hence we structure society with many have-nots, on the ground that it keeps the haves hard at the grindstone. We tend to regard innovation as some kind of fleeting accident to be carefully protected. And we create complex systems of law and property around the flowers of human creativity, lest we enjoy ourselves too much, and wind up in coffins we haven't paid for.
It's not clear that the world works this way any more. Innovation might become a surfeit, rather than a paucity, if our mechanisms to protect innovation were less obstreperous. We could have millions of artists rather than thousands of celebrities. The economics of distribution have changed, but our system of thinking hasn't.
One thing is certain, though. We're living on a crowded planet, and we're going to need more cooperation rather than less. This will mean more government rather than less, especially if the people who oppose government fritter their energies banging around in their ideological tin shacks.
My number one wish for better government would be a less bewildering and capricious legal system. My particular nightmare is that the environmental calamity arrives on schedule (fat chance), the technology is there to do something about it, but the lawyers can't work out how to divide the royalties, so the technology is not fully deployed, or too late to matter.
I've lived in the area for a long time, and never heard a good story about the VSE (RIP 1999), it's remains, but not the lingering stench, since composted into the CDNX.
Wikipedia just provided me with a funny story about the VSE I didn't know, but find all too typical.
The history of the exchange's index provides a standard case example of large errors arising from seemingly innocuous floating point calculations. The index was initialized at 1000 and subsequently updated and truncated to three decimal places on each trade. The accumulated truncations led to an erroneous loss of around 20 points per day. Over the weekend of November 25-28 1983, the error was corrected, raising the value of the index from its Friday closing figure of 524.811 to 1098.892
Are these the same people who are proposing to solve the fusion problem with 220 synchronized penises? Good god, I hope not.
For the record, here's what $500m typically buys you in British Columbia.
The Washington Marine Group sold the three ferries the company bought from B.C. Ferries for $19.8 million, to luxury yacht builder Abu Dhabi Mar.
Four cents on the dollar. That beats the old VSE hands down. Vancouver has a world-class ethnic cuisine, has enjoyed some decent success in video games and film production, but has a terrible track record with anything that floats.
Ballard Power being one of the more buoyant exceptions. I just did a search on "Ballard Power profit" and was pleasantly surprised to get a hit.
I suppose if General Slammer raises $500m to build the commercial scale reactor, they'll use our excellent BC shipyards to fabricate them. We're good. We can weld aluminum into structures less valuable than the original metal.
While I've never met a lumberjack I didn't like on a personal level, I have to say as voting collective, they're dumb as stumps. We inevitably get the government we deserve. Our big project always make work, but rarely make money.
In rural areas of BC, it's easy to spot the people with jobs at the local mill or the local mine: they've got more equity sitting in the driveway than in their shit-box house (4x4 trucks, boats, campers, skidoos, jet-skis, ATVs, etc.) Big nature, eh? You can't govern in this province without earning this vote.
We do have some nice mountains. Vancouver is planning a party to show this off. You might have heard of it. I think the plan is to lose a lot of money proving we're world class and shrewd at business.
The value of if/then for this mechanism is completely obvious. If the sun doesn't come up tomorrow, without fully taking this into account, the phase of the moon calculation will come out horribly wrong.
Computing was hard back in the day. Moses climbed Mount Sinai for some peace and quiet while he chiseled out his astrolabe's stack dump as part of a warranty claim, ate some strange mushrooms on the way up to relieve the tedium, and the rest is history. (One wrong chisel stroke and your claim is rejected completely, but that's another story.)
By the anally-uptight bifurcating decision path definition of computing, op amps aren't computers either, but it's truly amazing what a wizard can build with the little buggers.
From another perspective, the ancient Greeks deserve a major KISS bonus for making a mechanism even a fool could use. Instead, we complicate the definition of computer and take marks off.
I have a copy of Calendrical Calculations at hand. Opening to a random page, I transcribe this formula:
Or something like that. No conditionals in sight. I've used a computer to write complicated little programs with no conditionals or flow control and been tremendously happy when the answer came out as expected.
On the other hand, the equation for Easter involves a conditional expression, which looks like this:
if shifted-epact = 0 or (shifted-epact = 1 and 10 (g-year mod 19)}
In defence of the ancient Greeks, Easter hadn't been invented yet. But there were omens. While the Greeks were perfecting their teeth, the Romans were hard at it sharpening the nail. Tooth against nail, the nail prevailed.
To further complicate matters, it's known that one-dimensional cellular automata can function as a general purpose computer. However, the mechanism that executes a one-dimensional cellular automata can be built with no use of conditional mechanisms more complex than trivial boolean equations.
So we have it that evaluating (218.3164591 + 481267.88134236 * c -.0013268 * c^2 + c^3 / 538841 - c^4 / 65194000) makes you a calculator, whereas evaluating the boolean expression (a && b || b && ~c) makes you a general purpose computer (or at least capable of executing one). Perhaps this debate is silly.
In 100 BC, I suspect the idea of making something before you decided its purpose was not in wide currency. "General purpose" hadn't been invented yet. If it was being done, it wasn't been done under a creed. The ancients made a lot of pottery, sometimes (I once read) in volumes where it was easier to throw it away than wash it, but I'm sure they didn't consider mass production the founding credo of new-age industrialism.
Is solving a Sudoku a calculation or a computation? I could set up a machine to enumerate the solution space with no conditional logic aside from a stop condition when the displayed digits match the givens and the other Sudoku constraints are also satisfied. The mechanism would hardly be more complicated than an Enigma machine, which had no in-built conditional logic.
If you claim that it's the mechanism not the problem which distinguishes calculation from computation, who is to say the Greeks didn't have a general purpose understanding of how to implement single purpose mechanisms? Part of the deal is miniaturization. The general purpose implementation is a lot less attractive if it's the size of an aircraft hanger. Which would you build?
Some days I hope our civilization soon collapses, and the only device found 2000 years from now to commemorate our technology is an iPod.
"It's cool, but it doesn't really do much. Not even a two qubit superposition."
"Wait just a furry nanosecond there, this sand-baking thing is a lot more sophisticated than it first appears. However crude, I think the ancients were on to something with broader applications."
Stupid on so many levels my mind was reeling until that much needed shot of oxygen.
What kind of anger bubble does a person have to live in to fail to notice that causality calculus never works in the real world? What kind of dank basement does a person need to live in to fail to notice the chastity debate? I think it was the Zimbardo TED talk where he observes that 70% of American teenage girls who vow chastity before marriage succumbed to temptation within a year, and the majority of these do so without use of birth control. Subtracting out the sinners, the program is a complete success.
If society wished to alleviate the enormous waste of time and fuel of people stuck in traffic jams on heavily congested highways, we wouldn't let the highways become nearly so congested in the first place. Congestion based use fees would encourage corporations to be a lot more flexible about working hours. The cost to their employees would be right there on the bottom line for the bean counters to whittle upon.
But the typical population doesn't wish to see the government intrude on a perfectly good herd instinct, so things remain as they stand. It would certainly appease crowd psychology to send not an ambulance, but a portable witch-boiler to the scene of every accident on a congested interstate. Would need a roomy cauldron to accommodate any rubberneckers who fail to keep one eye in the forward direction.
As for the research result, I don't comprehend this at all. On a two-lane highway if two cars decide to drive side by side, the only way to pass on the right is on the shoulder. There are many drivers out there who lapse into synchrony with the car beside them. We have neither a rule nor a vigilante squad to effectively police this, despite the fact that this is the obvious root cause for much of the avoidable congestion on any highway where I've experienced a daily commute.
Quite frankly I find the terms of this study stupid.
I guess that explains why you didn't stop to extract the grain of truth in your pique of ennui.
Bonhomie is greatly increased in collaborations where little harm results from momentary lapses or honest mistakes. Software security is not one of those things. Likewise, deep sea divers are notorious for having a short fuse over charging a cylinder with the wrong gas at the wrong pressure. That kind of thing.
Personally, I'd like to dig up some tape recordings of the choice language involved while setting the pilings that resulted in the Pont du Gard standing for 2000 years.
Reminds me of a joke I came across in a relatively lame joke book in someone's basement.
