Well, for an organization like NSA or the DoD to start thinking about what a working quantum computer would mean for things like cryptography is not after such a computer has been found to exist.
That would be... suboptimal.
And, after all, since nobody knows whether such a thing is a practical possibility, it might take only one person having a really clever insight to make it possible. If such a thing is a practical impossibility, well, even so studying how it might behave may still be quite useful.
under the cover of "I'm just joking", which is no different from the frat boy who hits one of the "nerds" in the face, knocking him down, and then claims that the nerd has no legitimate grounds to be angry, let alone fight back. After all, "I was just messing with you"
Ah, you've put your finger on it. You just haven't put a name to it, and there is one.
It is called "cowardice".
Spreading ideas that you don't want to be really associated with is cowardice, it doesn't matter how you choose to evade being tarred with whatever is on your brush. You can claim you're just passing on what "some people say", or you're "just joking", or maybe you just quietly run your keys along the paint of some car with a bumper sticker you disagree with when you think nobody is looking. If you don't want to be associated with what you believe in, then there's probably not much good to say about whatever that is.
It may be that some sheep manage to convince the other sheep they're really just slightly woolier sheepdogs, but in the end, when it is time to be lead to the slaughter, they'll line up obediently with the rest of them. They'll probably be thinking, "This can't possibly be happening for real," and as insane as that may be, they've have got a lot of practice at denial when the time comes.
It's almost certainly NOT a parallax issue. The distance of the screen image from the touch surface is far too small. It might be a calibration problem, but I doubt that too.
The position reading on a conventional touch screen is the result of a hundred or more samples of resistance measurements from points around the periphery, which are then used to calculate a "best-fit" x-y position. What is most likely is that the sample measurements are being thrown off. This could be done by electrical noise, bad connections, an intermittent touch, a touch that is too light, a touch that is too heavy, or touching with too large a finger surface, etc.
Almost certainly it is an interaction between system faults and the physical idiosyncracies of voters. This would be quite independent of whether the users are voting system or computer literate.
Once, in the early days of office computing, I worked for a small OEM that had a number of installations in accounting firms. We began to get reports of reliability problems, which we reported up to the manufacturer. We got back a tech note saying that nylon stockings could cause static discharges (apparently the engineers testing the devices didn't wear 'em), and the work around was to either (a) not wear nylons or pantyhose or (b) regularly spray your legs with anti-static fabric softener. So, we had all the secretaries that every hour or two went into the ladies room to spray their legs with Downy.
The point is, hardware glitches happen in short manufacturing run devices that are used in unpredictable real-world circumstances. I bet the number of those old time devices shipped was fairly comparable to the number of voting machines a modern manufacturer ships, only they were used in much more predictable environments. Even so, a mere two years provided me with a lifetime of oddball stories, like the firm that had their minicomputer plugged into an outlet controlled by a dimmer switch; or the firm that was a cross from the auto wrecking yard that had huge brownouts and spikes when the big car lifting magnet was turned on and off.
User idiosyncrasies are even worse than you think when it comes to touch screens.
The most common type of technology uses two plates of glass of glass with conductive coatings, placed closely with conductive surfaces facing each other. IIRC, it takes the resistance from various points around the perimeter of the front piece to various points about the perimeter of the bottom piece, and does a best fit calculation for estimating the center point of the press after a couple of hundred samples (see, there was a reason they made you take linear algebra).
Some users seem to have the touch for throwing that calculation off; maybe they bounce their touch a bit, or touch so lightly the contact is intermittent. It's possible that mechanical or electronic faults interact with different users' touch to act different behaviors. Maybe a machine could test fine for the person setting it up, but bad for some number of voters.
It isn't hard to imagine the machines operating differently as they heat up, or depending of power supply noise or ambient RFI. It's quite easy to imagine beta copies of a machine passing all manufacturer tests, then sample machines passing all acceptance tests by the buyer with flying colors, but then dozens of machines out of every hundred bought acting flaky in real world conditions. The problems in the machines could well defy every pre-setup test and post mortem tests the owners can think of.
I've worked on mobile technology since the days of the Newton. Touch screen technology is pretty good, but mainly because the devices they are in are very nearly disposable. Most of them drift out of calibration sooner or later, and quite a few of them develop problems staying in calibration, or downright flakiness. I wouldn't use the technology some place where it was critical to avoid problems, e.g. in an operating room or voting booth.
The voting booth is a particularly tough environment because you get a lot of different users, none of whom vote every day. You need a ridiculously high degree of predictability in that kind of application because you can't recover from glitches.
Well, what about Apple? Apple has succeeded with music distribution because they have DRM (which makes copyright holders willing to sell on iTunes), which ALSO effectively locks the music sold to iTunes and iPod. Apple wins, copyright holder wins, user can live with it.
Once copyright holders get confident with the model of using low price and convenience to sell their content, they can be persuaded to do without DRM, particularly for content that is past its earning prime. If a song is not selling CDs any more, any revenue from an authorized MP3 download is gravy.
J2ME brings up an important point. A small developer can't avoid being tied to a third party who controls the relationship between the user and the developer. The nature of the mobile market is you pretty much have to target particular devices and carriers, even though this means a change in strategic direction by the gatekeepers controlling your access to the user can crush your project overnight. It's happened to projects I've worked on.
