I meant there would be death in the streets of the USA, not Bangladesh.
I'm sorry, but this is the same argument used by the RIAA / MPAA that says, "oh no.. Something's changed in our prefectly controlled little world, and now we don't know if we'll even be alive in x-years.. Please Mr. Congressman, stop this change now before it's too late."
Change is a mixture of good and bad.. Deal. It's not the end of the world, though it might be the end of a cushy job.
It's a feedback loop. The rich get richer by paying the poor less. Eventually the poor can't afford what the rich are selling, and then the rich have to either pay them more or lower prices.
This is industrialization-era thinking with unskilled labor as your "working/criminal class". There is an entire economics of human resources, and that too has a supply/demand curve. What's happening with globalization is that the supply curve sky-rockets. BUT, what again, you're thinking in terms of unskilled labor, where a worker is a commodity (indistinguishable from any other worker). In reality, each worker carrys with him/her a set of attributes and excess baggage. Employers will cost-analyize the benifits against the costs. Thus a domestic worker will always have a certain edge over a foreign worker. Will that edge equate to tripple the salary? Who knows. But in the case of Asian software developers, I've heard lots of horror stories, and thus any management worth their salt should do serious research first (There are soooo many hidden costs as have been pointed out in this thread).
So in response to this fragment, the rich only get richer as they have some competative edge over the working class. A worker could become a doctor, lawyer, engineer or any other skilled labor that is in high demand. Kings and the Aristocracy only did so well, because they were militarily superior (back in the days of knight-hood). They only maintained their wealth due to complacency among the subjects. In modern terms, the only reason that the rich like MicroSoft continue to get richer is because of our beloved ledgislators that support such non-free business practices (involving patents, copyright, and the enforcing of other non-free-market laws).
My point in this section is that the only way to prevent a company from going global is to ledislate, and hopefully it can be seen that this is a non-optimal solution.
This stuff about the standard of living balancing out is a crock.
You're over-simplifying. If America blocks it's borders with respect to international workers and immigration, then you can create an artificial environment where the only wealth that will leave a country are via imports. But if you're as wealthy as America, why would you ever need imports? Because other countries can make things better/cheaper. How can this be? Because your artifical sand-box has inneficiencies with respect to the rest of the world (who has more competition / employees who demand lower wages). Thus money that would be lost due to paying salaries abroad is instead lost due to exports (and America's trade deficit is annually enough to bankrupt entire nations). By importing human resources, American companies become competative and lesson the trade deficit. Theoretically, domestic production is cheaper than imported goods, and thus stands to raise purchasing power of consumers (which helps the economy). In reality, American prices are so much higher, that at best they would merely be competative, thereby not appretiably affecting consumers.
There's another angle. The best-of-the-best from India, China, etc feaverishly apply for Visa's to the US. Once here, they have a good chance at competing against otherwise complacent Americans. The heavily math/science focused educational system in other countries puts them at a distinct advantage with Americans (all else being equal). Most border states heavily complain about immigration problems, and indeed most immagrants move into the cities, heavily burdening us. Many modern countries (like France and the US) are burdened with the issue of immegration. When you have more citizens, your wealth is diluted. Legislation has often been the only resort to counter this phenomena.
But why do we even have this problem? It's because there is such a massive disparity in the wealth of other nations that it's worth the massive barrier to entry that a country imposes. This is why Mexicans / Chinese / Indians come here. If they had work and a decent living style in their home country, there would be less incentive to migrate. Remember, we're talking about the creme-de-la-creme that come here. If they had jobs (that gave them greater purchasing power) back home, then they'd possibly stay, thus reducing the immigration factor. In the case of Mexico, it's more the cunning / daring ones that make it here (and sneak about until they can find a way of being nationalized). Large scale cheap labor should still have a good effect there for similar reasons.
Go look up the GDP of Bangladesh in the CIA World Factbook. Compare that to the GDP of the United States. Add together and divide by two. That's what you get with globalism, sir.
You're suggesting (100 + 1) / 2 = 50 all around, but you're picking numbers out of the air. Because of the infrastructure, only a limited amount of physical labor can be transplanted to other regions. This is only a small percentage of the total GDP. Unemployment does reduce one's GDP, but it also has an effect of reducing salaries (think great depression). Reduced salaries avoids the inital issue at hand. These factors oppose one another so it's not as simplistic as you suggest. Further, wealtheir internationals have greater purchasing power, and thus increase export demand. If the organization is still head-quartered in the US, that means US tax revenue on work barely done in other nations.
In short, this is a natural process at work, that's gone on since the dawn of time... Darwinistic competition over scarce resources.. And only the most efficient will survive. So either give up a cushy job, or legislate and watch as your trade deficit continues to grow uncontrollably until there is little net asset left in the US. (Tarrifs don't work by they way, they're just a form of placibo to make out-raged citizens temporarily happy).
Just a nitpick (since I agree with your general point).
However, in the fork-per-request model, there's still a central process that touches each request and is vulnerable to change.
In pre-forking, process A sets up its environment which includes the service port, and a shared memory segment (often just a a simple memory mapped file). Then it creates a bunch of worker forks. At this point, the master could simply do a ps once every second and see if it's children have died, but usually the shared memory is used to communicate activity to the master.
Since all the forks maintain the file-handles, anyone of them can accept a job request. Thus they can be truely independent, and your issue of a corruptable monitor doesn't apply. (Except in the case that crashed workers need to be replaced by the monitor; but I feel that you were saying something different).
It's true that sharing information between the workers and the monitor has a potential for corruption, but I'm not seeing how mainframes avoid this (except if you're saying that the functionality normally handled by programmed IPC is relegated to other tasks).
I detest teachers like you who can't keep their personal politics out of the classroom.
I'd say that this is a big gray area. I think extreme tolerance is exactly what is hurting American society; tolerance assumes that deviance will turn out for the best; and that intervention will cause more trouble that it will rectify. Thankfully we're not so bad that we indifferently allow children to have sex on school grounds, but there has to be a balance. I think generalizations of what can and can't be done (such as the introduction of personal opinions (including politics)) are hurtful, because they legally restrain practically exceptible methods of helping students.
My problem is that you _have_ to introduce personal creativity in the classroom, but you are refrained from most natural methods (such as discipline, or other method of steering the student, when you know the allowed mechanisms will not work). It's taking away the straw and forcing them to make bricks.
The correct approach (in my opinion) is to have a proper balance and an allowable dynamic. Have peer review as an intricate element to operation, and allow extreme contempts by peers to be reconsiled at higher and higher levels as necessary. One rule does not fit all.
That a political science teacher favors one president (or candidate) and spends an assymetric amount of time on them would be considered introducing personal politics. But the reality is that you can't spend equal time on all subjects, and the teacher's personalization is essential in making a class worthwhile to a marginally interested student.
A technology class, likewise, will have to (at times) demonstrate failed works, and the causes for failures/successes. (Much like a political debate class must anaylize social mores) So long as a teacher does not require that a student recite a hate-mantra, I find it acceptible for a teacher to demonstrate their personal analysis of a subjective situation. If a particular community finds fault (which should come about through proper peer review), then that's another matter.
Schools do have people to admin their computers/networks, right?
At my high school, the CIS guy was WAY over-worked. In fact, we mostly had to teach ourselves CIS since he was always on call somewhere in the school (which made the slow lost and the swift excelent par-none).
Given budget cuts, the idea of adding additional staff probably doesn't sit well. And given the over-working of network staff, I would find it hard to impose additional burden apon them.
Apache/MySQL
Hehe.. Still don't think MySQL is up to the task of data-robustness (no roll-backs among other things). Given that grade-reports are more valuable than slash-comments, I'd argue for at least something like postgres (or better, even sybase/oracle on Linux).
The biggest problem is that you _have_ to garuntee data integrity and security. You have hackers in the high schools, and now you're giving them network access to machines that they probably play with at home. You're basically betting that your developers are more sophisticated than the students (or students non-school friends). At least with the proprietary systems, the likelihood of students even knowing what's under the hood was slim.
By the way, by network access, I'm referring to the fact that ethernet drops are generally available throughout the building (laptops can easily interface). Further, using Apache implies external access. (A developer _could_ forgo network access to the server, but this again implies that school sysadmins are sufficiently experienced). -Michael
Effectively, a particle in quantum physics is a cohesive bundle of energy. We measure the mass of that energy in "electron-volts" (eV), which you can think of as a electron-level volt-meter. It's similar to measuring the voltage of a battery; we can't directly see how much charge is in a battery, but we can see how hard it pushes / pulls a test charge. Likewise, we can't see how big a proton or electron or up-quark is, but we can see how it affects other particilars of similar size (e.g. an electron as a reference point). Due to the massive deviances in particular masses, it's hard to know for sure if a photon is truely massless (even though it carry's energy). post-modern quantum physics speculates that photons, neutrino's, and even gravitons have mass. (Yes, this does imply that gravity has a weight of it's own. More precisely, the emision of the force of gravity adds weight to the space between two particles.)