A women overhears some offensive language from two linemen working at the top of a telephone pole in front of her yard. She calls up the foreman at the power company and complains about the foul language. The supervisor contacts the two work men and demands an explanation. Joe explains: "Me and John were working up on this pole. He was up above me soldering some wires and dribbled some solder down the back of my shirt. So I yelled up 'Hey John, could you be a bit more careful up there'".
I've encountered a couple of civility-at-all-cost programmers in my day. The kind of guy who would ring you up from behind half a cement wall in downtown Mogadishu and go "hey, I didn't want to interrupt anyone but we're having a spot of trouble here... oh, the commander is taking lunch... no problem I'll call you back later". Needless to say, we didn't hit many deadlines.
Mihaly Csikszentmihalyi on TED talks about achieving flow states. He says a person enters a flow state in a situation when both challenge and skill are way above average (relative to the norms of that person), but the skill available is known to be sufficient to the challenge, even if the challenge is very high.
When I get deep into the top-right quadrant I don't have much patience with elementary errors, such a failing to free a pointer allocation, or a buffer overflow caused by misuse of the abysmal C string functions. If your standards are high, it's not that difficult to train yourself not to make this kind of mistake. In some ways I find the STL more difficult to achieve 100% correctness, since I can't keep Scott Meyer's oeuvre in my forebrain all at once.
How many buffer overflows are committed by people who are tying spongy ropes to their ankles and jumping off bridges? Is that figure elevation or altitude? Whatever. Let's just tie it and jump. That's how many people code. There's absolutely nothing about the terrible C string API more difficult than converting meters to feet, altitude to elevation, etc. The extra hoop is ominous and darned annoying, but it doesn't take a PhD to work it out correctly.
On some levels, there's not nearly enough brittleness among programmers. More brittleness among the community might have put the C string/buffer functions out of their misery twenty years ago.
When I'm deep in my flow quadrant, I truly resent the 1% of my brain allocated to safeguarding myself against abuse of the C language clstrfck() API. And I resent the other 500 paper cuts in the typical OS/programming language that confine my increasingly rare flights of inspiration to a depressingly paltry low-earth orbit. (And let's not kid ourselves that Java brings much to the table in the inspiration department. A flight of inspiration under a hermetic dome is not the life experience I'm seeking here.)
In "Man Without a Country" Vonnegut wills himself the following epitaph: "The only proof he needed for the existence of God was music".
He also says "What do men want? They want a lot of pals, and they wish people wouldn't be so mad at them."
If some day I arrive at a flow state where Art of the Fugue pours out of my keyboard, the very notion of palhood will lapse from my psyche, as will any concern I might otherwise feel for the emotional states of thin-skinned beer drinking leprechauns. Music first, pals second.
If you have to read all of your notes to search for something, then you are taking notes wrong. You should be able to flip through the pages and find what you are looking for pretty fast.
Yes, it's called a wiki, and it rhymes with quickie.
Even as a child in the 1970s, I was already a member of the digital age. My printed script was faster than my cursive script, and about 10x more legible. Unfortunately, my mind was still an order of magnitude faster than my hand, and letters from upcoming words (or sentences) would constantly spill out my fingers before whatever letter I had, fractions of second earlier, willed my fingers to transcribe at some dreadful sub-light speed.
I signed up for typing class at the first opportunity. My instructor's comment to my parents at the mid-term parent-teacher interview: the she-man types like a girl. I was doing 30 wpm on a manual Underwood six weeks into the course. Until then I was the kid whose cursive penmanship was so poor my teachers looked at me like I just crawled off the mitten-on-a-string short bus.
By the mid seventies (while still in elementary school) I had figured out that spelling only needed to be good enough to quickly and uniquely resolve to a single code-point (valid English word). I didn't know Shannon's theorem, but I knew enough to wait for it. I spelled homonyms correctly, and not much else. I knew that if Google could "suggest" the word 99 times out of a 100, the information content in my notes was sufficient to reconstruct the meaning. What I didn't know was that Google suggest was still two decades over the horizon.
When I enrolled in mathematics at university in the early 1980s, I essentially dropped out because it was not yet practical to transcribe math on a personal computer. I was more interested in programming anyway, but I wanted a solid background in information theory. TeX didn't run so well on an Osborne.
My hostility toward archival note-taking stayed with me until I set up my personal MediaWiki install three years ago. For the first time in my life, me and my note taking were on the same functional wavelength. It took about a year to generate enough content to be useful, and another year to work out the optimal structure.
Whenever I set out to learn a new skill, I shovel everything about the subject into my wiki like a bat out of hell. I'm gaining proficiency in new programming languages in half the time it used to take when I was younger and twice as sharp. Part of that is Google, part of that is the endless stream of breadcrumbs that tumble into belated structure.
Even better, I can learn a language in two weeks, finish a small program in another two weeks, not use the language again for another six months, and then return to the language and pick again at almost the same speed as I left off. It only takes about half a day for my wiki to regenerate the mental context I left behind, even if I've already forgotten the spelling of the assignment operator.
I sit here in slack-jawed amazement that anyone can function in a demanding capacity on a paper archival system. How could the human brain do that? Wonder of the world.
That was one of the jokes in the movie version of Russia House. They take it to the next level: the human anachronisms were using a chalk board. The movie doesn't show this, but I suspect their American counterparts were running most of their queries in COBOL and drowning in hard copy. Neither system at the time was what you'd call optimal.
To each his own, I guess. For me, 1975-2005 were the wilderness years. And then came my wiki. Somewhat perversely after all these years, my wiki gave me the first good reason to spell correctly (good reason = reason unrelated to the social grooming reflex). Last I looked, tsearch2 is not yet integrated with Google suggest, but even that might happen in due course.
When I need to brainstorm, I still sit down with a pen and a piece of paper.
photosynthesis being about 2%-4% in the real world
Photosynthesis in a colonial atmosphere, with no green shift.
Wikipedia:
Actual plants' photosynthetic efficiency varies with the frequency of the light being converted, light intensity, temperature and proportion of CO2 in the atmosphere, and can vary from 0.1% to 8%
Why is it the plants reflect the most energy intensive portion of the visible light spectrum? Maybe they can't handle the heat?
Regarding this thread as a whole, why is it that scepticism is so often given a free ride? I don't think 10% is out of range for an engineered solution. It's the engineering, not the claimed parameters, most deserving of scepticism.
There are lots of professional drivers who have to talk (pit crew, engineers, etc) while driving and they still can drive _competitively_ while doing so.
Don't strain yourself to bother contemplating the nature of cognition. For the most part, these professionals are talking *about* driving, so if anything, the object of focus is intensified rather than diminished.
There have been many studies about distraction in an office environment. I find it ten times as distracting when an office mate is discussion personal business on the phone while I'm coding than any technical conversation the person might be having. It's also immensely distracting if you're hearing half of a phone conversation where you're clueless about the agenda/emotional state of the unheard party. However, if he's talking about the intricacies of FPGA clock domains, no problem, even if I'm coding a web application.
Here's a choice to mull over next time you go under the knife. Have the surgeon explain the surgery to half a dozen medical students in the gallery as he works, or have him explain to the nurses the nature of ball spin on the putting green as pertains to his most recent golfing adventure.
Here's a great example of how poorly your average _professional_ multitasks on issues unrelated to work:
Larry: Excuse me, but what the hell's going on out here? Crash Davis: Well, Nuke's scared because his eyelids are jammed and his old man's here. We need a live... is it a live rooster? [Jose nods] Crash Davis: We need a live rooster to take the curse off Jose's glove and nobody seems to know what to get Millie or Jimmy for their wedding present.
Slashdot Mirror
...is it s diffuse and decentralized nature, a network of networks, not a single network. An organization or individual with the power to "fix" the internet would have the power to destroy it or lock it down.
Ahem, buddy. To what extent is "fixing" the internet going to amount to replacing cash with plastic? Cash is bad. It contains traces of cocaine, and a guy can get mugged for carrying it around.
Plastic is good. Corporations keep records of your transactions, which they provide free to any capable hacker. You can earn credit with plastic, and one day become entitled to upkeep a mortgage.
Even if a shirt and tie internet is someday invented and deployed to the masses, I'm hoping a portal remains to the bad side of town for those of us still clinging to coin.