J2ME has been strangled over the last two years by the move to smart phone. You either can't buy an implementation (even though they exist), or the implementations are priced to make you pass them up. I think this is because the carriers are interested in non-net-neutrality: they want to lock you into THEIR music service, THEIR picture services. Windows smart phones have to have their registry diddled to even allow apps not blessed by the carriers to be installed.
We're at a tricky point in mobile development. Android is promising, but there's no choice of devices or carriers. PhoneME, the open source j2ME project, is promising, but there is no prospect of a release any time soon. iPhone provides an opportunity to reach an early-adopter audience, but your fate is totally controlled by Apple's whims.
What I'd do is this: develop in Java for J2ME on Blackberries, but hide the fact from the vast majority of your code. Ban J2ME specific APIs except for a small number of files. This goes against the kind of models we are given when we learn a platform, models which showcase an API. That's fine for tutorials, but bad for production code. Review code and ask, if this had to run on Android, would it have to be changed? If it isn't in a package whose job is gluing your app to the underlying platform, refactor. Inversion of Control and AbstractFactory are your friends here.
The idea is that you want to be able to port to Android, or even AWT or Swing, by targeting a small number of files in well identified packages. If an API is not in both Android and J2ME, it get wrapped and put in the platform ghetto packages.
The reason I'd do it this way is that it gives me a reasonable current market, but also a maximal future market. When J2ME is available on the iPhone, I can sell on the iPhone. When PhoneME releases, I'm there. When more Android is available on more devices and carriers, supporting Android is not out of the question.
This advice assumes you are in this for the long term. If you just want to have fun, get your feet wet with mobile development and make a few bucks, the iPhone Apps store is not a bad scenario for you.
Re:MPG is an obsolete measurement
on
1000-mph Car Planned
·
· Score: 5, Insightful
Asking the point of engineering feats like this is like asking the point of sex being enjoyable. The point of sex being enjoyable is to encourage procreation. The point of engineering being enjoyable is to encourage creativity.
Of course, engineers like to see their creations at work, doing useful things, just like chefs love to see people eating. But speaking as an engineer who grew up in a restauranting family, you've got to be a little bit insane to go into either business. Nobody would become a chef unless they had a bizarre compulsion to cook. My brother went into that business, and you literally can't keep him away from the stove or the grill if there is cooking going on. The only reason he can sit still in a restaurant, I think, is professional interest in other aspects of the diner's experience, but even then he can't resist the temptation to host the meal, to buy drinks, to make suggestions for what to select from the menu. Some of his buddies have actually put full restaurant kitchens in their garages and spend their time off cooking.
When I visit my relatives, on the other hand, I find myself fixing their computer problems. I can't not fix their problems, even though I hate dealing with those kinds of messes. If cars were as easy to work on from general knowledge as they were forty years ago, I'd probably be fixing their cars too. I'm just addicted to the satisfaction of getting everything sorted out.
Huh? Why does NOT having a higher power deprive life of value? And if life has no value intrinsically, then why does a higher power "give" it any value at all?
Not acknowledging the existence of a higher power invalidates a number of arguments that life has value. That's not at all the same as depriving life of value.
If life has value, the invalidation of a large class of arguments to that effect only means that this fact must be demonstrated differently. On the other hand, if life has NO value, then contrary arguments that it HAS value are necessarily disprovable in some way. However, not all disproofs of such counter arguments are necessarily valid, they are just correct. A third possibility is that existence of value in life cannot be proved or disproved within the terms of discourse. In that case one can consistently take it as axiomatic one way or the other.
That's where things get interesting. Different moralities could be constructed around either alternative. For those who take the position that life has value, this value is, in a sense, rooted in the arguments to that effect. Refuting the arguments thus really does deprive life of its value.
On the other hand, for those who take it for granted that life has no value, the success of a counter argument to their views entails the prospect of their magnificently bleak psychological landscape being cluttered with fluffy white bunny and pink valentine hearts.
Those old Tektronix scopes were great -- the kind of tool you get when it is designed entirely by people who use it every day. I remember when scopes started to get smaller and certainly more capable and by every objective measure better. What they weren't is as much of a pleasure to use.
It's like the difference between an old analog and new digital alarm clock. The digital clock keeps better time and can do lots of tricks like change the alarm time for the weekend, but there's always a tiny bit of irritation involved in using the features.
Actually, American "Bud" is not in my opinion a bad beer. It's a boring beer. A "lawnmower" beer. I don't drink it because I don't have unlimited room in my diet for alcohol and calories, not because the taste is dreadful.
My main gripe with Bud is not its lightness, which is neither here nor there as far as I'm concerned. It's the near total lack of any distinctive aroma to its hops. Most Americans who now "hops" as an ingredient in Bud probably have no idea that hops are flowers.
The Czech beer on which Bud is flavored (which is reportedly impossible to duplicate without just the right water) is famous for its hop flavor. Pre-prohibition American beers derived from this style were lighter in body, already using maize or rice, but if old recipes are to be believed, they were probably more hoppy than modern "Bud". After Prohibition, I suspect that a thirsty and less educated American market was not ready for a strong hop flavor, and so the American style of lager was born.
In a way, the flavor of Bud, insipid though it may be, is a technical marvel. Given the alcohol content and light flavor, the tiniest problem with fermentation would be extremely noticeable. I've brewed Russian Imperial Stouts where you could probably pickle a dead skunk in the wort and nobody would notice. Presuming you could get it into the wort; the hydrometer just sat on top of the wort as if it were a bucket of wet cement.