The substance of the particle is subject to debate. String theorists believe (if I'm not mistaken), that all particles are made of strings of something (which we'll never know), and that those strings wrap around space (which we also can't know it's consistency)- warping it and being stretched by it.
Another point of view is that of Ether, which we tend to hold on to, since quntum physics is so similar to our percieved world that it would be a shame that such patterns could not be known to persist at different scales. One theory that I like is called Aethero-kinematics. It's based on the idea that tiny hard balls (perfectly elastic, like steel) bounce about in different patterns (mostly vortexs, like in a drain). All energy is in the form of the kinetic energy present from these bouncing balls. The cohesion allows for quantum particles. The augmentation / contraction of mass (via Einsteins special relativity) is explained away the same as Mach-theory (where an the air-resistance increases exponentially as you exceed the speed of sound). The "speed of light" is merely the average velocity of the balls. The explained reason why we can't perceive relative motion against the ether of space is that earth is not moving with respect to the ether about it; nothing does. Motion is only ever a small fraction of a difference in speed from it's surrounding ether. Lastly, the concept of experimentally determined transverse nature of light is nicely explained away in Aethero-kinematics in common sence ways. (having to do with the probability distribution of collisions of particles in an ideal gass)
Modern quantum physics simply ignores the what's and hows of particles, and simply says they exist with certained measured properties.. That's it, that's all, that's ugly. Because of this, I tend to look at models like the above (so long as they fit the experimental data) as a way of putting my mind at ease. The problem is that until the theory's demonstrate validity, we can't take the analogies they present (ideal gas, or strings) too far in extrapolation / interpolation.
As for waves (also questioned in this thread): a wave is a regular periodic fluxuation. Longitudal waves are like a wripple in a violin string or cresting waves on the ocean. If you just look at a single water molecule, however, you'll see that it doesn't move forward, but instead up and down (just like a boat). You could also look at a police-car flashing light. The color of the light slowly fluxuates from red to blue and back again in a definite period. If you took a cardboard box and punched a hole through it, you'd see on a wall the color fluxuation. If you look more closely, the fluxuation is merely caused by a rotation of two light bulbs. Photonic transverse waves are the fluxuation of the state of the photon from electric to magnetic (hense the phrase, electro-magnetic). An electron sitting still has only an electric field (which applies force to other adjacent electric objects (pretty much anything but a neutron; and even it, if you break it down into quarks). When an electron moves in a circle, it applies a strange perpendicular force which only affects other spining electrons. You can understand that it's different than charge because two electrons are attracted to each other when they counter-rotate (or rotate, I forget which). It turns out that rotation has nothing to do with it; it's the motion of the electrons (but the math gets harder). So here are two completely independent characteristics of a charged particle. As it turns out the transmission of photons accounts for both activities, so the photon is both a messenger particle for magnetic fields and charged-fields (electric-fields). Since a photon must always travel at the speed of light (relative to it's medium), it should be apparent that it works within a magnetic context (e.g. charge in motion). It seems that the photon fluxuates between the two in a sinusoidal pattern with respect to time (independent of it's physical motion). The "frequency" of the photon is the speed at which it oscilates a full transition between electric and magnetic. Such a periodic transverse wave-pattern has many astonishing properties. Most notibly that the same beam of photons when reflecting back apon itself can have interference patterns; namely that the waves can cancel each other out (or amplify one another). The best example of this is to take a beam of monochromatic polarized light and send it through a cardboard box with two slits on it. On the other side of the box, you should see a periodic pattern of light and dark spots.
I'm not a physisist, but I am an electrical engineer, so I have more than a lay understanding of the principles.
if RF lighting exists at "[a] frequency that humans can't even see" then how [snip] can it illuminate anything?
Color is directly related to the frequency of the wave of a photon of light. (we'll ignore for now how a photon can be both a partical and a wave). Visible light is a tiny frequency range; IR, microwave, and radio are below it, and ultra-violet, gamma and X-ray are above it.
Additionally, the energy of a photon is directly proportional to its frequency. E = hf (h is a constant of proportionality; planks constant divided by 2*pi).
When photons hit atoms, they're usually obsorbed. The electrons only obsorb a quantum amount of energy, and give off any remainder. Eventually the electrons re-emit this stored energy. The neat part is that all the emitted photons will have the exact same energy. Different molecules will have different discrete values of energy, and thus give off different colors.
It's possible that multilple low-energy photons can hit the same atom, and cause it to raise it's energy to different levels (each jump will require a different amount of energy, and thus give off different colored photons). It's highly likely that the energy would quickly be given off, but it's possible that the energy will continue to build until when it does give off it's energy, you've achieved a significantly higher frequency photon.
While I haven't looked at Fusion's method, it's completely plausible to accumulate lower energy photons as I've described above.
Sorry, but Linux is not anywhere close to being as good for the average dumb computer user as windows. People will always like it because of its simplicity, which is its weakness for those who "get" technology.
Yeah!! I can think of an argument to this. First of all, the initial comparison was between DOS. Tell me that DOS is easier to use than Linux.
Now look at modern RH Linux (and it's brethren). Installation is easier today than DOS HW/SW was; assuming you have the proper hardware (which has always been a problem). If all that "grandma" needs is a button that connects them to AOL / MSN and a subsequent button for email / word-processing / web-browsing. You can _easily_ do that with existing GUI capability (once AOL, MSN get on the Linux platform).
It should be trivial to have Linux boot-up without even a login prompt.
The ONLY thing that Linux can't completely do is provide usibility transparency to the MS platform (sadly they've even been trying; which I find pathetic). But that's the nature of competition. We don't expect that a saturn coupe and a lexus SUV are going to drive the same way, why should we "expect" that software all run the same way (while we should always expect that there is some level of intutition for each manner).
As for proof, I've already gotten several of my past girlfriends to "use" linux while at my house. They know which icons to hit, and are largely familiar with netscape. Aside from running maple, and games that are targeted to the windows platform, much of the office productivity suite are becommign viable. The complexities are all being hidden (I still refuse to use nautilus; I need to find a way of disabling it to save memory). The geek-argument is quickly eroding.
Competition doesn't normally involve KEEPING others from competing
Don't know if I agree.. A capitalist is merely concerned with maximizing profit. From a purely mathematical / logical point of view, any and everything is a possible course of action.. The only constraints are the physical world, which includes artificial policing. e.g., from a business point of view, what's the most I can "get away with". Bribe, and cut-backs have long been a business practice (long before the evil yet legitimate practices of MS). Hell, just like at any city where the Maphia has influence. Capitalism is really survival of the fittest, and that often means those that are able to think outside the box will win. (note that it's not the same as acting outside the box).
So in summary, competition involves everything including murder (just think about how capitalism works when you're stuck on a mountain side with only a small ration of food and 10 starving people). Jail is just as financially bad as bankrupsy, so illicit activity followed by conviction isn't any more of a risk to investors than betting the farm on some last-ditch scheme before being squashed by your competition.
Does this justify it? Course not.. That's why we have the publicly funded SEC, Trade commission, police, etc.
Just a small nit-pick here.. Capitalism is based on rational decisions. Those decisions are based on supposedly practical values. The most practical value is self preservation. Thus an essential element in capitalism is that one will do what is best for themself. If any decision means that throughout the course of one's life, they will be better off doing something that, but that action hurts society as a whole, they will definately choose the selfish direction. It is irrational (non-optimal decision making) to do otherwise.. It is only when hurting society will eventually hurt them in the long run (reigning anarchy where they will most likely die at a young age), that an individual must choose non-selfishly.
Greed is a hard word to quantify. You could say it's an obsessive level of consumption at the expense of others. But it's hard to differentiate this from rational action. I'd rather characterize it as a compulsory (obsessive) behavior, where optimal decision (e.g. rationalism) are not made. To promote a consumption (including mere acrument of assets) whose' diminishing marginal returns are not considered. In other words, since each additional dollar should have less value than preceeding ones, there should be a point where other elements of one's personal life are more valuably persued that that of the next dollar. Obsession is where we persue this devalued dollar in spite of other needs.