The fly in the ointment is that sometimes the obvious won't peacefully coexist.
Arrow's impossibility theorem
With email, we want some semblance of anonymity, the ability to cold-call (write to someone you've never written to before, who hasn't written to you, either), yet no ability to churn poo in mass quantities.
This is surprisingly difficult to engineer. With voting systems, first past the post is known to have more flaws than average, yet we persist with it on the grounds, I suppose, that people deserve the fruits of their inability to emotionally comprehend a system that works.
Plurality voting system
For simplicity, every alternative system is lumped under the heading "the Italian model". This scares most people more than the mafia.
I was explaining to my sweetie the other day that math is all about spending hours to crack tiny grains of rice. Many of the people who struggle with math get caught up in the manipulations. The big ideas are tiny: positional number system, the digit zero, and challenge-response proof structure (aka calculus).
Let me explain that last one. Continuity was a tough nut to crack. All that infinity, how do you stop? It turns out, you don't actually show that the slope equals a value (that would be stopping, and stopping is verboten), you instead show that error bound can be made arbitrarily small (for any epsilon challenge, a delta response exists). It's a small idea, but essential, and rest of calculus follows directly.
From Arabic numerals
Fibonacci, a mathematician born in the Republic of Pisa who had studied in Bejaia (Bougie), Algeria, promoted the Indian numeral system in Europe with his book Liber Abaci, which was written in 1202
This late date never fails to stun me. So much for the obvious being obvious.
My own proposal, which I contemplated in idle moments some years back but never fully fleshed out, is that we add computational cost to the email syn packet (aka the "cold call"). email messages part of a back and forth exchange could linked cryptographically by any of the methods that prevent hostile packet insertion in e.g. ssh sessions. The details are difficult and exceed my attention span, but it has obviously been done.
The receiving mail host could inspect the incoming message, determine that the packet is a syn packet (not linked within an established exchange) and then decide to impose a computational cost on the sending machine: please factor this product of two large primes, then I will trust you enough to relay this message.
The essential feature is that this functions as challenge response: the imposed cost (product length) can simply scale as a function of how bad the spam problem becomes. If the amortized computational cost imposed exceeds the expected return, the economic incentive to push spam will vanish. It's far easier for the receiving host to generate the prime product than the sending machine to perform the factorization.
There are other asymmetrical math problems if this has some defect. It could equally be solving SETI frames or protein folding, if those have no fatal flaws. (A determination which is best left to specialists.) The size of the prime product challenge could rise and fall in a manner similar to TCP/IP congestion control: if more spam gets through, cost escalates, until the spamholes declare a loss and bugger off again.
The game theoretic proposition from the spammers perspective is this: the receiving hosts can band together to make pushing spam arbitrarily expensive.
Some legit mail (cold call subset) might entail a mail host devoting hours to a factorization challenge. Ideally this computation would be delegated back to the email origin. I'd happily let my system grind for day to authenticate one outb
The fundamental aspect of the Wikipedia concept was the fact that there wasn't a bureaucratic layer between your information and the world.
If you believed that, you believed that the purpose of Wikipedia was to provide a permanent sandbox to adolescence. Grow a pair, dude, a pair of lobes.
Perhaps you've noticed that the legal professional has this slander and liable thing. Now that the English Wikipedia is out of the boost phase, it only makes sense to put some checks and balances in place so that some juvenile pimple hiding behind a NAT the size of Singapore can't write harmful material about living persons on the basis of shoot first, ask questions later. Have you ever noticed how many times "George is a faggot" has been added to Wikipedia articles?
Banking grew up (and then regressed again), the wild west grew up, why shouldn't Wikipedia? If you ask me, it's pretty amazing they stuck to their wild west guns for as long as they did.
From Wikipedia:
Within fifty years of the Great Chicago Fire of 1871, the population had tripled to over 3 million.
Three million smells like change. Around about 1920, America experimented with prohibition.
Prohibition in the United States
OK, so Wikipedia won't get it right the first time. But shouldn't they try? Or should it be doomed to live in its mother's basement for all eternity so that you can personally identity with Wales, for the rest of his life, never admitting "our first conceit is no longer right".
Yeah that last one is a real warning message, pick a jar of peanuts and read it for your self. Anyone ignorant enough to not know that a jar of peanuts contains peanuts needs a lifetime treatment at the local electrical shock therapy center.
There are so many things wrong with this sentiment, from an engineering perspective, I hardly know where to begin. This flies in the face of almost everything we've learned about software development in over the last thirty years. But I'm posting belatedly, so I won't belabour the point.
1. common sense is not common
2. use case blindness: not everyone is standing in a brightly lit retail store, carefully contemplating the interior contents of each individual bottle, with a full slate of ordinary human senses
3. deniability in the edge case (we thought it was just as obvious with these sugar frosted chocolate coated peanut clusters as with the plain roast peanuts, so we didn't put the label on the bottle)
4. adding complexity to little advantage
I just love reading code with lots of condition logic to handle errors were no useful recovery is possible.
Simple: if it has peanuts, label it as containing peanuts
Complex: if it has peanuts, label it as containing peanuts, unless the product is obviously peanuts, and in case the interpretation of obvious ever needs to be litigated, here's fifty pages of criteria to legally define "obvious peanuthood".
Yeah, sure, let's reformulate the entire legal system so we can indulge in brief orgasms of self-satisfaction when the lazy idiot next to us does himself avoidable harm and wails in primal dismay. Let me tell you, I enjoy watching lazy idiots get their comeuppance. There's no greater joy in life than watching some stupid fool attempt to weave his way through heavy congestion with a series of phone-booth lane changes, on reflex and attitude and no anticipation, only to find himself mired at the center of dead stop while everyone he nipped around crawls slowly past in the next lane.
But I don't go around advocating catastrophically stupid engineering practise to enshrine comeuppance as a constitutional virtue.
After a long career in the tourism and non-for-profit sector, my brother became fond of the statement "the first rats off a sinking ship are the best swimmers". When this happens, the contrast ratio tends to increase, leaving a dysfunctional organization ever more dysfunctional. If the best and brightest of rural America are heading into the cities, the same applies. Averaged across America as a whole, nothing has changed, but you do have a slightly smaller, more woeful bucket.
Arthur Benjamin suggests we have the wrong focus in our math education. We should be teaching statistics, not calculus.
Arthur Benjamin's formula for changing math education
A typical person, after learning some calculus in high school, applies this skill precisely zero times in the rest of their adult life.
Statistics, however, is something we encounter on a daily basis, such as this article, with its potentially bamboozled statistical claim (did it properly account for a selective migration effect? Impossible to say from the story summary.)
People tend to have a relatively poor intuitive grasp on statistics, yet it impacts many of our daily decisions. Worse, even among those who have a reasonable grasp of statistics, few have a solid handle on robust statistics, which can be surprisingly subtle.
Bart Kosko (Edge.org is link challenged.)
Does everyone know the old joke that you can take the dumbest guy from a room of 50 pound foreheads and move him into a room of evolution deniers, and the average IQ in both rooms increases (really). This is just to point out that it matters how you draw the lines, as every corrupt politician knows instinctively. It doesn't mean that a single additional person voted in favour of the corrupt politico, yet moving the line can still result in victory.
On another front, urban migration is a fact of the modern world.
Stewart Brand on squatter cities
I have a friend who paddles at an elite level. As the club where she presently rows, the coach recently decided to split the top six athletes three each in an A and B boat. Two things happened: A) the race time averaged across the two crews improved, B) neither boat medalled. The six elite athletes were not impressed.
In Canada, we're inclined toward this kind of social experimentation. I deliberately live on the edge of a slightly seedy area of town, because I oppose further polarization (seedy by Canadian standards is no great hazard to life and limb).
In America, the balance is tipping so that one more year of life for some rich old white fart is procured at great expense, while a far cheaper intervention for an inner city black kid, who might live another twenty years with the benefit of treatment, is often neglected.
Here's an interesting question for debate: how does our widespread statistical ignorance bias social policy? If schools taught statistics instead of calculus, would the coefficient on power-law wealth distribution change one way or the other?