I'm more into ciders and meads these days; they're harder to get. Good cider is especially hard to come by; I favor a dry cider brewed with champagne yeast. What got me off of brewing beer was Long Trail's IPA. I couldn't imagine any IPA I made coming out better, so why bother?
I agree. The solution to land use problems is planning.
This doesn't make the technology useless though. You jut have to imagine all the thing you could you could do with a large, self-powered, all terrain robotic platform. Among other things, it could be a self-propelled toolbox. Or if you filled the thing with batteries and let it sit charging most of the time, it could be a walking emergency power source.
I'm imagining that some day we might have giant robotic ants that build infrastructure projects like railways and bridges.
The one concept everybody should learn from economics 101 is not supply and demand, it's opportunity cost.
Well, they should learn supply and demand, and more to the point the idea of marginal costs and benefits.
In the coming depression, if there is one, people will be desperate for money, that is true. That is because the market value of their labor is low. So the opportunity cost of blogging is small, the number of people doing it large, and the market value of a blog small (as it is now) except when the blog stands out for some reason.
In a situation where labor is scarce, given full employment, time spent on amusing yourself on a blog is time not spent making lots of easy money. People aren't going to be looking for creative ways to draw attention to themselves either.
I inserted a flash card with the software, including the operating system, which was given to me by an officer of the electoral court minutes before the election started.
If you can corrupt a representative of the judge who is responsible for declaring if the vote is correct, does it matter if the box is electronic or paper?
Well, that's the wrong question. The question is how does the judge and his representative know that the flash card contains what it is purports to? How do they know that machine will do what they intend it to when the flash card is inserted? For that matter, how do they know the machine cannot be altered after the flash card is inserted?
These kinds of questions are command and fundamental to computer security. Phishing is just a technically crude version of a common paradigm: trick people into thinking they are doing one thing with a certain party when they are doing something else with a different party. Very sophisticated forms of the same attack can take in much more sophisticated people.
The judge, his representative and you can all, in complete honesty, think you are doing one thing, but if any part of the system, from the voting machines themselves to the software used to prepare the data for it, if any part of the system cannot be trusted then you don't literally don't know for sure what you are doing.
There are all kinds of way that electronic machines, if they can be trusted, can prevent specific kinds of fraud. But the most fundamental and important tool for dealing with fraud is detection. With a pure electronic system, there is no way to demonstrate that the results given reflect the votes people cast. Even if the difference is a matter of a technical glitch, it is not necessarily detectable.
Where bad totals created by bugs are detectable, the true totals can't be reconstructed, which changes elections. It is therefore possible to deliberately alter elections by introducing innocent looking technical faults that spoil the votes in some way that is statistically useful -- e.g., when the machines are not maintained, or when more than a certain number of votes are recorded..
Well, actually it did add something: the widgets on each tab can be listed, tree style, by clicking on the "+" next to the tab name, which shows why the tabs were moved to the left.
So, it's not a pointless exercise by any means. As you point out, if you don't have many widgets to manage, then it's a waste of space. The flip side is that the more widgets you have to manage, the better this layout is relative to the old one.
The change was a bit disorienting, for a few seconds at least until I realized that I was still on the portal rather than in some changed version of gmail.
With respect to resizing after you visit some kind of wide format unfriendly site -- well, maybe that's a problem, but I'm not sure it's iGoogle's. The format works fine for working with portal widgets, maybe not so good when you launch a site that is designed for 800x600 displays.
Ford and GM going belly up wouldn't magically cause them and their billions of assets, be it physical or intellectual to disappear.
What you'd have is companies breaking away from the main brand and being sold off. Ford would let Mazda go, GM Saab and so on.
What would die would be the GM/Ford brand names along with the pension plans and other UAW union benefits. Which frankly is a good thing for the US auto industry in the long run.
Well, actually the brand names would not necessarily die. To some degree, the process of dismemberment of these companies into their various assets has already begun, especially at Ford. Ford was a pioneer at mass production, at a vertically integrated manufacturing giant that did everything after the mining and smelting of the metals in its cars, up to franchising the retail selling of its cars. It also flirted in recent years with the idea of going the other way, towards becoming something like Apple, which only keeps enough of a toehold in manufacturing to keep its cutting edge design mentality grounded in reality.
Companies like Ford, GM (or AMD for that matter) are constantly struggling to solve an unsolvable puzzle: how to make more money than the next guy. So they're continually absorbing and spinning off companies, divisions, and functions. It is a financial optimization ethos, far removed from the craft ethos, epitomized by the potter who digs his own clay, grinds his own pigments, and generally has a hand in every possible aspect of production so as to have complete control of the product.
The very existence of Tesla motors is a testament to this. Anybody with money and an idea can create a car brand; they can buy design and components and practically everything else put assembling the pieces together from somebody else.
You don't need vertical integration to be a car company. But you do need credit.
If your product works, or at least appears to, and you have a sound plan for getting it to market, where it will be purchased, then SOMEONE will loan you the money.
Actually, that not being true could be one definition of a "credit crisis".
I will now perform for your amazement, my amazing one sentence and one paragraph explanations of "the credit crisis". Please note that these do not constitute economic advice nor advice on treasury policy, they are intended strictly for entertainment purposes only.
One sentence explanation:
Investors sometimes choose to ignore opportunities for what looks like relatively certain profit because what is rational economical behavior on the part of an individual or enterprise in times of relative certainty differs from what is rational in times of relative uncertainty.