The problem is that while a human being is ill suited to make use of that last made dollar, a company isn't. In fact, the opposite is true. Each additional dollar made is a vehicle for persuite of even more productivity. It therefore has an exponential effect. Remember, so long as hurting others does not hurt yourself in the long-run (including possibly your decendents), then it is still rational to act selfishly. Thus a monopolistic company that squashes competition and locks every man-woman-and-child into a sort of tax (due to annually required upgrades), then enforces license compliance, then the raising of prices, hurts the hell out of external cultures and organizations (countries even), BUT still benefits in the end. Even if they are boycotted, affected with ledgislation, they will still have acquired more total present-day-dollars, then if they'd constrained themselves.
This is efficient and rational capitalism.
The problem is that MS should not exist as it does today according to classical capitalistic theory. The ONLY reason MS exists today as a monopoly is because of copyright law. A "tool" emposed to artifically help capitalism. MS does not produce windows more efficiently than a potential competitor. MS does not write database code that is better than competitors (nor even more cheaply). It merely has a strangle-hold thanks to it's artifically enforced licences and intellectual property. Non-disclosure-agreements, and non-terminating copyright labels stand in the way of "effecient" production of the computer software good.
Any society that has to ammend it's fundamental tenent (such as American freedom, squelched by business-bolstering copyright/patent-law) obviously demonstrates it's lack of fundamental viability.
If Capitalism can't exist without artifically constraining laws, then capitalism is a failure (if not today, then some-day). Personally, I don't think capitalism is a failure. I think that in light of modern work-paradigms, it would actually be possible to live in a capitalistic society free of copyright and patent law (of course this even means GPL would have to go). It throws away work-for-profit mentality in exchance for work-for-compensation. It would still be possible to profiteer, but you wouldn't be able to rest on your laurals.. You'd have to have a business plan where you could at least for a time, be more efficient at production of a good/service than any free-rider. If this is so strange, then look at what we say must happen with the music industry. We say that they have to wake up and accept the change. But it's no different then saying that software or medical drugs would have to give up manufacturing royalties (and thereby hurt venture capitalism).
Who then would bother spending billion dollars on medical research, or on a nuclear power-plant management operating system, if as soon as their done, a generic company will sell at bottom dollar, and thus equate a massive loss. I think the answer is very obvious. If people are afflicted with a massive disease, and we need to spend a billion dollars, then people will demand sanctuary from the government, which will therefore grant money to an organization which would then hopefully find a solution. The findings (successful or not) would be open to other organizations (since it was publicly funded). You're paying a "service" fee. A nuclear power-plant would likewise have enough value to pay a company to make custom Nuclear-power-compliant software. That the next power plant could then steal the code and use it at no additional cost is irrelevant.. The first nuclear power plant couldn't operate otherwise. Further, the second plant would have to pay service-fees to reverse engineer and recustomize the code to fit their need. The actual reduction in cost to the second plant would actually not be substantial.
If tomorrow we abolished copyright law, the first thing that would happen would be that stock prices would go rock-bottom. Share-holders would know that their IP-based profits would dissapear. Especially in the medical and computer fields where the information is most of the final result. But these companies wouldn't disssapear. First, they'd spend billions of dollars on litigation to recover their previous nest-eggs.. Many companies would go belly up in this last-ditch effort (assuming they weren't successful, which is irrelevent to this discussion). Some would simply adapt. They'd convert to a truely-free-market economy (e.g. migrating to service or minimial-cost production, or marketing (such as inferior brands who's label is all that matters; e.g. the textile industry)).
All commercial software companies would go belly up immediately, since it would no longer be illegal to simply take a single MS CD and install it on every machine in the world. The only thing MS could do would be to enforce registration (which is fine by darwinean capitalistic ideology), but this wouldn't last since hacked MS products would eventually appear which thwart registration attempts.
Strangely enough, this wouldn't hurt society as a whole because we already have sufficient advances in technology and medicine to survive. What would definately hurt would be the economy. But that would only be temporary.
It's true that we'd have a stagnent economy for a while as we learned to shift our resources from high-risk venture capital to other methods, and it's likely that we couldn't achieve the same order of magnitude financial capability as with constrained capitalism, but we have to look at the larger picture. As a society (and thereby a governing body such as congress), what are we trying to achieve? Growth? That's a selfish endeavor; not something society as a whole should be concerned about. Do we really care that we have more people next year than we do today? Do we really care that the total number of dollars in our society grows each year (even though our personal stock-piles are stagnent with respect to inflation). Do we really care that one country grows exponentially with respect to all other countries in the world? (A popular bastardization of capitalism is that you must not only compete with your peers, but measurably squash them). I say that these things should not matter. What I consider altruistic measures for a ledgislature are the accomplishment of zero non-transferring unemployment (meaning the only unemployment in a society should be that of those seeking to change their current work-status). If everyone is successfully employed and inflation is marginal, then times are good. If we move from a venture capitalistic society to a service-based society, then we obviously have a demand for employement. Further, artificial weath (as from get-rich schemes based on IP; e.g. many forms of venture capitalism) is reduced which dramatically curtails inflation. While I can't currently prove it, I am under the belief that inflation is dramatically reduced when you reduce the number of monopolies that are paracitic to the general population. While some monopolies are enevitable (telephone / cable industry, etc), the fewer monopolies you have (assuming competative markets can replace them), the less excess each citizen must pay (eaten from economic welfare), and thus the less it costs to live. A reduction in the number of forces encouraging higher costs of living should constrain inflation.
I'd be interested to hear further arguments for or against a free-market economy (again, I'd like to reinforce that we don't live in one). I know full well that no politician would cut off his own neck in suggesting curtailing IP-laws, but it doesn't hurt to contemplate such a society.
Open Source software is insanely susceptible to hackers/bugs/etc. Moreso than even M$.
Explain then how there have been few viruses that have had success against linux/BSD servers, while IIS + winX has a new security headline weekly. You could make the argument that the proliferation of Windows machines makes them more suseptible to an attack, and therefore if Linux was 99% of the market, we'd see few Windows viruses and lots of Linux viruses.. But I'm talking about servers where Apache has a >50% share (at least at one point in time). Apache is recognized as one of the most robust services out there, and it's obviously a critical piece of the cracker puzzle. OpenSSL + OpenApache + OpenLinux (with and open firewall design) are extremely robust (though not impenetrible). Most of the security holes that I'm aware of deal with user-level applications. If you're trying to hack a bank, you can be sure that user-level applications will not be a factor. Should a bank use winXP Pro + IIS then? (Sadly I know a few that do).
I just couldn't sit by with such an unsubstantiated comment went by.
Second, MS is known for their relatively LOW prices in commercial software (want to compare Oracle pricing anyone?).
Once apon a time they had low prices. But not anymore, and your statement is becomming increasingly false. As their software actually gets close to companies like Oracle / IBM in terms of capability and robustness, their prices are reflecting it. Don't forget to take into account upgrading into the cost. A nice expensive SUN hardware work-station has been running the same version of Sybase for years at our company, and there hasn't been a compelling reason to upgrade; that's thanks to the mostly settled UNIX standard. MS APIs are ever-forward looking and thus obsoleting of previous standards (read interpolibility and network security methods).
You may not like any software that MS does, but others are able to see some of the achievments that MS has made.
I fully acknowledge that MS has contributed a lot, and provided a realistic alternative to UNIX / custom-solutions. Such innovations as MS Works, MS Access, and virus-friendly do-anything office-suites have contributed leaps and bounds to the industry. (note my synicism)
The problem is the same as if you had to choose which company to work for. Company A has low pay, marginal benifits, not particularly exciting projects, but have open and moral business practices. Company B is very radical, very growth-oriented; it has a lot of flash and excitement. It's already number one in two or three fields, and it has it's sights set on dominating other markets. The business men are ruthleess, smart, and know how to turn a buck.
Well, it would seem that most people would prefer Company B, if for nothing else than it's like a shiny toy to a child. But, risk-takers, statistically fail more often - including the probability that illicit legal affairs go on. Look at Enron, MS (by all rights, if Bush hadn't stepped in, their stock would be significantly less valuable today as they'd have been broken up like the Baby Bells), etc. Ruthlessness is a sure-sign of anything goes, which is another word for morally-indifferent. Such action works great for the here-and-now, but says nothing about one's future.
Thus the conservative employee would be best advised to choose company A, and do their best to make it competative. While in our Attention Deficit Disorder society (including stock-investors), it's entirely possible that good companies will be jettisend in favor of the Blitzkriek companies, you still have a good measure of reliability.