Yesterday, cleaning out the back room, I stumbled across an old photo album my partner had never seen. I told her how much I hated my grade 4 year. I didn't know it at the time, but I was forced to write with a pen for the first time during the Arab Oil Embargo. For me, it was the elementary school classroom pencil and eraser embargo.
The attractive Ms Pinder also wanted me to adopt a cursive script. I had a form of written dyslexia: letters from any word that might complete my sentence would jump the queue in the middle of whatever word I was laboriously spooling out. I couldn't slow myself down enough to use either a pencil or a pen, but at least with a pencil I had a fighting chance.
It happens I lived in a small town just a country road away from where a classic computer nerd had grown up, long before this meme was established. He was already off in the big world helping to invent APL, but I would visit his parents and play in his old bedroom with his amazing robot cars and stuff. His father used to tell me the story about the first day he left the house after purchasing an early edition black and white television. When he came home, his son had every piece of it, down to the last tube, laid out on the living room carpet. His father described him as having the messiest room he'd ever seen (I felt I had a shot to compete with that one) but that when it came to his wires, they were laid out like he was taking dictation from God. My Dad had hung out with him leading an after school church group in his high school graduation year and he had shown my dad, who also a bit of engineering school, some mod stuff about computers.
Since I was an easily bored child, one night when I was making trouble as an eight year old, my dad randomly started to show me stuff he had learned from D. He illustrated the binary number system with an egg carton and some black marbles. I got it right away. Afterwards, whenever I got a boring arithmetic problem in school, I would first change it to some other random base system, solve it, then change it back to decimal. OK, you wanted me to show my working, there it is. I was sending out major distress signals (hey, I'm a little bored here) but the stun wardens of the 1970s were unable to clue in.
Ink and cursive writing and obsession with spelling drove me to new heights of frustration. For a ten year old in 1973, I had a pretty forward view of computers. I knew the spell checker was coming, I just didn't know when, or in exactly what form. The book about the nature of algorithms my dad had checked out of the university library made this clear to me: if you could define a mechanical procedure, a computer would certainly do it. The only apparent road block was actually getting my hands on such a machine. Three years later I got my hand on a TI-30, it was the best I could manage, though I did also manage to get the 8008 data sheets from a military surplus mail order outfit. I just didn't want to mow the entire subdivision all summer in order to own one, and even if I did own one, it wasn't going to spell check my essays.
So there I am, surrounded my cultural artifacts from the future, with not much hands-on opportunity, speculating wistfully about exactly what I could get away with in school, given the future existence of these machines.
My attitude was this: if I've managed to get enough of the right letters out of my trick fingers that the teacher unambiguously knows what word I've intended, then I've done enough. What's she ragging on me about? I was ripe for a copy of Shannon's 1948 monograph "A Mathematical Theory of Communication", which I would have devoured as a young child. I was already thinking hard about numeric representation and English words as code points in a larger representation space. The fly in the ointment, I realized, is that the homonyms would continue to be a problem long after the computers arrived. So I worked hard to spell the homonyms correctly (and the plural
The truth about people's food opinions are like people's opinions on global warming - there really isn't enough empirical evidence to prove anything beyond any doubt.
I don't find that insightful.
There's no controversy that human activity is adding CO2 to the atmosphere, and no controversy I've ever run into that a sustained increase in CO2 level has a positive impact on terrestrial heat capture due to insolation. There's perhaps minor controversy over how much of the added CO2 remains in the atmosphere (and for how long), and a fair amount of controversy over how quickly the additional heating resulting from this leads to macroscopic climate change.
There's enormous controversy over whether the macroscopic changes will prove cataclysmic. Few scientists are formally trained in cataclysm, so I fail to understand why so many insist on weighing into this side of the debate (but it does seem to factor into granting application approvals). I do wish that any scientist who referenced Chernobyl in cataclysmic terms would recuse themselves from the debate over the likely consequences of global warming. Nor do I completely grasp how human removal of a lake in one place, and the human addition of a lake in another place are both automatically tagged as harm to the environment. In any case, there's relatively little controversy about the fact or mechanism of global warming, and a great deal of controversy over its rate, interpretation and human consequence.
Likewise, there's no controversy that human metabolism contains homeostatic feedback pathways, though there is a great raging controversy over the situations in which (and degree to which) the homeostatic mechanisms prevail over gluttony, fate, and life style.
That said, I've never seen it claimed that the elimination of soft drinks from a diet made anyone fatter. (Even artificial sweeteners have been linked to adverse changes in caloric balance.) Neither have I seen it claimed that a moderately brisk 40 minute walk three times a week dealt any healthy person's metabolism a dire consequence (fibromyalgia is no-one's idea of good health). I suspect there's a large group of people who if three times a week substituted a 30 minute walk in place of consuming a half litre of cola, would shed ten pounds rather casually.
Most other interventions seem to have a plenitude of YMMV asterisks. Metabolism is a complex system.
http://en.wikipedia.org/wiki/File:Neo-Laffer-Curve.svg
You can't afford to have a rational conversation with the public about the merits and demerits of your product, because your job is on the line if the sales curve slacks, so you've got to do your damnedest to sell the thing, no matter how awful it is.
Yes, but you sacrifice the goose (credibility with the public) as a byproduct of living to cuss another day. For me this is the central subject matter of this thread: that the race to the bottom has exceeded itself to such a degree that the press release has become just a fancy spam packet in the mind of the thinking public.
My definition of hell is entering a profession where my victories and my defeats verge on indistinguishable. Monetize any lingering shred of credibility, and once the credibility is gone, move on, like a prostitute whose desirability is inversely proportional to her career earnings.
From a game theory perspective, in a game of competent rivals this outcome should not exist: the cost of manufacturing the PR exceeds the value imparted to those who read it (at arms length with nose between fingers).
In the real world, there are sheep to be fleeced. A relatively small proportion of the public is essentially suckered by PR verbiage into providing just enough marginal profit to employ the hopeless schmuck who churned the thing out, so despite having reached bottom, it doesn't mercifully fade away.
I say this knowing that the conversion rate on penis enlargement is shockingly large. I'm just not capable of wrapping my mind around the vast septic system of human gullibility. We're a depressingly dumb animal once the primal pathways light up and the PR now functions as an unpleasant nasty-gram of human nature.
I'm stunned, I don't know why, to see people debating this as if this is the first time the issue has crossed their consciousness. News flash: this has been in the public water supply for at least two decades now. It's important, and if you haven't given it some thought long ago, you're not taking life seriously, you're just a woodpusher in the game theory of human realpolitik.
It boils down to a very simple premise: that entropy is a munition.
If you have some large chunk (say 100MB) of random bits in a file on your computer, there is no way to prove that there isn't some password that will decrypt this block of bits into meaningful information. Any chunk of information content which looks like pure entropy can be accused of harboring munitions, if you're trying to hit the preservation of society nerve, or child pornography, if you're trying to hit the righteousness of the flesh nerve (we all care about flesh). Steganography is the art of boiling a thin soup: very small amount of pure entropy hidden in a huge amount of tedious backdrop (say 200GB of licit pink matter).
If you have a large quantity of real physical entropy, there is of course no way to produce a password, and neither is there any way to prove that the entropy is real.
The authorities find this unbearable, so we are now deep into guilt by association. Caught hanging out with random bits, go directly to jail.
Any public discussion of the matter would conclude that our social concept of judicial fairness is incompatible with this new guilt by association model. What kind of society would declare entropy a munition? How would we all go about scrubbing anything that looks like entropy from our electronic records? It's not clear it is possible to comply with the implications of this, even if greater society drank the Orwellian Spook-Aid.
Hence the secrecy. If the spooks destroy 1000 innocent lives in the course of protecting society as we know it, it appears to be a cost we're going to have to bear.
The easy way to cease to think seriously about this is to invoke Stalinist escalation: that 1000 lives is soon 30 million lives.
Don't be so hasty. Sun Tsu beheaded one giggling princess to make every other princess march with the discipline of soldiers. For his needs, one was enough.
The credit industry doesn't work on principles much better than our agents of darkness. The suits have succeeded in labeling credential fraud as identity theft. Note the slight shift in blame here: it's not the design of VISA at fault (which could hardly be worse), it's your fault for offering up your digits in the first place (well, you can't use your VISA card without doing so, but why niggle?)