One paragraph explanation:
Many important companies (banks of course, but also big companies) generate and spend huge amounts of cash. However, this is not in the form of things like dollar bills; even small businesses the "cash" on their ledgers represents a right to demand credit or actual physical dollars from a bank. Businesses don't want dollar bills: cash is credit. Large businesses who deal in stupendous flows of cash don't want that cash on their ledgers for more than a microsecond, because even with modest inflation that translates into millions of dollars lost. So they turn cash into securities: investments that return enough to offset inflation (or a bit more ideally), but can be turned with a great deal of certainty to a known amount of cash (which of course is usually just credit but could be sacks of dollars) at a specific future date. "A specific future date" is not the same as "right this very instant"; so businesses do have a small amount of actual cash -- well, credit at a bank really -- on their ledgers, but it really isn't enough for large cash volume companies like Home Depot or Coke and certainly not enough for banks. So they get credit and extend credit, and credit is the primary currency used in the economy. Both operation depend on the certainty of that promise of future cash: to secure credit, and ensure that upcoming obligations can be covered. When securities held by company A become uncertain, it cannot get credit from company B; nor can it extend credit to company C without possibly failing to meet its short term obligations (a condition known as bankruptcy). This has a ripple effect because C uses A's debt as a security, and B needs A's credit to meet its own short term obligations. So credit is frozen in limbo, and since credit is the primary currency used in the economy, cash stops moving, even to safe investments, because a safe investment is not safe for a company that needs the cash before the investment matures.
What really happened is that the LHC destroyed the universe, but then put it back almost the same way as it was before, at least close enough that nobody noticed it.
"But," I hear you ask, "how could the LHC put the universe back together from inside a universe that, speaking rather loosely, did not at the moment exist?"
Well, an equivalent (from an observational standpoint) way of looking at it is that the LHC created a nearly exact duplicate parallel universe at the same time it destroyed the one it was currently residing in. However, it would be totally pointless to create an exact duplicate, otherwise how would you know you actually did it? So it... left out a bit. Specifically the bit that was containing the liquid helium in LHC'.
Well, I can answer the reason why ACPI works "so well" on Windows. It "works so well" because the BIOS writers' job isn't done until the computer jumps through all the necessary ACPI hoops -- on the current versions of Windows.
This isn't quite the same as making a standards compliant implementation. It is making an implementation that does everything (or the important things) the standard says should be doable provided you are dealing with on Windows.
Secondly, there's no guarantee that whatever atrocities you commit will work on the next version of Windows. I've seen a number of ACPI DSDTs uncompiled, and they do a lot of version and OS dependent stuff.
You'd think the point of a standard would be to provide a uniform interface, that whatever operating system twiddles some lever gets the same result. However the result is dependent on what the ACPI implementation thinks your OS is. So this makes it possible, indeed very easy, to provide an ACPI implementation that is broken depending on which operating system it believes is running the show. Indeed, sometimes the fix for Linux based ACPI problems is to give a phony identification string to ACPI, to tell ACPI that your Linux kernel is really Windows XP.
Maybe there's some profound reason why the things ACPI are supposed to do in a situation have to be (a) tied a specific operating system and version yet (b) hide the details of the differences from the operating system. Or maybe it's just a crappy standard. The fact that it makes all kinds of wonderful things happen doesn't make it a good standard.
How on earth can the entire command staff of the Enterprise be that young?...
An adolescent captain just looks wrong...
It's just one of those experiences that everybody (or at least the fortunate ones) eventually have, like the first time you go to the doctor and he's younger than you are.
The first time I saw Star Trek TOS, it was in the original run on NBC. I was six years old at the time. At the time, 28 would have seemed quite old.
I'm not sure what the technical differences would be between being able to delete an app on the user's device "on a whim", and being able to delete an app on the user's device "for a good reason". A good reason might be to stop the spread of mal-ware. Think of what spyware could do to somebody, especially if it had access to the GPS.
Normally, I'm against vendors having this kind of control over users' platforms, but since you can presumably boot a patched version of the OS from microSD, or possibly even reflash your phone, it seems like if this becomes a problem for some users third parties can provide fixes.
I just edited the registry on my WM6 phone to allow unsigned apps to be installed. Thats pretty bad, in my opinion, making users buy apps that are blessed by the vendor, but it's fixable. Allowing a vendor to kill an app is dangerous, although there might be a good reason for this in some cases. However, I'm guessing it should be fixable.
I find it kinda hard to swallow that Sulu took an additional 25 years to rank captain. Being that he was in the same inner circle, as Captain Kirk and friends.
Depends on how dysfunctional the organization is.
My wife once interviewed a guy who was retiring from the military after years as a non-com. She'd ask, "do you have experience in such and so?" and he'd answer things like "no, not at all," or "a bit, but I wasn't very successful at it." She's a decent soul, and finally she had to stop him and remind him that he was applying for a job, not discussing his qualifications with an officer who had a complete and comprehensive paper trail on him and every other person being considered. Even so, I imagine within the military a lot depends on who you work for and with, especially in peacetime.
One problem every organization has is recognizing competent team players, ones who do their job so well that they don't get noticed, especially if they are around people who do get noticed for their feats of last ditch derring-do. I'd say that if you want an explanation for why Sulu doesn't become captain until he's almost retirement age, consider that Starfleet promoted Kirk to captain when he was thirty. Given what we know about Kirk, this indicates that frequent brushes with danger, survived by the skin of your teeth, is a fast track for promotion in the fleet.