The same sort of comparison SHOULD go on when choosen business associates (including who's administrative products you purchase). You wouldn't purchase a database from a company that you've never heard of; least you eventually lose support. Likewise, dealing with a company who's made a publicly known habit of coersing it's customers (strong arming IBM, schools, etc via maliscious audit-threats) whenever they determine probable gain is a definate factor in the risk-analysis.
For many of us "heretical slashdotters", the risk-assesment is at the forefront of our minds. We stand to lose x,y,z as we chant our mantra over and over. That common businesses don't always take these factors into consideration is what we really fight. That they decided that the risk-cost isn't higher than other opportunity costs, is fine; so long as it is part of their calculation.
I assume you're talking about the velocity of money (where-apon a dollar is respent many times over, so long as it never sits still). While it's true that investing money in an institutions which immediately re-issues can produce a >1.0 velocity during average to strong economic times, when the economy eventually falters (which it didn't really do during our little American recession), that multiplier contracts and can go < 1.0. At which point, the artificially inflated money suddenly dissapears.
The larger the mulitplier, the more it hurts during a real recession.
As part of the multiplier, is the idea of financial float; checks are written that are immediately deposited by the federal reserve into bank A, but funds are not immediately withdrawn from bank B. Thus extra money temporarily exists. So long as the number of transactional cash-flow remains constant, this temporarily created money sticks around. Once a recession hits, purchasing slows/stops, and checks clear...
Since recessions are cyclical, anything that artificially heightens an economic boom will effectively over-stear the economy and make the ensuing correction more powerful.
I personally believe the 90's represented an incredible growth of efficiency in our economy. Banks were allowed greater freedoms, computers reduced transactional costs, demand shifted from expensive goods to high-profit-margin-goods (like software solutions, and raw experienced human labor such as IT and effective middle-management). Since people were generally required to do more, their higher pay was justified; we simply did more work per day than we did in the 80's as a whole.
I believe this growth allowed us to avoid a catastrophic recession / depression on our economic correction. Aside from another socio-technological revolution, we can't sustain such growth as in the 90's, and thus such augmented multipliers are not ultimately to our benifit.
Disclaimer: I've taken several undergraduate economics courses, but I don't claim to be an expert, so comment-accordingly.
Second, I love this. They don't even have a THEORY on why this works. It just does.
Well, I'm an undergraduate Electrical Engineer, so I only have superficial understandings of how semi-conductors interact with light, but it doesn't seem too great a stretch of the imagination.
First, semi-conductors work based on the principle of the band-gap (which they even mentioned) (correct me if I'm wrong with any of this, I'm doing it straight from rusty memory).
A little background: The outer 8 electrons held by an atom are the most important (the valence) - They are responsible for the bonding of other atoms. The configuration of all the electron orbitals in free space is nicely geometric; the first two electrons form a spherical shell (s-shell), the second 6 form dumb-bells in each of three axis's (p-shell). These types of configurations affect the geometries of the connection of the atoms. Configurations get more complex as the number of electrons grow (which is somewhat independent of the atomic number (number of protons), but such ionized atoms are unstable; especially when the number of electrons differs dramatically from the #protons). The important thing to understand here is that each additional electron takes more energy. Instead of worrying about the geometries, you can plot each electron orbital at a different (successively higher) energy level. Different atoms (characterized by atomic-number and even, to a small degree, the number of neutrons present), have differing characteristic energy-levels. The discrete nature of atoms includes the probabilistic nature wherein electrons have an extremely high probability of occupying the exact energy levels (which can be thought of as the distance away from the center of the nucleus). There is a chance that an electron will pass through any point around the shell of an atom, but it's highly unlikely that it will deviate from its characteristic point.
But, since different atoms have different characteristic levels, warping an atom will warp its points. Warping can occur by simply placing two atoms near each other (such as in an ionic or covalent bond). As it happens, when you squeeze atoms closer and closer together, the discrete lines that represent the energy levels start to merge together. Eventually the 8 outer valence bands merge into one continuous band... As you squeeze them even closer together, this band breaks into two continuous pieces. As you get even closer together, these pieces get further and further apart (I would presume that eventually one of these bands starts to merge with preceding energy levels, but that's not relevant here). This gap of continuous energy levels is called the band-gap.
As it turns out, in perfectly bonded atoms (those where every electron in the valence layer are bonded, and each atom has exactly 8 outer electrons; such as carbon, Silicon, etc) we have a total of 4 electrons that fill the inner continuous shell and 4 electrons that are void in the outer continuous shell. BUT, that outer shell is looped across neighboring atoms. When a diamond-lattice is organized (which is as close as you can possible get multiple atoms to sit next to each other), you have the greatest band-gap you can get for that particular element. Different elements (or even molecules) that can form the diamond-lattice will have differing characteristic band-gaps. What we have here are 4 electrons that are tightly tied to a core atom, and 4 potentially absorbed electrons that can freely be shared across every single atom in the entire crystalline lattice. In semi-conductor crystals, the problem is that every electron is accounted for so there are no free electrons to put into the outer band (which could roam free as current through an almost zero-resistance substrate; due mostly to quantum effects). Impurities are therefore inserted into the crystalline lattice which act as ionic donators of electrons or ionic acceptors of electrons (namely atoms not in the 4-column of the periodic table). Thermal excitation (heat) causes an electron to be ripped from donor atoms and those which are then quickly swept up in the outer-most continuous band.
Normally, electrons must have a precise energy-value in order to live in an atomic orbital. When an atom absorbs an electron, it gives off a photon of the remainder of the energy. To change orbital-levels, it has to accept a photon of exactly the correct amount of energy. It can accept a larger energy photon, but it will again give off the remainder of energy. Eventually that excited electron will fall back to its lower energy level, giving off another photon which will have the exact energy as the distance between the two energy levels.
In the continuous region of these silicon atoms, excitation between energy levels isn't apparent, since an electron can have any value within the region. The only difference is that separating the band gap... An electron from the inside can jump into the outer band if it's given at least enough energy to make the jump... This gap is usually enormous for semi-conductors. I believe its 1.2 electron Volts for Silicon, and 10 electron Volts for Carbon. The 1.2V is within the range of thermal excitation. That means that heat (in the form of vibrating atoms in the crystal) is enough to shake an electron free; e.g. jump the gap (like water successfully spitting to the lid of a boiling pot). In carbon, however, room-temperature heat is no where near enough to make the jump. This property (along with others) is why we don't use carbon-based semi-conductors. Germanium and silicon are much more practical in our particular earth climate.
There is another aspect to the band-gap that is relevant to our discussion. Each electron has not only an associated energy, but a quantum-form of momentum. You must not only have conservation of energy, but conservation of momentum. I'm a little fuzzy on this topic, but this momentum is represented by the letter k, and we can plot energy verses k for different things. For semi-conductors, we get parabolas, and inverted parabolas, but with discrete points. This says that while we have a continuous set of energy levels within a region, we have only a certain set of allowable energy+momentum values for the electrons. And like the discrete energy levels of atomic orbitals, you can only have one electron occupying a given state. In a rather unfulfilling way, I'll stop talking about things that I don't fully understand and simply say that this multitude of characteristic parabolas says that in order to have an electron jump, you have to not only have a precise amount of energy absorbed or emitted, but you have to be able to transition your momentum somehow. Energy transition occurs through photons, and momentum transition occurs through phonons - which is energy present in lattice vibrations (e.g. packets of heat).
Gallium arsenide is an example where the lowest point of the upper parabola and the highest point of the lower inverted parabola are aligned with respect to momentum. This means that the smallest amount of energy needed to make an electron jump the gap requires zero change in momentum. Because of this, gallium arsenide crystals easily will easily absorb or emit light with no dependence on the heat of a lattice. For this and other reasons, GaAs is great for laser diodes. Silicon, on the other hand requires a momentum change for its lowest energy transfer. Thus lots of heat is generated and absorbed; (Not to mention that silicon doesn't conduct heat as well as some other semi-conductors).
Given this superficial description, what I get out of this is that heat of a certain resonant point (in the form of vibrating atoms in the crystal) could provide the proper momentum shift needed for efficient electron excitation. You'd still need to provide photonic energy for the transition, but you'd have a perfect combination of heat + light absorption. Eventually (due to statistical decay), the electrons would fall back to their lower-energy-level states. But they'd give off light of specific frequencies.
Putting all this together, my initial impression from the article was that that the tungsten injection into a silicon substrate change the characteristic e-k curves enough to absorb the phonon-heat generated by IR light. The result is a 60% efficient absorption of the heat + light (e.g. nearly perfect efficiency). That energy is retransmitted as diode light (e.g. an exact energy level transition, producing a constant level of energy photons, which requires an equally constant frequency of light).