I hand pieces of information about myself to thousands of institutions. If the information is gathered and used against me, somehow I'm to blame, not the thousands of institutions who regard protecting the sensitive information they demanded from me as a cost center to be outsourced to India.
The great line in Brazil is "Confess quickly, or you'll jeopardize your credit rating."
Our credit system is nearly as arbitrary and secretive as this business of guilt by entropy. Innocent before proven guilty. The credit system is exempt from our normal social protections against slander. Any merchant can file a damaging untruth about me with little basis in fact, few avenues of complaint, and no ultimate liability whatsoever. The rating agencies will then spread this slander around and I can't prosecute them for spreading damaging falsehoods about me, even if I finally prove that the original merchant lied, and no sensible agency would persist in believing the original claim.
If we're not up in arms about the violation of our social norms concerning slander implicit to the credit industry, I don't harbour much hope that cottage outrage in this forum over incrimination by entropy is going to make any dent in the real world.
Stay tuned for the next exciting chapter, where encryption keys are extracte
The level of skill it implies, the time and the money, is out of reach of any ordinary user.
This would be an extraordinarily hard sentiment to formally define. How far back in history does one need to go to say the same about literacy? How far would one need to travel in the present world? Long before your definition reaches bedrock, it all becomes relative to a social construct.
Two of the great innovations of our number system (positional representation, and the digit zero) were incredible aids to making numeracy less "out of reach" for ever larger segments of the population.
The upward swing of the innovation cycle is making the dog walk in the first place, however badly. If the technology becomes pervasive, this is followed by the outward swing, making the technology ordinary. This point was also neglected by the post who suggested that open source needs to be better than closed source to overcome the adoption hurdle. But that only applies in the boost phase of the innovation cycle, not to ball point pens.
The desktop OS is halfway through the commoditization cycle already. For a broad audience, the browser already matters more. Decisions are increasingly driven not by what is best, but by what is most hassle free. Cost is not driving the bus. There are some pieces of software offered with no monetary cost I won't install because the software doesn't do enough for me to justify reading the license agreement, and the organization hasn't maintained a reputation where I'm willing to install the software on trust. (Ubuntu seems to be actively scouting the boundary with this latest move.)
Speaking of licenses, IANAL => literacy ain't worth much. The governing rules of society are out of reach to those governed. So that's what "out of reach" gets you as a debating tactic: absolutely nothing. Out of reach is not merely taken for granted, it's a governing principle.
The hassle factor is asserting control over the behaviour of our installed software has less to do with the learning the C language and more to do with byproducts of the software engineering life cycle. We're at the point in the innovation cycle where invention of the digit zero would be incredibly welcome.
The problem is that our software has an emotional IQ which in the animal kingdom would be dead square in the quadrant "too dumb to live".
A mother bear gets a bit testy about the space between herself and her bear cubs. I get a mite testy about a software installer shuffling around a system configuration that was carefully tweaked. To cite the most extreme example, I once lost nearly a month in a software development process because some stupid Microsoft JET accessory (don't ask) swapped a defective DLL in place of a DLL I had carefully chosen to be compatible with some other quirky POS (thanks, Microsoft).
In the animal kingdom, you take one look at momma bear, then you consider your survival odds if she gets testy about your next foot step. This the emotional IQ our software needs to develop. This is difficult at the present moment, because my software is blind: it doesn't know who the fuck I am. It doesn't know I have "bad mother fucker" tattooed on my wallet. (Note to the Sun Java installer: the next time you install the Yahoo toolbar because I forget to click off the Yahoo button--after clicking it off 13 out of 15 times already--I'm going to rip out your giblets and engage in a pagan ritual.)
This blindness is a byproduct of an inferior technology: our software engineering and release life cycle. As software engineers, we haven't yet figured out how to function effectively in a world where _every_ end user software install takes into account a personality profile of the software victim.
Testiness factors
1. maybe it's cool ... ...
2. what the heck is a DLL and why should I care?
5. don't mess with it if I don't understand it
10. mother bear
11. Godzilla
Having lived through
Stop reading, start thinking and feeling.
This makes a lot of sense, but given that Slashdot functions as a write-only medium, I doubt your advice will be widely heeded.
I used to think that literature was the human invention which freed us from our smallness of mind. Now that drama has been replaced by reality, and literature has been replaced by pop psych, could it be that we're better off turning the page on turning the page? The distilled essence of 10 billion lives lived is heavy going. Who has the time?
Half the commentary track on Criterion's Solaris (Tarkovsky) was about the female matrix's success at "becoming more human" and her human companion's inability/reluctance to love. What the hell does "more human" actually mean I wondered to myself. I guess that's what the movie was trying to tell me.
Then I watched The Last King of Scotland. Idi Amin struck me as frightfully human (though the Garrigan simulacrum did not).
I find this notion that there is some intrinsic state of "more human" emotional activation too Zen like for my taste. The presumption that the average person has a reliable in-built compass for pursuing this state strikes me as akin to "just keep driving around until you've memorized all the streets, then you'll know where you're going". There's no substitute for self-reliance--if you haven't lost the ball (or run out of gas) before enlightenment dawns.
A map of the human spirit might be a better starting point for those of us whose days are numbered. Some maps are better than others. Pop psych is the literature of the drowned.
Well, stop drowning and start swimming. I agree completely.
Guess which one crime dropped in?
Throw me a bone here. Are you counting suicide among the crime statistics?
Sorry to sound like a Microsoft fanboi or whatever, but Word is a more powerful tool than most give it credit for or bother to figure out, since a lot of its capability is kind of "hidden" to make it user friendly out of the box.
The number one skill for getting along with Word "out of the box" is learned helplessness. Let it randomly move everything around whenever it gets the whim to do so. If your fonts mysteriously change, just call it art.
I often help other proficient computer users unmangle their Word documents when they get a corpse bride formatting artifact. Sometimes I just have to use the mouse to delete something that the keyboard delete fails to fully remove, or use a left delete instead of a right delete. Sometimes I boil the corpse in Open Office, then return it to the scene of the crime, where it behaves fine ... until the next corpse bride resurrection.
Why should a simple word processor have a steep learning curve before one achieves any semblance of predictability? AmiPro made it possible to write a quick ten page document without encountering a single undead artifact. WordStar made that possible. Back in 1985. Why not Word?
The commentary track for Last King of Scotland notes that foisting your failures onto the shoulders of others is a symptom of megalomania.
Idi Amin: Yes, you are my advisor. You are the only one I can trust in here. You should have told me not to throw the Asians out, in the first place.
Nicholas Garrigan: I DID!
Idi Amin: But you did not persuade me, Nicholas. You did not persuade me!
Amin was a difficult man to work with. Word is a difficult program to work with. They both have a fondness for dancing bullets.
I'll happily add Word to my toolkit the day Microsoft publishes a suite of 10,000 regression tests enforcing predictability, not for the benefit of some obscure legacy macro package, but from the perspective of the end user writing simple documents.
Article contains the terms "ATM Machine" and "PIN Number". Read at your own risk.
Languages are shaped by cognitive cost. This is what Steven Pinker seems not to get. There _is_ an innate language instinct, it's just not what he thinks it is. What we all share is the ability to introspect the cognitive cost of figuring out "WTH is this dude trying to convey?"
One of the key insights on language is that Lempel-Ziv compression never transmits the compression dictionary. The dictionary is implied because the compression program and the decompression program share the same dictionary construction heuristic. This is a trick you can pull off only if the two sides of the channel share the same cognitive architecture. There are no shortage of examples out there of how fast communication breaks down when the parties begin with fundamentally different premises on how to structure the categories of thought.
Here's another fundamental question: what portion of the brain's cognitive activity is devoted to power management? For one thing, glucose is precious resource, and the brain is a chug-a-lug organ where it comes to glucose consumption. For another, the brain is costly to cool. From the real-time perspective (which governed 5.999 million years of human evolution), there's not much use firing up the abstract-noun chocolate factory when you need a survival response in under 100ms.