Slashdot Mirror
Well, for an organization like NSA or the DoD to start thinking about what a working quantum computer would mean for things like cryptography is not after such a computer has been found to exist.
That would be ... suboptimal.
And, after all, since nobody knows whether such a thing is a practical possibility, it might take only one person having a really clever insight to make it possible. If such a thing is a practical impossibility, well, even so studying how it might behave may still be quite useful.
Ah, you've put your finger on it. You just haven't put a name to it, and there is one.
It is called "cowardice".
Spreading ideas that you don't want to be really associated with is cowardice, it doesn't matter how you choose to evade being tarred with whatever is on your brush. You can claim you're just passing on what "some people say", or you're "just joking", or maybe you just quietly run your keys along the paint of some car with a bumper sticker you disagree with when you think nobody is looking. If you don't want to be associated with what you believe in, then there's probably not much good to say about whatever that is.
It may be that some sheep manage to convince the other sheep they're really just slightly woolier sheepdogs, but in the end, when it is time to be lead to the slaughter, they'll line up obediently with the rest of them. They'll probably be thinking, "This can't possibly be happening for real," and as insane as that may be, they've have got a lot of practice at denial when the time comes.
It's almost certainly NOT a parallax issue. The distance of the screen image from the touch surface is far too small. It might be a calibration problem, but I doubt that too.
The position reading on a conventional touch screen is the result of a hundred or more samples of resistance measurements from points around the periphery, which are then used to calculate a "best-fit" x-y position. What is most likely is that the sample measurements are being thrown off. This could be done by electrical noise, bad connections, an intermittent touch, a touch that is too light, a touch that is too heavy, or touching with too large a finger surface, etc.
Almost certainly it is an interaction between system faults and the physical idiosyncracies of voters. This would be quite independent of whether the users are voting system or computer literate.
Once, in the early days of office computing, I worked for a small OEM that had a number of installations in accounting firms. We began to get reports of reliability problems, which we reported up to the manufacturer. We got back a tech note saying that nylon stockings could cause static discharges (apparently the engineers testing the devices didn't wear 'em), and the work around was to either (a) not wear nylons or pantyhose or (b) regularly spray your legs with anti-static fabric softener. So, we had all the secretaries that every hour or two went into the ladies room to spray their legs with Downy.
The point is, hardware glitches happen in short manufacturing run devices that are used in unpredictable real-world circumstances. I bet the number of those old time devices shipped was fairly comparable to the number of voting machines a modern manufacturer ships, only they were used in much more predictable environments. Even so, a mere two years provided me with a lifetime of oddball stories, like the firm that had their minicomputer plugged into an outlet controlled by a dimmer switch; or the firm that was a cross from the auto wrecking yard that had huge brownouts and spikes when the big car lifting magnet was turned on and off.
User idiosyncrasies are even worse than you think when it comes to touch screens.
The most common type of technology uses two plates of glass of glass with conductive coatings, placed closely with conductive surfaces facing each other. IIRC, it takes the resistance from various points around the perimeter of the front piece to various points about the perimeter of the bottom piece, and does a best fit calculation for estimating the center point of the press after a couple of hundred samples (see, there was a reason they made you take linear algebra).
Some users seem to have the touch for throwing that calculation off; maybe they bounce their touch a bit, or touch so lightly the contact is intermittent. It's possible that mechanical or electronic faults interact with different users' touch to act different behaviors. Maybe a machine could test fine for the person setting it up, but bad for some number of voters.
It isn't hard to imagine the machines operating differently as they heat up, or depending of power supply noise or ambient RFI. It's quite easy to imagine beta copies of a machine passing all manufacturer tests, then sample machines passing all acceptance tests by the buyer with flying colors, but then dozens of machines out of every hundred bought acting flaky in real world conditions. The problems in the machines could well defy every pre-setup test and post mortem tests the owners can think of.
I've worked on mobile technology since the days of the Newton. Touch screen technology is pretty good, but mainly because the devices they are in are very nearly disposable. Most of them drift out of calibration sooner or later, and quite a few of them develop problems staying in calibration, or downright flakiness. I wouldn't use the technology some place where it was critical to avoid problems, e.g. in an operating room or voting booth.
The voting booth is a particularly tough environment because you get a lot of different users, none of whom vote every day. You need a ridiculously high degree of predictability in that kind of application because you can't recover from glitches.
Well, what about Apple? Apple has succeeded with music distribution because they have DRM (which makes copyright holders willing to sell on iTunes), which ALSO effectively locks the music sold to iTunes and iPod. Apple wins, copyright holder wins, user can live with it.
Once copyright holders get confident with the model of using low price and convenience to sell their content, they can be persuaded to do without DRM, particularly for content that is past its earning prime. If a song is not selling CDs any more, any revenue from an authorized MP3 download is gravy.
J2ME brings up an important point. A small developer can't avoid being tied to a third party who controls the relationship between the user and the developer. The nature of the mobile market is you pretty much have to target particular devices and carriers, even though this means a change in strategic direction by the gatekeepers controlling your access to the user can crush your project overnight. It's happened to projects I've worked on.
J2ME has been strangled over the last two years by the move to smart phone. You either can't buy an implementation (even though they exist), or the implementations are priced to make you pass them up. I think this is because the carriers are interested in non-net-neutrality: they want to lock you into THEIR music service, THEIR picture services. Windows smart phones have to have their registry diddled to even allow apps not blessed by the carriers to be installed.