What I don't know at the moment is if this is actually emitting mono-chromatic light, or if a multitude of frequencies (e.g. white-light) permeates. The only way I could see white-light emitting is if the standard tungsten light-bulb is making it, and the Tungsten semi-conductor is amplifying a particular frequency.
My understanding is that they said that IR-frequencies are synonymous with "heat". They frequently used the term black-body radiation. I remember IR-HEAT being associated with green-house effects; the angle of refraction is low for IR and glass, for example. So when sunlight enters your car (at a direct angle), it bounces off things but hits the glass on the inside at too great of an angle, and thus bounces back inwards, amplifying the total heat.
Not that I'm satisfactorily answering your question, but throwing out some food for thought.
Even better, perl has several SQL-emulators that can work on flat files or berkley DBs. Using this approach, you could apply portable complex queries / relational constraints easily to legacy data.
and Linux becomes illegal (you can't have efective copy prevention with source code), what will the movie studios (the same who pushed the SSSCA in the first place) use then?
Well, that assumes that the sound-cards themselves don't hold keys like DVD players and thus the audio format of the furture isn't encrypted with audio cards exclusively being able to decrypt them.
You had several valid points, some of which I even acknowledged.. But still felt the need to further comment.
8 bits?! Why 8 bits? You make it sounds like this is atomic, when it's not. At all. If you're going to go for the theoretical minimum, go for 1 bit.
Nothing magical about 8bits, except that we like multiples of 8. An 8bit add requires two 8bit numbers = 16input bits, and 8 output bits. Ignoring optimization and implementation complexities, that's on the order of 65 thousand transistors for a two-level adder. That's definately do-able in today's technology.
Two 16bit numbers on the other hand, would likewise require on the order of 4 billion transistors, which it not currently possible with today's technology (my guess is that by the time it is, we'll be using 128bit ALU's).
My point was to find the most practical pure combinational logic unit. Thus to suggest a 1-bit machine implies that we miscommunicated.
Point A)consolidate in c-structs? Programmer laziness?.... However, that can often slow down the code...
An alpha has a nice little byte-swapper which relatively efficiently extracts a byte out of a multi-byte word. It's computationally slower to work with than just testing the whole word, but when you take into account bandwidth / cache-space considerations, it's a boon. It can't be easy for a compiler to optimize away: struct {
int f_flag1, f_flag2; }...
Can it drop the flags down to 1 byte?
While it's not always true for each architecture, you should find higher performance swapping out the int's with char's.
I'm not familiar enough with the state-of-the-art in compiler technology, so perhaps this can be determined. I was just generalizing on this type of point, nothing more.
Point E)
The points you mentioned here would be included in the sort of tricks that have been devised over the years to avoid upping the word. It just furthers the idea that larger word-sizes are expensive and worth side-stepping.
How on Earth did Apple manage the migration from 68k to PowerPC then?
Very simple. Lack of competition.. They held a monopoly on ALL OS / motherboards. The only real competition that I'm aware of were the 3'rd parties that sold the various chips / expansion cards.
When you have a vertical monopoly like that, you can coordinate an architectural shift. If Intel decided to start a new CPU line, and it turned out to not provide the best bang-for-the-buck, then AMD/Cyrix competitors could supply legacy and current MS-products a better alternative. Intel would have lost all that money. From the other side, MS is spread so thin that they don't have the time to rework their core to be optimal on multiple platforms (look at the death of NT on any non x86 platform). The lack of a compelling reason for someone to purcase the alternative platform says that MS shouldn't devote too many resources in that direction, which of course kills it off. Hense platform architects are at the mercy of software people, who must provide killer apps for that platform.. If any major killer app isn't immediately available, then a domino effect of lost support will occur; and more importantly, business people understand this a priori.
This is actually a lot more exciting then it might first appear. On the one hand, you have a controlling hand-of-God who enforces their will. So long as they can project a bottom line that benifits their customer, they can make radical changes (shedding virtually all of it's former self). On the other hand, we have multiple independant organizations, who each act in their own best interest. In monopolized environments, changes are swift and clean (but not always in the best interests of everyone). In independent environments, no organization can squander or otherwise take too great a risk. Efficiency is upheld, since only rational decisions can be made (involving the mutual benifit of progress). The side effects are a slowing of evolution, and an accumulation of "useless appendages". On the other hand, it provides an incredible level of trust on everbody's part that the architecture has staying power; that it'll weather the storm of change, instead of flippantly changing with the current mood; throwing 3'rd party interests aside when it's convinient (read Apple's resinding licences over the years).
Personally, I think an architecture that has "grown" over the years is more remarkable then one that simple borrows the best ideas that come out of universities.
Note that I'm not really advocating x86's. I'm just admiring it's successful history. The proliferation of UNIX and the general ability to recompile source could theoretically alleviate a "better" platforms' lack of killer apps, and thus perpetuate a radical acceleration of architectural designs.. Go Linux!!
Note I'm replying several sub-posts all at once here.
All other things being equal, a processors with larger word size (instruction sizes and address sizes) will be faster than those with smaller, though, depending on application, the results can be negligible or even worse, especially if compilers and programs aren't properly optimized.
I disagree from an architectural standpoint. In an ideal world, we'd all have 8-bit machines. All our arithmatic would be insanely fast; we'd be able to use combinational logic to allow two probagation levels for ANY operation (add, sub, mul, div, sqrt, log, etc). That's because it's cost effective to do so; a minimal set of possible outcomes. I'm not completely sure, but I'll speculate that it's possible to arbitrarily generate an arbitrarily sized number from just these 8 bits; though most likely it would be programatically (even if done via micro-code), and thus would be non-optimal for larger than 8-bit data-sets. So obviously, as we've been able to, we've increased the data-length throughout history as we've demonstrated a need.
Contrary to the impression that's given in these posts, a larger word size fundamentally is slower in calculating smaller values. Sticking with higher performance two-stage combinational logic requires an exponentially increasing number of transistors. Breaking the logic up into tiers allows designers to trade the number of transistors for the number of probagation delays. The more delays, the slower the clock; the more transistors, the less practical the design (due to heat, cost, and feasibility of fabrication). Pipelining somewhat helps aleviate the issue of extreme probagation delay, but it's impossible to achieve 100% efficiency, and thus you're practically garunteed slower operation for deeper pipelines. What's more, pipelining requires additional probagation layers for buffering, so you take an immediate performance hit; speculating that you'll achieve greater over-all performance.
In an ideal architecture, you'd minimize the probagation delays for each instructional unit, but practical measures say you must group most, if not all, of the CPU such that the slowest part drives the system. (P4's are nice in that they sub-divide the clock for the simpler integer units).
Combining the two ends, we can better appretiate the trade-off.. If we're performing large-valued arithmetic which is slow programatically (emulated 64bit), then it's worth the extra cost (towards the speed of each operation, and in terms of the number of transistors). In other words, one hardware 64bit add is most certainly faster than than several assembly language instructions that piece together 32bit values. BUT, now all your 32bit arithmetic is slower (unless you have separate 32bit/64bit logic cores).
It's possible to design 32/64bit cores that only take as many clock-ticks to complete as necessary, and thus 32bit arith isn't horribly slower, but there are definately additional probagation delays. The augmentation to 64bit can never increase the speed of a 32bit operation. (Any speed ups must be due to over-all advances in computational efficiency, which should benifit a pure 32bit core even more).
The trade-off must then be a statistical one. We cost out the largest word size that provides benifit. You're going to have arbitrarily large ALU operations (just look at encryption), so choose a cutoff where a certain percentage of all operations occur at that high of a word size. This is how we moved from 8 to 16 bit, and then the painful shift from 16 to 32 bits. And for server-targeted machines, the shift has already been cost-out to adopt 64bits. The desktop has not yet made sufficient requirements to adopt 64bits, though the underlying x86 CPUs are being shared in server-space which is nudging 64bit's acceptance.
Another important factor (which is presumably obvious in concept) is that a higher word-size has a greater probability of wasted space. A 1-bit boolean, for example, wasts 63bits.. Booleans are very common, and though they can easily be consolidated in c-struct's, such is rarely the case, since there are memory alignment issues (and flat-out laziness on the part of programmers). The wasted word-space also affects the instructions. Rarely do you actually see 64bit aligned CPU-instructions (except in VLIW or in places that the data-word-size was irrelevant). Such a situation would have massive implications towards performance. But one serious consideration is that the population of 64bit constants using a 32bit instructional word is expensive. Now you have to perform at least 3 (probably 4 or 5) instructions just to load a constant. Suddenly "a++" starts to look scary (at least when non-optimal compilers are used). In all cases sub-word-size'd instructional arguments are permissable to the delight of compiler designers, but there are still classes of problems that thwart this.. Namely memory addressing...