There's another truism here: fool me once, shame on you, fool me twice, shame on me. (Or, if you've spent forty years fouling your spark plugs, "fool me once, shame on -- shame on you. Fool me -- you can't get fooled again.")
When you get surprised by a lion, first you need to act, secondly, you need to record, to avert recurrence, after deferred reflection.
However, the brain does not record broad-spectrum. There's just too much. It's easy to build a PVR these days with 1TB of storage. I still haven't seen one where the tuner is replaced by a DC-to-daylight recording mode.
You can't defer deciding what to record for very long. So this is an obligatory cognitive function when your brain is already heavily loaded. At high enough stress levels, the recording function does shut down. Assessing and responding to cognitive burden is a mission-critical survival function. This is a key foundation for language learning.
A child doesn't need a special gene to discover the linguistic consequences of garden path sentence structures. "Oh damn, my mind when the wrong direction, and I wasted cognitive effort". Thus a child can self-infer a constraint on viable grammatical form, even if, in the manner of an LZW dictionary, the constraint is never explicitly conveyed from the language proficient to the language learner. The underlying assumption that makes this work in practise is that the architectural model of the child's brain resembles that of the rest of the population. This is 99% satisfied by being a member of the same species, without any weird genetic Pinkerisms.
As the language convention becomes more sophisticated, some parameters in the ambiguity resolution process become social constructs. Given a conflict between two heuristics, which takes priority? The important thing to realize about socially determined linguistic parameters is that they tend to vary across discourse settings. Experts have slightly different rules among themselves than apply in heterogeneous settings, where, e.g. half the people involved are ESL.
There was a thread here the other day on the consequences of a non-specialist treating guilt and liability as vaguely synonymous in exactly the wrong forum (wrists cuffed to ankles by the minions of RIAA).
A person incapable of pedanticism is not likely to succeed with either law or software. (This is one of the reasons why the IANAL meme on slashdot annoys the hell out of me: if the law is too complex to be successfully interpreted by a concentrated group of the weediest pedants on planet earth, just maybe perhaps the root c
That they are scrambling to censor is proof this is absolutly 100% legit.
Played much at the $1000 no-limit Hold'em table? Drowned any witches lately? Three weeks overdue on your first assignment from the first week of Introduction to Counter-Intelligence 101? Completely missed the confirm/deny coin-flip that has become the sole entertainment value of MLB in the juiced-ball era?
Obama has publicly opposed waterboading, but the fervently uptight spooks would still like know what the baddies are thinking. Why don't you apply for the job, you seem to possess special powers of seeing right through any kind of mask a person wraps around his/her motives.
But hey, run a spell checker on your application form, you don't want to be excluded for the wrong reason.
I had good initial success with ltspice under Wine. Not perfect: some dialog boxes don't focus input fields normally (but if you're persistent, you can get it to work).
No problem with the simulations. The command line simulation is more powerful than I at first suspected. I especially appreciated the ability to do multiple plots in parallel. Some basic logic elements were missing (multiple input muxes, IIRC). So there was a startup curve learning how to make my own.
Once annoyance was the plotting setup. The default background is black, which makes screen plots great, but sucks for printed output. You can change the background to white (you also have to make all the colours darker for contrast), then it works fine for printing, but your monitor display is nowhere near as good, and I never found a way to automate the switch. It needs a way to set up the monitor and printer palettes independently.
There is an RLGC simulator for coaxial transmission lines, but it's not fully general. According to my notes: "At least two of RLGC must be non-zero. If G is non-zero, L and C must be zero." I understand there is a way to transform coaxial parameters so you don't need G non-zero, but I haven't yet learned how to do this, so this was a bit annoying.
At once point an obscure error message from ltspice exactly matched the string in source code I found online for Spice 3f4. I believe the underlying simulation code is forked largely unchanged from 3f4.
I've done a lot of embedded programming, mostly in C/restricted C++. I often contribute to the schematic design at a block level, or for a key circuit element critical to the technology, but I'm far from an everyday EE guy. So you can either conclude that ltspice was easily adopted by a non-specialist (well, an elite non-specialist) or, if you have less success than I did, you can conclude that I just didn't push it very hard.
I get tense whenever I come across this kind of CYA posturing which tends to invoke more double standards than a house of mirrors.
More charges may be filed in HP case
Ousted Chairwoman Patricia Dunn and former ethics chief Kevin Hunsaker surrendered, were booked and released Thursday, a day after being charged -- along with three private investigators -- with felonies for their roles in HP's spying scandal.
Ethics officer dragged off in handcuffs, did it really hurt HP's business? What was it all about? It was about keeping their dirty laundry behind closed doors, no matter how appalling or borderline illegal.
I'm more of a KTB than a CYA. KTB = kill the bastards
CentOS is an example of "life happens". Interesting how many uninformed people who just caught their first whiff of this immediately chime in to explain that they should have tried harder to keep a lid on this, without even checking that maybe this box already had a tick mark.
The underlying assumption is that making an effort to keep a scandal behind closed doors will always work if the people involved warrant respect. Similar to the belief system of the HP executives. In their quest not to be damned by their knee-jerk shareholders for not trying hard enough to bung the leak, they went all the way to extra-legal.
I'm tremendously unimpressed by people who maintain their social standing by undermining the credibility of our public institutions.
Fundamentally, most support contracts are a tax on social insecurity. Am I the only person on slashdot with a vastly better track record at debugging failed software on my own steam than getting assistance over a telephone support hot-line?
I've had a few excellent external support experiences. On the other hand, in the time it takes to fight through the telephone support system of a typical company to a person who actually knows something I could have reversed engineered the antikythera.
Or in the case of HP, when one of their printer drivers made a friend's Windows 2000 machine unbootable, and their driver uninstaller refused to run under safe mode because it demanded a higher screen resolution that safe-mode VGA, and then my note about this on their support website (which took me 15 minutes to compose) returned "404 not found" after I pressed submit. So much for big, redundant iron. The dudes can't even keep their customer-support web server running (for customer_type==peon). Nothing screams "we care" like "404 not found".
That incident with HP cost me half a day of my life. End result of my support escalation: "404 corporation doesn't give a damn".
I once spent half a day debugging an obscure failure of EAccelerator in which tried dozens of Apache settings, ended up hand compiling, and then ended up running the whole thing under strace, finally filing a bug report which was incorporated upstream.
Which of those two experiences do you think I'm willing to repeat? Which of those experiences made me feel lower than an earth-worm? Which of those experiences made me feel like a useful member of the human race?
The most interesting property of the CYA reflex is how quickly and thoroughly it vaporises irrelevant considerations such as life quality.
The human brain was designed with a kill-switch on life quality. That's an amazing artifact from six million years of evolution under the parameters of nasty, brutish, and short. And still we marvel at the human capacity for genocide. Our CYA reflex is not the main player in this, but if you follow the wiring, I'm sure it's highly interconnected with the culpable wetware.
Government exists for one reason: To deprive individuals of the freedom of choice.
Quick, call the doctor, we've got a galloping case of polarization disorder.
Actually, governments exist because societies without governments are ungovernable.
Paul Collier on the "bottom billion"
Paul Collier argues that most nations which experience a natural resource windfall experience a few great years, then end up worse off than before.
The countries where this doesn't happen are countries like Canada, which has a complex system of checks and balances. The "instant democracies" which have the vote, but none of the other trappings of effective government, aren't so lucky.
One of the reasons this works in Canada is that we accept what the government is there to accomplish, even if we grumble as much as any other nation about the obvious inefficiency in how this transpires.
The growing problems in rich nations with health care delivery runs much deeper than government. Our medical technology has reached the state where the last five years of our life expectancy (in failing health) is capable of consuming most of the wealth generated in our working years.
Most people wish to live as long as possible. Without checks and balances, the logical outcome is that 100% of the GDP is ultimately devoted to life expectancy and medical intervention. We're far from running out of ways to make health care more expensive for incrementally less return.
The way it works in America, as I understand it, is that a lower-middle income wage-slave works their ass off for 40 years, saves up enough money for a modest retirement, soon breaks a hip or experiences some other common medical ailment of the golden years, and is then hustled back to work at Wal-Mart for another decade to pay it off.