We're at a tricky point in mobile development. Android is promising, but there's no choice of devices or carriers. PhoneME, the open source j2ME project, is promising, but there is no prospect of a release any time soon. iPhone provides an opportunity to reach an early-adopter audience, but your fate is totally controlled by Apple's whims.
What I'd do is this: develop in Java for J2ME on Blackberries, but hide the fact from the vast majority of your code. Ban J2ME specific APIs except for a small number of files. This goes against the kind of models we are given when we learn a platform, models which showcase an API. That's fine for tutorials, but bad for production code. Review code and ask, if this had to run on Android, would it have to be changed? If it isn't in a package whose job is gluing your app to the underlying platform, refactor. Inversion of Control and AbstractFactory are your friends here.
The idea is that you want to be able to port to Android, or even AWT or Swing, by targeting a small number of files in well identified packages. If an API is not in both Android and J2ME, it get wrapped and put in the platform ghetto packages.
The reason I'd do it this way is that it gives me a reasonable current market, but also a maximal future market. When J2ME is available on the iPhone, I can sell on the iPhone. When PhoneME releases, I'm there. When more Android is available on more devices and carriers, supporting Android is not out of the question.
This advice assumes you are in this for the long term. If you just want to have fun, get your feet wet with mobile development and make a few bucks, the iPhone Apps store is not a bad scenario for you.
Asking the point of engineering feats like this is like asking the point of sex being enjoyable. The point of sex being enjoyable is to encourage procreation. The point of engineering being enjoyable is to encourage creativity.
Of course, engineers like to see their creations at work, doing useful things, just like chefs love to see people eating. But speaking as an engineer who grew up in a restauranting family, you've got to be a little bit insane to go into either business. Nobody would become a chef unless they had a bizarre compulsion to cook. My brother went into that business, and you literally can't keep him away from the stove or the grill if there is cooking going on. The only reason he can sit still in a restaurant, I think, is professional interest in other aspects of the diner's experience, but even then he can't resist the temptation to host the meal, to buy drinks, to make suggestions for what to select from the menu. Some of his buddies have actually put full restaurant kitchens in their garages and spend their time off cooking.
When I visit my relatives, on the other hand, I find myself fixing their computer problems. I can't not fix their problems, even though I hate dealing with those kinds of messes. If cars were as easy to work on from general knowledge as they were forty years ago, I'd probably be fixing their cars too. I'm just addicted to the satisfaction of getting everything sorted out.
Not acknowledging the existence of a higher power invalidates a number of arguments that life has value. That's not at all the same as depriving life of value.
If life has value, the invalidation of a large class of arguments to that effect only means that this fact must be demonstrated differently. On the other hand, if life has NO value, then contrary arguments that it HAS value are necessarily disprovable in some way. However, not all disproofs of such counter arguments are necessarily valid, they are just correct. A third possibility is that existence of value in life cannot be proved or disproved within the terms of discourse. In that case one can consistently take it as axiomatic one way or the other.
That's where things get interesting. Different moralities could be constructed around either alternative. For those who take the position that life has value, this value is, in a sense, rooted in the arguments to that effect. Refuting the arguments thus really does deprive life of its value.
On the other hand, for those who take it for granted that life has no value, the success of a counter argument to their views entails the prospect of their magnificently bleak psychological landscape being cluttered with fluffy white bunny and pink valentine hearts.
Those old Tektronix scopes were great -- the kind of tool you get when it is designed entirely by people who use it every day. I remember when scopes started to get smaller and certainly more capable and by every objective measure better. What they weren't is as much of a pleasure to use.
It's like the difference between an old analog and new digital alarm clock. The digital clock keeps better time and can do lots of tricks like change the alarm time for the weekend, but there's always a tiny bit of irritation involved in using the features.
It's not much of a research university if lab monkey can't spit in the eye of an IT drone any day of the week.
Yousa keepa dis up, and me no more lissa' you.
Actually, American "Bud" is not in my opinion a bad beer. It's a boring beer. A "lawnmower" beer. I don't drink it because I don't have unlimited room in my diet for alcohol and calories, not because the taste is dreadful.
My main gripe with Bud is not its lightness, which is neither here nor there as far as I'm concerned. It's the near total lack of any distinctive aroma to its hops. Most Americans who now "hops" as an ingredient in Bud probably have no idea that hops are flowers.
The Czech beer on which Bud is flavored (which is reportedly impossible to duplicate without just the right water) is famous for its hop flavor. Pre-prohibition American beers derived from this style were lighter in body, already using maize or rice, but if old recipes are to be believed, they were probably more hoppy than modern "Bud". After Prohibition, I suspect that a thirsty and less educated American market was not ready for a strong hop flavor, and so the American style of lager was born.
In a way, the flavor of Bud, insipid though it may be, is a technical marvel. Given the alcohol content and light flavor, the tiniest problem with fermentation would be extremely noticeable. I've brewed Russian Imperial Stouts where you could probably pickle a dead skunk in the wort and nobody would notice. Presuming you could get it into the wort; the hydrometer just sat on top of the wort as if it were a bucket of wet cement.
I'm more into ciders and meads these days; they're harder to get. Good cider is especially hard to come by; I favor a dry cider brewed with champagne yeast. What got me off of brewing beer was Long Trail's IPA. I couldn't imagine any IPA I made coming out better, so why bother?
I agree. The solution to land use problems is planning.
This doesn't make the technology useless though. You jut have to imagine all the thing you could you could do with a large, self-powered, all terrain robotic platform. Among other things, it could be a self-propelled toolbox. Or if you filled the thing with batteries and let it sit charging most of the time, it could be a walking emergency power source.