Memory addressing is arguably the strongest supporter of 64bit architectures. The 4GB limit is already apon us on desk-top machines (I have half a gig on all my home machines, and I don't need it). When you add swap-space, it's entirely possible for modern desk-tops to run enough apps to desire 4+Gig of memory. (Especially considering that large chunks of the address space are wasted). Aside from the various tricks designers have employed over the years to avoid augmenting the address space (8086's segment-registers, 80386's segment-selectors, OS's swapping out apps completely from memory, etc), it's arguably slower to emulate larger address spaces.
In addition to the above arguments against larger address spaces, there is massive cache polution; doubling the word-length, literrally halves the usefulness of a cache-line-load, unless you were previously emulating a larger word-size. You can only load 4 words on a pentium-class cache-line-fill instead of 8. Your bandwidth requirements litterally double (unless you don't standardize at one word-length).
Now in contrast, there are a few advantages. If your minimal word-size is larger, then the number of address pins that you need are reduced. But this is really independent of the core word-size. Pentiums have long required 64bits for their external bus, and use even larger cache-line sizes. Thus most of the advantages attributed to this argument are moot.
Theoretically, an architecture can be designed to split an ALU such that it acts as either 1 64bit unit or 2 32bit units. This is especially true for vector-cores (which are already up to 128bits for main-stream processors). In general, however, there is still the trade-off here, since additional logic-probagations are required which slow down the general case of only a single But comparing the X-box and the Nintendo 64, which were released many years apart won't buy you much of a conclusion other than current processors are generally faster than older processors.
I'd like to address the nature of large bit-sizes with respect to graphics. While this isn't my expertise to the extent of the above, this primarily affects the bus width. In graphics, you commonly have multi-integer structures (red, blue, green, alpha (opacity), Z-depth (the depth into the screen the geometrical object that drew this dot is), stencil, etc). The entire structure is usually just 16bits, 32bits, 64bits, etc. The larger the structure, both the more features you can pack into it, and the more accurate each individual number can be. Thus saying that an architecture is 128bits purely based on this is very misleading. What's even worse are labeling the bus-width numbers (e.g. 128, 256). That's like calling the Pentium I a 64bit CPU, just because it has a 64bit bus (used purely for cache-line burst fills). Yes it makes it go faster, but so does shortening the length of each wire; it's not really innovative. I'll throw this in, but I'm starting to get in over my head; Graphics units (especially the filtering parts) make heavy use of hard-wired combinational logic units. The number of bits going into these units is really meaningless (how many thousands of wires go into the control logic portion of a CPU?). Thus the ability of a custom pieces of hardware to utilize larger bit-depths is unimpressive. What would be impressive would be to say that a Graphics unit does it's integer / floating arithmetic in 128bits so as to minimize error (even though the input/output might only be 8 or 16bits per atomic unit).
Course congressmen know full well that they'd lose their barganing chips; the I'll scratch your back if you scratch mine. The cornerstone of US democracy - compromise.
I remember cheering when the president got line-item veto power, and thus could signify the death of pork. Unfortunately this gave the president too much power, and was overturned.
Still, somehow a campaign finance bill was recently passed so anything could happen.
So does that mean it has 3.6 Terabytes of swap space?
Cute, but swap is optional as far as I know.. When you have enough memory, or when you want real-time (at least in solaris or qnx), then simply leave it out.
There was once a project to embed a RAMBUS controller into an intel CPU. I was hesitent to post the above because I wasn't 100% sure that the P4 indeed had it. Unfortunately I don't have time to research and find out what article (from over a year ago) prompted this.
While it's blatently obvious that embedding motherboard components will make that particular implementation faster, it does little to allow future advancesments. In what-ever article I did read about the RAMBUS controller, it didn't say whether the feature could be bypassed so as to utilize a more modern controller somewhere down the line. I remember feeling frustrated over that matter. What if CPU's came out with AGP 2x controllers in them when AGP first started. It would have slowed progression of AGP 4 and now 8x, simply because most cards would be sold to existing 2x-based systems.
I don't have a problem with embedding technology, so long as it can be upgraded (see the nForce chipset). Likewise with MS's IE, MediaPlayer or MSIM. I have no problem with them putting it on their OS, so long as I can 100% disable it if/when I choose Mozilla / winamp / AIM. Unfortunately I don't know how well you can shut down media player in XP; especially given XP's alledged taddle-tale role. Not to mention the insult of having MSIM reactivate when viewing hotmail.
What's more, I'm sure that such goodies in the windows line encourage them to jack the price up. This is the problem.. We're paying for a bundle of goods that we can't fully use, and we're not given the option of purchasing a streamed line version at a lower rate. (e.g. OEMs could ship winamp + Mozilla + Macaffe and shave $10 - $50 bucks off the OS).
Slashdot Mirror
I'm sorry, but this is the same argument used by the RIAA / MPAA that says, "oh no.. Something's changed in our prefectly controlled little world, and now we don't know if we'll even be alive in x-years.. Please Mr. Congressman, stop this change now before it's too late."
Change is a mixture of good and bad.. Deal. It's not the end of the world, though it might be the end of a cushy job.
This is industrialization-era thinking with unskilled labor as your "working/criminal class". There is an entire economics of human resources, and that too has a supply/demand curve. What's happening with globalization is that the supply curve sky-rockets. BUT, what again, you're thinking in terms of unskilled labor, where a worker is a commodity (indistinguishable from any other worker). In reality, each worker carrys with him/her a set of attributes and excess baggage. Employers will cost-analyize the benifits against the costs. Thus a domestic worker will always have a certain edge over a foreign worker. Will that edge equate to tripple the salary? Who knows. But in the case of Asian software developers, I've heard lots of horror stories, and thus any management worth their salt should do serious research first (There are soooo many hidden costs as have been pointed out in this thread).
So in response to this fragment, the rich only get richer as they have some competative edge over the working class. A worker could become a doctor, lawyer, engineer or any other skilled labor that is in high demand. Kings and the Aristocracy only did so well, because they were militarily superior (back in the days of knight-hood). They only maintained their wealth due to complacency among the subjects. In modern terms, the only reason that the rich like MicroSoft continue to get richer is because of our beloved ledgislators that support such non-free business practices (involving patents, copyright, and the enforcing of other non-free-market laws).
My point in this section is that the only way to prevent a company from going global is to ledislate, and hopefully it can be seen that this is a non-optimal solution.
You're over-simplifying. If America blocks it's borders with respect to international workers and immigration, then you can create an artificial environment where the only wealth that will leave a country are via imports. But if you're as wealthy as America, why would you ever need imports? Because other countries can make things better/cheaper. How can this be? Because your artifical sand-box has inneficiencies with respect to the rest of the world (who has more competition / employees who demand lower wages). Thus money that would be lost due to paying salaries abroad is instead lost due to exports (and America's trade deficit is annually enough to bankrupt entire nations). By importing human resources, American companies become competative and lesson the trade deficit. Theoretically, domestic production is cheaper than imported goods, and thus stands to raise purchasing power of consumers (which helps the economy). In reality, American prices are so much higher, that at best they would merely be competative, thereby not appretiably affecting consumers.
There's another angle. The best-of-the-best from India, China, etc feaverishly apply for Visa's to the US. Once here, they have a good chance at competing against otherwise complacent Americans. The heavily math/science focused educational system in other countries puts them at a distinct advantage with Americans (all else being equal). Most border states heavily complain about immigration problems, and indeed most immagrants move into the cities, heavily burdening us. Many modern countries (like France and the US) are burdened with the issue of immegration. When you have more citizens, your wealth is diluted. Legislation has often been the only resort to counter this phenomena.
But why do we even have this problem? It's because there is such a massive disparity in the wealth of other nations that it's worth the massive barrier to entry that a country imposes. This is why Mexicans / Chinese / Indians come here. If they had work and a decent living style in their home country, there would be less incentive to migrate. Remember, we're talking about the creme-de-la-creme that come here. If they had jobs (that gave them greater purchasing power) back home, then they'd possibly stay, thus reducing the immigration factor. In the case of Mexico, it's more the cunning / daring ones that make it here (and sneak about until they can find a way of being nationalized). Large scale cheap labor should still have a good effect there for similar reasons.