Politics in America has always cultivated a large pool of docile and desperate workers. I don't know if this has its root in the slave trade or not. I do know that it remains easy in America to end up out in the cold.
Washington Mutual
How many of the workers spoke up about the clearly ludicrous lending practises? At risk of losing their health coverage? Or some form of power against peon litigation? Not likely.
Historically, a productive economy was seen as a fragile thing, as if the economic miracle might just as easily evaporate. Hence we structure society with many have-nots, on the ground that it keeps the haves hard at the grindstone. We tend to regard innovation as some kind of fleeting accident to be carefully protected. And we create complex systems of law and property around the flowers of human creativity, lest we enjoy ourselves too much, and wind up in coffins we haven't paid for.
It's not clear that the world works this way any more. Innovation might become a surfeit, rather than a paucity, if our mechanisms to protect innovation were less obstreperous. We could have millions of artists rather than thousands of celebrities. The economics of distribution have changed, but our system of thinking hasn't.
One thing is certain, though. We're living on a crowded planet, and we're going to need more cooperation rather than less. This will mean more government rather than less, especially if the people who oppose government fritter their energies banging around in their ideological tin shacks.
My number one wish for better government would be a less bewildering and capricious legal system. My particular nightmare is that the environmental calamity arrives on schedule (fat chance), the technology is there to do something about it, but the lawyers can't work out how to divide the royalties, so the technology is not fully deployed, or too late to matter.
I've lived in the area for a long time, and never heard a good story about the VSE (RIP 1999), it's remains, but not the lingering stench, since composted into the CDNX.
Wikipedia just provided me with a funny story about the VSE I didn't know, but find all too typical.
The history of the exchange's index provides a standard case example of large errors arising from seemingly innocuous floating point calculations. The index was initialized at 1000 and subsequently updated and truncated to three decimal places on each trade. The accumulated truncations led to an erroneous loss of around 20 points per day. Over the weekend of November 25-28 1983, the error was corrected, raising the value of the index from its Friday closing figure of 524.811 to 1098.892
Are these the same people who are proposing to solve the fusion problem with 220 synchronized penises? Good god, I hope not.
For the record, here's what $500m typically buys you in British Columbia.
Fast Ferry Scandal
Amazing, just eight hours ago, a local newspaper is reporting that these vessels have been flipped for $20m.
PacifiCat ferries resold overseas
The Washington Marine Group sold the three ferries the company bought from B.C. Ferries for $19.8 million, to luxury yacht builder Abu Dhabi Mar.
Four cents on the dollar. That beats the old VSE hands down. Vancouver has a world-class ethnic cuisine, has enjoyed some decent success in video games and film production, but has a terrible track record with anything that floats.
Ballard Power being one of the more buoyant exceptions. I just did a search on "Ballard Power profit" and was pleasantly surprised to get a hit.
Ballard Power reports modest profit in Q2
I suppose if General Slammer raises $500m to build the commercial scale reactor, they'll use our excellent BC shipyards to fabricate them. We're good. We can weld aluminum into structures less valuable than the original metal.
While I've never met a lumberjack I didn't like on a personal level, I have to say as voting collective, they're dumb as stumps. We inevitably get the government we deserve. Our big project always make work, but rarely make money.
In rural areas of BC, it's easy to spot the people with jobs at the local mill or the local mine: they've got more equity sitting in the driveway than in their shit-box house (4x4 trucks, boats, campers, skidoos, jet-skis, ATVs, etc.) Big nature, eh? You can't govern in this province without earning this vote.
We do have some nice mountains. Vancouver is planning a party to show this off. You might have heard of it. I think the plan is to lose a lot of money proving we're world class and shrewd at business.
The value of if/then for this mechanism is completely obvious. If the sun doesn't come up tomorrow, without fully taking this into account, the phase of the moon calculation will come out horribly wrong.
Computing was hard back in the day. Moses climbed Mount Sinai for some peace and quiet while he chiseled out his astrolabe's stack dump as part of a warranty claim, ate some strange mushrooms on the way up to relieve the tedium, and the rest is history. (One wrong chisel stroke and your claim is rejected completely, but that's another story.)
By the anally-uptight bifurcating decision path definition of computing, op amps aren't computers either, but it's truly amazing what a wizard can build with the little buggers.
From another perspective, the ancient Greeks deserve a major KISS bonus for making a mechanism even a fool could use. Instead, we complicate the definition of computer and take marks off.
I have a copy of Calendrical Calculations at hand. Opening to a random page, I transcribe this formula:
mean-moon = degrees .0013268 * c^2 + c^3 / 538841 - c^4 / 65194000)
(218.3164591 + 481267.88134236 * c -
Or something like that. No conditionals in sight. I've used a computer to write complicated little programs with no conditionals or flow control and been tremendously happy when the answer came out as expected.
On the other hand, the equation for Easter involves a conditional expression, which looks like this:
if shifted-epact = 0 or (shifted-epact = 1 and 10 (g-year mod 19)}
In defence of the ancient Greeks, Easter hadn't been invented yet. But there were omens. While the Greeks were perfecting their teeth, the Romans were hard at it sharpening the nail. Tooth against nail, the nail prevailed.
To further complicate matters, it's known that one-dimensional cellular automata can function as a general purpose computer. However, the mechanism that executes a one-dimensional cellular automata can be built with no use of conditional mechanisms more complex than trivial boolean equations.
So we have it that evaluating (218.3164591 + 481267.88134236 * c - .0013268 * c^2 + c^3 / 538841 - c^4 / 65194000) makes you a calculator, whereas evaluating the boolean expression (a && b || b && ~c) makes you a general purpose computer (or at least capable of executing one). Perhaps this debate is silly.
In 100 BC, I suspect the idea of making something before you decided its purpose was not in wide currency. "General purpose" hadn't been invented yet. If it was being done, it wasn't been done under a creed. The ancients made a lot of pottery, sometimes (I once read) in volumes where it was easier to throw it away than wash it, but I'm sure they didn't consider mass production the founding credo of new-age industrialism.
Is solving a Sudoku a calculation or a computation? I could set up a machine to enumerate the solution space with no conditional logic aside from a stop condition when the displayed digits match the givens and the other Sudoku constraints are also satisfied. The mechanism would hardly be more complicated than an Enigma machine, which had no in-built conditional logic.
If you claim that it's the mechanism not the problem which distinguishes calculation from computation, who is to say the Greeks didn't have a general purpose understanding of how to implement single purpose mechanisms? Part of the deal is miniaturization. The general purpose implementation is a lot less attractive if it's the size of an aircraft hanger. Which would you build?
Some days I hope our civilization soon collapses, and the only device found 2000 years from now to commemorate our technology is an iPod.
"It's cool, but it doesn't really do much. Not even a two qubit superposition."
"Wait just a furry nanosecond there, this sand-baking thing is a lot more sophisticated than it first appears. However crude, I think the ancients were on to something with broader applications."
Quite frankly I find your idea stupid.
Stupid on so many levels my mind was reeling until that much needed shot of oxygen.
What kind of anger bubble does a person have to live in to fail to notice that causality calculus never works in the real world? What kind of dank basement does a person need to live in to fail to notice the chastity debate? I think it was the Zimbardo TED talk where he observes that 70% of American teenage girls who vow chastity before marriage succumbed to temptation within a year, and the majority of these do so without use of birth control. Subtracting out the sinners, the program is a complete success.
If society wished to alleviate the enormous waste of time and fuel of people stuck in traffic jams on heavily congested highways, we wouldn't let the highways become nearly so congested in the first place. Congestion based use fees would encourage corporations to be a lot more flexible about working hours. The cost to their employees would be right there on the bottom line for the bean counters to whittle upon.
But the typical population doesn't wish to see the government intrude on a perfectly good herd instinct, so things remain as they stand. It would certainly appease crowd psychology to send not an ambulance, but a portable witch-boiler to the scene of every accident on a congested interstate. Would need a roomy cauldron to accommodate any rubberneckers who fail to keep one eye in the forward direction.
As for the research result, I don't comprehend this at all. On a two-lane highway if two cars decide to drive side by side, the only way to pass on the right is on the shoulder. There are many drivers out there who lapse into synchrony with the car beside them. We have neither a rule nor a vigilante squad to effectively police this, despite the fact that this is the obvious root cause for much of the avoidable congestion on any highway where I've experienced a daily commute.