I'm imagining that some day we might have giant robotic ants that build infrastructure projects like railways and bridges.
This article summary is not very informative. The very least they could do is tell us which ten million lines of code Linux has surpassed.
I would mod the story "economically naive".
The one concept everybody should learn from economics 101 is not supply and demand, it's opportunity cost.
Well, they should learn supply and demand, and more to the point the idea of marginal costs and benefits.
In the coming depression, if there is one, people will be desperate for money, that is true. That is because the market value of their labor is low. So the opportunity cost of blogging is small, the number of people doing it large, and the market value of a blog small (as it is now) except when the blog stands out for some reason.
In a situation where labor is scarce, given full employment, time spent on amusing yourself on a blog is time not spent making lots of easy money. People aren't going to be looking for creative ways to draw attention to themselves either.
I inserted a flash card with the software, including the operating system, which was given to me by an officer of the electoral court minutes before the election started.
If you can corrupt a representative of the judge who is responsible for declaring if the vote is correct, does it matter if the box is electronic or paper?
Well, that's the wrong question. The question is how does the judge and his representative know that the flash card contains what it is purports to? How do they know that machine will do what they intend it to when the flash card is inserted? For that matter, how do they know the machine cannot be altered after the flash card is inserted?
These kinds of questions are command and fundamental to computer security. Phishing is just a technically crude version of a common paradigm: trick people into thinking they are doing one thing with a certain party when they are doing something else with a different party. Very sophisticated forms of the same attack can take in much more sophisticated people.
The judge, his representative and you can all, in complete honesty, think you are doing one thing, but if any part of the system, from the voting machines themselves to the software used to prepare the data for it, if any part of the system cannot be trusted then you don't literally don't know for sure what you are doing.
There are all kinds of way that electronic machines, if they can be trusted, can prevent specific kinds of fraud. But the most fundamental and important tool for dealing with fraud is detection. With a pure electronic system, there is no way to demonstrate that the results given reflect the votes people cast. Even if the difference is a matter of a technical glitch, it is not necessarily detectable.
Where bad totals created by bugs are detectable, the true totals can't be reconstructed, which changes elections. It is therefore possible to deliberately alter elections by introducing innocent looking technical faults that spoil the votes in some way that is statistically useful -- e.g., when the machines are not maintained, or when more than a certain number of votes are recorded..
Well, actually it did add something: the widgets on each tab can be listed, tree style, by clicking on the "+" next to the tab name, which shows why the tabs were moved to the left.
So, it's not a pointless exercise by any means. As you point out, if you don't have many widgets to manage, then it's a waste of space. The flip side is that the more widgets you have to manage, the better this layout is relative to the old one.
The change was a bit disorienting, for a few seconds at least until I realized that I was still on the portal rather than in some changed version of gmail.
With respect to resizing after you visit some kind of wide format unfriendly site -- well, maybe that's a problem, but I'm not sure it's iGoogle's. The format works fine for working with portal widgets, maybe not so good when you launch a site that is designed for 800x600 displays.
Ford and GM going belly up wouldn't magically cause them and their billions of assets, be it physical or intellectual to disappear.
What you'd have is companies breaking away from the main brand and being sold off. Ford would let Mazda go, GM Saab and so on.
What would die would be the GM/Ford brand names along with the pension plans and other UAW union benefits. Which frankly is a good thing for the US auto industry in the long run.
Well, actually the brand names would not necessarily die. To some degree, the process of dismemberment of these companies into their various assets has already begun, especially at Ford. Ford was a pioneer at mass production, at a vertically integrated manufacturing giant that did everything after the mining and smelting of the metals in its cars, up to franchising the retail selling of its cars. It also flirted in recent years with the idea of going the other way, towards becoming something like Apple, which only keeps enough of a toehold in manufacturing to keep its cutting edge design mentality grounded in reality.
Companies like Ford, GM (or AMD for that matter) are constantly struggling to solve an unsolvable puzzle: how to make more money than the next guy. So they're continually absorbing and spinning off companies, divisions, and functions. It is a financial optimization ethos, far removed from the craft ethos, epitomized by the potter who digs his own clay, grinds his own pigments, and generally has a hand in every possible aspect of production so as to have complete control of the product.
The very existence of Tesla motors is a testament to this. Anybody with money and an idea can create a car brand; they can buy design and components and practically everything else put assembling the pieces together from somebody else.
You don't need vertical integration to be a car company. But you do need credit.
If your product works, or at least appears to, and you have a sound plan for getting it to market, where it will be purchased, then SOMEONE will loan you the money.
Actually, that not being true could be one definition of a "credit crisis".
I will now perform for your amazement, my amazing one sentence and one paragraph explanations of "the credit crisis". Please note that these do not constitute economic advice nor advice on treasury policy, they are intended strictly for entertainment purposes only.
One sentence explanation:
Investors sometimes choose to ignore opportunities for what looks like relatively certain profit because what is rational economical behavior on the part of an individual or enterprise in times of relative certainty differs from what is rational in times of relative uncertainty.