You're suggesting (100 + 1) / 2 = 50 all around, but you're picking numbers out of the air. Because of the infrastructure, only a limited amount of physical labor can be transplanted to other regions. This is only a small percentage of the total GDP. Unemployment does reduce one's GDP, but it also has an effect of reducing salaries (think great depression). Reduced salaries avoids the inital issue at hand. These factors oppose one another so it's not as simplistic as you suggest. Further, wealtheir internationals have greater purchasing power, and thus increase export demand. If the organization is still head-quartered in the US, that means US tax revenue on work barely done in other nations.
In short, this is a natural process at work, that's gone on since the dawn of time... Darwinistic competition over scarce resources.. And only the most efficient will survive. So either give up a cushy job, or legislate and watch as your trade deficit continues to grow uncontrollably until there is little net asset left in the US. (Tarrifs don't work by they way, they're just a form of placibo to make out-raged citizens temporarily happy).
-Michael
In pre-forking, process A sets up its environment which includes the service port, and a shared memory segment (often just a a simple memory mapped file). Then it creates a bunch of worker forks. At this point, the master could simply do a ps once every second and see if it's children have died, but usually the shared memory is used to communicate activity to the master.
Since all the forks maintain the file-handles, anyone of them can accept a job request. Thus they can be truely independent, and your issue of a corruptable monitor doesn't apply. (Except in the case that crashed workers need to be replaced by the monitor; but I feel that you were saying something different).
It's true that sharing information between the workers and the monitor has a potential for corruption, but I'm not seeing how mainframes avoid this (except if you're saying that the functionality normally handled by programmed IPC is relegated to other tasks).
I'd say that this is a big gray area. I think extreme tolerance is exactly what is hurting American society; tolerance assumes that deviance will turn out for the best; and that intervention will cause more trouble that it will rectify. Thankfully we're not so bad that we indifferently allow children to have sex on school grounds, but there has to be a balance. I think generalizations of what can and can't be done (such as the introduction of personal opinions (including politics)) are hurtful, because they legally restrain practically exceptible methods of helping students.
My problem is that you _have_ to introduce personal creativity in the classroom, but you are refrained from most natural methods (such as discipline, or other method of steering the student, when you know the allowed mechanisms will not work). It's taking away the straw and forcing them to make bricks.
The correct approach (in my opinion) is to have a proper balance and an allowable dynamic. Have peer review as an intricate element to operation, and allow extreme contempts by peers to be reconsiled at higher and higher levels as necessary. One rule does not fit all.
That a political science teacher favors one president (or candidate) and spends an assymetric amount of time on them would be considered introducing personal politics. But the reality is that you can't spend equal time on all subjects, and the teacher's personalization is essential in making a class worthwhile to a marginally interested student.
A technology class, likewise, will have to (at times) demonstrate failed works, and the causes for failures
At my high school, the CIS guy was WAY over-worked. In fact, we mostly had to teach ourselves CIS since he was always on call somewhere in the school (which made the slow lost and the swift excelent par-none).
Given budget cuts, the idea of adding additional staff probably doesn't sit well. And given the over-working of network staff, I would find it hard to impose additional burden apon them.
Hehe.. Still don't think MySQL is up to the task of data-robustness (no roll-backs among other things). Given that grade-reports are more valuable than slash-comments, I'd argue for at least something like postgres (or better, even sybase/oracle on Linux).
The biggest problem is that you _have_ to garuntee data integrity and security. You have hackers in the high schools, and now you're giving them network access to machines that they probably play with at home. You're basically betting that your developers are more sophisticated than the students (or students non-school friends). At least with the proprietary systems, the likelihood of students even knowing what's under the hood was slim.
By the way, by network access, I'm referring to the fact that ethernet drops are generally available throughout the building (laptops can easily interface). Further, using Apache implies external access. (A developer _could_ forgo network access to the server, but this again implies that school sysadmins are sufficiently experienced).
-Michael
Effectively, a particle in quantum physics is a cohesive bundle of energy. We measure the mass of that energy in "electron-volts" (eV), which you can think of as a electron-level volt-meter. It's similar to measuring the voltage of a battery; we can't directly see how much charge is in a battery, but we can see how hard it pushes / pulls a test charge. Likewise, we can't see how big a proton or electron or up-quark is, but we can see how it affects other particilars of similar size (e.g. an electron as a reference point). Due to the massive deviances in particular masses, it's hard to know for sure if a photon is truely massless (even though it carry's energy). post-modern quantum physics speculates that photons, neutrino's, and even gravitons have mass. (Yes, this does imply that gravity has a weight of it's own. More precisely, the emision of the force of gravity adds weight to the space between two particles.)
The substance of the particle is subject to debate. String theorists believe (if I'm not mistaken), that all particles are made of strings of something (which we'll never know), and that those strings wrap around space (which we also can't know it's consistency)- warping it and being stretched by it.
Another point of view is that of Ether, which we tend to hold on to, since quntum physics is so similar to our percieved world that it would be a shame that such patterns could not be known to persist at different scales. One theory that I like is called
Aethero-kinematics. It's based on the idea that tiny hard balls (perfectly elastic, like steel) bounce about in different patterns (mostly vortexs, like in a drain). All energy is in the form of the kinetic energy present from these bouncing balls. The cohesion allows for quantum particles. The augmentation / contraction of mass (via Einsteins special relativity) is explained away the same as Mach-theory (where an the air-resistance increases exponentially as you exceed the speed of sound). The "speed of light" is merely the average velocity of the balls. The explained reason why we can't perceive relative motion against the ether of space is that earth is not moving with respect to the ether about it; nothing does. Motion is only ever a small fraction of a difference in speed from it's surrounding ether. Lastly, the concept of experimentally determined transverse nature of light is nicely explained away in Aethero-kinematics in common sence ways. (having to do with the probability distribution of collisions of particles in an ideal gass)
Modern quantum physics simply ignores the what's and hows of particles, and simply says they exist with certained measured properties.. That's it, that's all, that's ugly. Because of this, I tend to look at models like the above (so long as they fit the experimental data) as a way of putting my mind at ease. The problem is that until the theory's demonstrate validity, we can't take the analogies they present (ideal gas, or strings) too far in extrapolation / interpolation.
As for waves (also questioned in this thread): a wave is a regular periodic fluxuation. Longitudal waves are like a wripple in a violin string or cresting waves on the ocean. If you just look at a single water molecule, however, you'll see that it doesn't move forward, but instead up and down (just like a boat). You could also look at a police-car flashing light. The color of the light slowly fluxuates from red to blue and back again in a definite period. If you took a cardboard box and punched a hole through it, you'd see on a wall the color fluxuation. If you look more closely, the fluxuation is merely caused by a rotation of two light bulbs. Photonic transverse waves are the fluxuation of the state of the photon from electric to magnetic (hense the phrase, electro-magnetic). An electron sitting still has only an electric field (which applies force to other adjacent electric objects (pretty much anything but a neutron; and even it, if you break it down into quarks). When an electron moves in a circle, it applies a strange perpendicular force which only affects other spining electrons. You can understand that it's different than charge because two electrons are attracted to each other when they counter-rotate (or rotate, I forget which). It turns out that rotation has nothing to do with it; it's the motion of the electrons (but the math gets harder). So here are two completely independent characteristics of a charged particle. As it turns out the transmission of photons accounts for both activities, so the photon is both a messenger particle for magnetic fields and charged-fields (electric-fields). Since a photon must always travel at the speed of light (relative to it's medium), it should be apparent that it works within a magnetic context (e.g. charge in motion). It seems that the photon fluxuates between the two in a sinusoidal pattern with respect to time (independent of it's physical motion). The "frequency" of the photon is the speed at which it oscilates a full transition between electric and magnetic. Such a periodic transverse wave-pattern has many astonishing properties. Most notibly that the same beam of photons when reflecting back apon itself can have interference patterns; namely that the waves can cancel each other out (or amplify one another). The best example of this is to take a beam of monochromatic polarized light and send it through a cardboard box with two slits on it. On the other side of the box, you should see a periodic pattern of light and dark spots.
I'm not a physisist, but I am an electrical engineer, so I have more than a lay understanding of the principles.
-Michael
Color is directly related to the frequency of the wave of a photon of light. (we'll ignore for now how a photon can be both a partical and a wave). Visible light is a tiny frequency range; IR, microwave, and radio are below it, and ultra-violet, gamma and X-ray are above it.
Additionally, the energy of a photon is directly proportional to its frequency.
E = hf (h is a constant of proportionality; planks constant divided by 2*pi).
When photons hit atoms, they're usually obsorbed. The electrons only obsorb a quantum amount of energy, and give off any remainder. Eventually the electrons re-emit this stored energy. The neat part is that all the emitted photons will have the exact same energy. Different molecules will have different discrete values of energy, and thus give off different colors.