Quite frankly I find the terms of this study stupid.
I'm tired of that silly myth.
I guess that explains why you didn't stop to extract the grain of truth in your pique of ennui.
Bonhomie is greatly increased in collaborations where little harm results from momentary lapses or honest mistakes. Software security is not one of those things. Likewise, deep sea divers are notorious for having a short fuse over charging a cylinder with the wrong gas at the wrong pressure. That kind of thing.
Personally, I'd like to dig up some tape recordings of the choice language involved while setting the pilings that resulted in the Pont du Gard standing for 2000 years.
Reminds me of a joke I came across in a relatively lame joke book in someone's basement.
A women overhears some offensive language from two linemen working at the top of a telephone pole in front of her yard. She calls up the foreman at the power company and complains about the foul language. The supervisor contacts the two work men and demands an explanation. Joe explains: "Me and John were working up on this pole. He was up above me soldering some wires and dribbled some solder down the back of my shirt. So I yelled up 'Hey John, could you be a bit more careful up there'".
I've encountered a couple of civility-at-all-cost programmers in my day. The kind of guy who would ring you up from behind half a cement wall in downtown Mogadishu and go "hey, I didn't want to interrupt anyone but we're having a spot of trouble here ... oh, the commander is taking lunch ... no problem I'll call you back later". Needless to say, we didn't hit many deadlines.
Mihaly Csikszentmihalyi on TED talks about achieving flow states. He says a person enters a flow state in a situation when both challenge and skill are way above average (relative to the norms of that person), but the skill available is known to be sufficient to the challenge, even if the challenge is very high.
When I get deep into the top-right quadrant I don't have much patience with elementary errors, such a failing to free a pointer allocation, or a buffer overflow caused by misuse of the abysmal C string functions. If your standards are high, it's not that difficult to train yourself not to make this kind of mistake. In some ways I find the STL more difficult to achieve 100% correctness, since I can't keep Scott Meyer's oeuvre in my forebrain all at once.
How many buffer overflows are committed by people who are tying spongy ropes to their ankles and jumping off bridges? Is that figure elevation or altitude? Whatever. Let's just tie it and jump. That's how many people code. There's absolutely nothing about the terrible C string API more difficult than converting meters to feet, altitude to elevation, etc. The extra hoop is ominous and darned annoying, but it doesn't take a PhD to work it out correctly.
On some levels, there's not nearly enough brittleness among programmers. More brittleness among the community might have put the C string/buffer functions out of their misery twenty years ago.
When I'm deep in my flow quadrant, I truly resent the 1% of my brain allocated to safeguarding myself against abuse of the C language clstrfck() API. And I resent the other 500 paper cuts in the typical OS/programming language that confine my increasingly rare flights of inspiration to a depressingly paltry low-earth orbit. (And let's not kid ourselves that Java brings much to the table in the inspiration department. A flight of inspiration under a hermetic dome is not the life experience I'm seeking here.)
In "Man Without a Country" Vonnegut wills himself the following epitaph: "The only proof he needed for the existence of God was music".
He also says "What do men want? They want a lot of pals, and they wish people wouldn't be so mad at them."
If some day I arrive at a flow state where Art of the Fugue pours out of my keyboard, the very notion of palhood will lapse from my psyche, as will any concern I might otherwise feel for the emotional states of thin-skinned beer drinking leprechauns. Music first, pals second.
If you have to read all of your notes to search for something, then you are taking notes wrong. You should be able to flip through the pages and find what you are looking for pretty fast.
Yes, it's called a wiki, and it rhymes with quickie.
Even as a child in the 1970s, I was already a member of the digital age. My printed script was faster than my cursive script, and about 10x more legible. Unfortunately, my mind was still an order of magnitude faster than my hand, and letters from upcoming words (or sentences) would constantly spill out my fingers before whatever letter I had, fractions of second earlier, willed my fingers to transcribe at some dreadful sub-light speed.
I signed up for typing class at the first opportunity. My instructor's comment to my parents at the mid-term parent-teacher interview: the she-man types like a girl. I was doing 30 wpm on a manual Underwood six weeks into the course. Until then I was the kid whose cursive penmanship was so poor my teachers looked at me like I just crawled off the mitten-on-a-string short bus.
By the mid seventies (while still in elementary school) I had figured out that spelling only needed to be good enough to quickly and uniquely resolve to a single code-point (valid English word). I didn't know Shannon's theorem, but I knew enough to wait for it. I spelled homonyms correctly, and not much else. I knew that if Google could "suggest" the word 99 times out of a 100, the information content in my notes was sufficient to reconstruct the meaning. What I didn't know was that Google suggest was still two decades over the horizon.
When I enrolled in mathematics at university in the early 1980s, I essentially dropped out because it was not yet practical to transcribe math on a personal computer. I was more interested in programming anyway, but I wanted a solid background in information theory. TeX didn't run so well on an Osborne.
My hostility toward archival note-taking stayed with me until I set up my personal MediaWiki install three years ago. For the first time in my life, me and my note taking were on the same functional wavelength. It took about a year to generate enough content to be useful, and another year to work out the optimal structure.
Whenever I set out to learn a new skill, I shovel everything about the subject into my wiki like a bat out of hell. I'm gaining proficiency in new programming languages in half the time it used to take when I was younger and twice as sharp. Part of that is Google, part of that is the endless stream of breadcrumbs that tumble into belated structure.
Even better, I can learn a language in two weeks, finish a small program in another two weeks, not use the language again for another six months, and then return to the language and pick again at almost the same speed as I left off. It only takes about half a day for my wiki to regenerate the mental context I left behind, even if I've already forgotten the spelling of the assignment operator.
I sit here in slack-jawed amazement that anyone can function in a demanding capacity on a paper archival system. How could the human brain do that? Wonder of the world.
That was one of the jokes in the movie version of Russia House. They take it to the next level: the human anachronisms were using a chalk board. The movie doesn't show this, but I suspect their American counterparts were running most of their queries in COBOL and drowning in hard copy. Neither system at the time was what you'd call optimal.
To each his own, I guess. For me, 1975-2005 were the wilderness years. And then came my wiki. Somewhat perversely after all these years, my wiki gave me the first good reason to spell correctly (good reason = reason unrelated to the social grooming reflex). Last I looked, tsearch2 is not yet integrated with Google suggest, but even that might happen in due course.
When I need to brainstorm, I still sit down with a pen and a piece of paper.
Half the time I'll sketch
Photosynthesis in a colonial atmosphere, with no green shift.
Wikipedia:
Why is it the plants reflect the most energy intensive portion of the visible light spectrum? Maybe they can't handle the heat?
Regarding this thread as a whole, why is it that scepticism is so often given a free ride? I don't think 10% is out of range for an engineered solution. It's the engineering, not the claimed parameters, most deserving of scepticism.
There are lots of professional drivers who have to talk (pit crew, engineers, etc) while driving and they still can drive _competitively_ while doing so.
Don't strain yourself to bother contemplating the nature of cognition. For the most part, these professionals are talking *about* driving, so if anything, the object of focus is intensified rather than diminished.
There have been many studies about distraction in an office environment. I find it ten times as distracting when an office mate is discussion personal business on the phone while I'm coding than any technical conversation the person might be having. It's also immensely distracting if you're hearing half of a phone conversation where you're clueless about the agenda/emotional state of the unheard party. However, if he's talking about the intricacies of FPGA clock domains, no problem, even if I'm coding a web application.
Here's a choice to mull over next time you go under the knife. Have the surgeon explain the surgery to half a dozen medical students in the gallery as he works, or have him explain to the nurses the nature of ball spin on the putting green as pertains to his most recent golfing adventure.
Here's a great example of how poorly your average _professional_ multitasks on issues unrelated to work:
Larry: Excuse me, but what the hell's going on out here? ... is it a live rooster? [Jose nods]
Crash Davis: Well, Nuke's scared because his eyelids are jammed and his old man's here. We need a live
Crash Davis: We need a live rooster to take the curse off Jose's glove and nobody seems to know what to get Millie or Jimmy for their wedding present.