One paragraph explanation:
Many important companies (banks of course, but also big companies) generate and spend huge amounts of cash. However, this is not in the form of things like dollar bills; even small businesses the "cash" on their ledgers represents a right to demand credit or actual physical dollars from a bank. Businesses don't want dollar bills: cash is credit. Large businesses who deal in stupendous flows of cash don't want that cash on their ledgers for more than a microsecond, because even with modest inflation that translates into millions of dollars lost. So they turn cash into securities: investments that return enough to offset inflation (or a bit more ideally), but can be turned with a great deal of certainty to a known amount of cash (which of course is usually just credit but could be sacks of dollars) at a specific future date. "A specific future date" is not the same as "right this very instant"; so businesses do have a small amount of actual cash -- well, credit at a bank really -- on their ledgers, but it really isn't enough for large cash volume companies like Home Depot or Coke and certainly not enough for banks. So they get credit and extend credit, and credit is the primary currency used in the economy. Both operation depend on the certainty of that promise of future cash: to secure credit, and ensure that upcoming obligations can be covered. When securities held by company A become uncertain, it cannot get credit from company B; nor can it extend credit to company C without possibly failing to meet its short term obligations (a condition known as bankruptcy). This has a ripple effect because C uses A's debt as a security, and B needs A's credit to meet its own short term obligations. So credit is frozen in limbo, and since credit is the primary currency used in the economy, cash stops moving, even to safe investments, because a safe investment is not safe for a company that needs the cash before the investment matures.
What really happened is that the LHC destroyed the universe, but then put it back almost the same way as it was before, at least close enough that nobody noticed it.
"But," I hear you ask, "how could the LHC put the universe back together from inside a universe that, speaking rather loosely, did not at the moment exist?"
Well, an equivalent (from an observational standpoint) way of looking at it is that the LHC created a nearly exact duplicate parallel universe at the same time it destroyed the one it was currently residing in. However, it would be totally pointless to create an exact duplicate, otherwise how would you know you actually did it? So it ... left out a bit. Specifically the bit that was containing the liquid helium in LHC'.
It really is one hell of a parlor trick.
Well, I can answer the reason why ACPI works "so well" on Windows. It "works so well" because the BIOS writers' job isn't done until the computer jumps through all the necessary ACPI hoops -- on the current versions of Windows.
This isn't quite the same as making a standards compliant implementation. It is making an implementation that does everything (or the important things) the standard says should be doable provided you are dealing with on Windows.
Secondly, there's no guarantee that whatever atrocities you commit will work on the next version of Windows. I've seen a number of ACPI DSDTs uncompiled, and they do a lot of version and OS dependent stuff.
You'd think the point of a standard would be to provide a uniform interface, that whatever operating system twiddles some lever gets the same result. However the result is dependent on what the ACPI implementation thinks your OS is. So this makes it possible, indeed very easy, to provide an ACPI implementation that is broken depending on which operating system it believes is running the show. Indeed, sometimes the fix for Linux based ACPI problems is to give a phony identification string to ACPI, to tell ACPI that your Linux kernel is really Windows XP.
Maybe there's some profound reason why the things ACPI are supposed to do in a situation have to be (a) tied a specific operating system and version yet (b) hide the details of the differences from the operating system. Or maybe it's just a crappy standard. The fact that it makes all kinds of wonderful things happen doesn't make it a good standard.
How on earth can the entire command staff of the Enterprise be that young? ...
An adolescent captain just looks wrong...
It's just one of those experiences that everybody (or at least the fortunate ones) eventually have, like the first time you go to the doctor and he's younger than you are.
The first time I saw Star Trek TOS, it was in the original run on NBC. I was six years old at the time. At the time, 28 would have seemed quite old.
A constitution class ship has neither an infinitely thin skin, nor is it 100% hollow, nor is it a perfect box. Your air volume calculation needs work.
On the other hand, you have to take into account the fact that warp drive is made possible by an arrangement of massless, frictionless pulleys.
I'm not sure what the technical differences would be between being able to delete an app on the user's device "on a whim", and being able to delete an app on the user's device "for a good reason". A good reason might be to stop the spread of mal-ware. Think of what spyware could do to somebody, especially if it had access to the GPS.
Normally, I'm against vendors having this kind of control over users' platforms, but since you can presumably boot a patched version of the OS from microSD, or possibly even reflash your phone, it seems like if this becomes a problem for some users third parties can provide fixes.
I just edited the registry on my WM6 phone to allow unsigned apps to be installed. Thats pretty bad, in my opinion, making users buy apps that are blessed by the vendor, but it's fixable. Allowing a vendor to kill an app is dangerous, although there might be a good reason for this in some cases. However, I'm guessing it should be fixable.
I find it kinda hard to swallow that Sulu took an additional 25 years to rank captain. Being that he was in the same inner circle, as Captain Kirk and friends.
Depends on how dysfunctional the organization is.
My wife once interviewed a guy who was retiring from the military after years as a non-com. She'd ask, "do you have experience in such and so?" and he'd answer things like "no, not at all," or "a bit, but I wasn't very successful at it." She's a decent soul, and finally she had to stop him and remind him that he was applying for a job, not discussing his qualifications with an officer who had a complete and comprehensive paper trail on him and every other person being considered. Even so, I imagine within the military a lot depends on who you work for and with, especially in peacetime.
One problem every organization has is recognizing competent team players, ones who do their job so well that they don't get noticed, especially if they are around people who do get noticed for their feats of last ditch derring-do. I'd say that if you want an explanation for why Sulu doesn't become captain until he's almost retirement age, consider that Starfleet promoted Kirk to captain when he was thirty. Given what we know about Kirk, this indicates that frequent brushes with danger, survived by the skin of your teeth, is a fast track for promotion in the fleet.