It's possible that multilple low-energy photons can hit the same atom, and cause it to raise it's energy to different levels (each jump will require a different amount of energy, and thus give off different colored photons). It's highly likely that the energy would quickly be given off, but it's possible that the energy will continue to build until when it does give off it's energy, you've achieved a significantly higher frequency photon.
While I haven't looked at Fusion's method, it's completely plausible to accumulate lower energy photons as I've described above.
-Michael
Yeah!! I can think of an argument to this. First of all, the initial comparison was between DOS. Tell me that DOS is easier to use than Linux.
Now look at modern RH Linux (and it's brethren). Installation is easier today than DOS HW/SW was; assuming you have the proper hardware (which has always been a problem). If all that "grandma" needs is a button that connects them to AOL / MSN and a subsequent button for email / word-processing / web-browsing. You can _easily_ do that with existing GUI capability (once AOL, MSN get on the Linux platform).
It should be trivial to have Linux boot-up without even a login prompt.
The ONLY thing that Linux can't completely do is provide usibility transparency to the MS platform (sadly they've even been trying; which I find pathetic). But that's the nature of competition. We don't expect that a saturn coupe and a lexus SUV are going to drive the same way, why should we "expect" that software all run the same way (while we should always expect that there is some level of intutition for each manner).
As for proof, I've already gotten several of my past girlfriends to "use" linux while at my house. They know which icons to hit, and are largely familiar with netscape. Aside from running maple, and games that are targeted to the windows platform, much of the office productivity suite are becommign viable. The complexities are all being hidden (I still refuse to use nautilus; I need to find a way of disabling it to save memory). The geek-argument is quickly eroding.
-Michael
Don't know if I agree.. A capitalist is merely concerned with maximizing profit. From a purely mathematical / logical point of view, any and everything is a possible course of action.. The only constraints are the physical world, which includes artificial policing. e.g., from a business point of view, what's the most I can "get away with". Bribe, and cut-backs have long been a business practice (long before the evil yet legitimate practices of MS). Hell, just like at any city where the Maphia has influence. Capitalism is really survival of the fittest, and that often means those that are able to think outside the box will win. (note that it's not the same as acting outside the box).
So in summary, competition involves everything including murder (just think about how capitalism works when you're stuck on a mountain side with only a small ration of food and 10 starving people). Jail is just as financially bad as bankrupsy, so illicit activity followed by conviction isn't any more of a risk to investors than betting the farm on some last-ditch scheme before being squashed by your competition.
Does this justify it? Course not.. That's why we have the publicly funded SEC, Trade commission, police, etc.
-Michael
Basically, greed got in the way of capitalism.
Just a small nit-pick here.. Capitalism is based on rational decisions. Those decisions are based on supposedly practical values. The most practical value is self preservation. Thus an essential element in capitalism is that one will do what is best for themself. If any decision means that throughout the course of one's life, they will be better off doing something that, but that action hurts society as a whole, they will definately choose the selfish direction. It is irrational (non-optimal decision making) to do otherwise.. It is only when hurting society will eventually hurt them in the long run (reigning anarchy where they will most likely die at a young age), that an individual must choose non-selfishly.
Greed is a hard word to quantify. You could say it's an obsessive level of consumption at the expense of others. But it's hard to differentiate this from rational action. I'd rather characterize it as a compulsory (obsessive) behavior, where optimal decision (e.g. rationalism) are not made. To promote a consumption (including mere acrument of assets) whose' diminishing marginal returns are not considered. In other words, since each additional dollar should have less value than preceeding ones, there should be a point where other elements of one's personal life are more valuably persued that that of the next dollar. Obsession is where we persue this devalued dollar in spite of other needs.
The problem is that while a human being is ill suited to make use of that last made dollar, a company isn't. In fact, the opposite is true. Each additional dollar made is a vehicle for persuite of even more productivity. It therefore has an exponential effect. Remember, so long as hurting others does not hurt yourself in the long-run (including possibly your decendents), then it is still rational to act selfishly. Thus a monopolistic company that squashes competition and locks every man-woman-and-child into a sort of tax (due to annually required upgrades), then enforces license compliance, then the raising of prices, hurts the hell out of external cultures and organizations (countries even), BUT still benefits in the end. Even if they are boycotted, affected with ledgislation, they will still have acquired more total present-day-dollars, then if they'd constrained themselves.
This is efficient and rational capitalism.
The problem is that MS should not exist as it does today according to classical capitalistic theory. The ONLY reason MS exists today as a monopoly is because of copyright law. A "tool" emposed to artifically help capitalism. MS does not produce windows more efficiently than a potential competitor. MS does not write database code that is better than competitors (nor even more cheaply). It merely has a strangle-hold thanks to it's artifically enforced licences and intellectual property. Non-disclosure-agreements, and non-terminating copyright labels stand in the way of "effecient" production of the computer software good.
Any society that has to ammend it's fundamental tenent (such as American freedom, squelched by business-bolstering copyright/patent-law) obviously demonstrates it's lack of fundamental viability.
If Capitalism can't exist without artifically constraining laws, then capitalism is a failure (if not today, then some-day). Personally, I don't think capitalism is a failure. I think that in light of modern work-paradigms, it would actually be possible to live in a capitalistic society free of copyright and patent law (of course this even means GPL would have to go). It throws away work-for-profit mentality in exchance for work-for-compensation. It would still be possible to profiteer, but you wouldn't be able to rest on your laurals.. You'd have to have a business plan where you could at least for a time, be more efficient at production of a good/service than any free-rider. If this is so strange, then look at what we say must happen with the music industry. We say that they have to wake up and accept the change. But it's no different then saying that software or medical drugs would have to give up manufacturing royalties (and thereby hurt venture capitalism).
Who then would bother spending billion dollars on medical research, or on a nuclear power-plant management operating system, if as soon as their done, a generic company will sell at bottom dollar, and thus equate a massive loss. I think the answer is very obvious. If people are afflicted with a massive disease, and we need to spend a billion dollars, then people will demand sanctuary from the government, which will therefore grant money to an organization which would then hopefully find a solution. The findings (successful or not) would be open to other organizations (since it was publicly funded). You're paying a "service" fee. A nuclear power-plant would likewise have enough value to pay a company to make custom Nuclear-power-compliant software. That the next power plant could then steal the code and use it at no additional cost is irrelevant.. The first nuclear power plant couldn't operate otherwise. Further, the second plant would have to pay service-fees to reverse engineer and recustomize the code to fit their need. The actual reduction in cost to the second plant would actually not be substantial.
If tomorrow we abolished copyright law, the first thing that would happen would be that stock prices would go rock-bottom. Share-holders would know that their IP-based profits would dissapear. Especially in the medical and computer fields where the information is most of the final result. But these companies wouldn't disssapear. First, they'd spend billions of dollars on litigation to recover their previous nest-eggs.. Many companies would go belly up in this last-ditch effort (assuming they weren't successful, which is irrelevent to this discussion). Some would simply adapt. They'd convert to a truely-free-market economy (e.g. migrating to service or minimial-cost production, or marketing (such as inferior brands who's label is all that matters; e.g. the textile industry)).
All commercial software companies would go belly up immediately, since it would no longer be illegal to simply take a single MS CD and install it on every machine in the world. The only thing MS could do would be to enforce registration (which is fine by darwinean capitalistic ideology), but this wouldn't last since hacked MS products would eventually appear which thwart registration attempts.
Strangely enough, this wouldn't hurt society as a whole because we already have sufficient advances in technology and medicine to survive. What would definately hurt would be the economy. But that would only be temporary.
It's true that we'd have a stagnent economy for a while as we learned to shift our resources from high-risk venture capital to other methods, and it's likely that we couldn't achieve the same order of magnitude financial capability as with constrained capitalism, but we have to look at the larger picture. As a society (and thereby a governing body such as congress), what are we trying to achieve? Growth? That's a selfish endeavor; not something society as a whole should be concerned about. Do we really care that we have more people next year than we do today? Do we really care that the total number of dollars in our society grows each year (even though our personal stock-piles are stagnent with respect to inflation). Do we really care that one country grows exponentially with respect to all other countries in the world? (A popular bastardization of capitalism is that you must not only compete with your peers, but measurably squash them). I say that these things should not matter. What I consider altruistic measures for a ledgislature are the accomplishment of zero non-transferring unemployment (meaning the only unemployment in a society should be that of those seeking to change their current work-status). If everyone is successfully employed and inflation is marginal, then times are good. If we move from a venture capitalistic society to a service-based society, then we obviously have a demand for employement. Further, artificial weath (as from get-rich schemes based on IP; e.g. many forms of venture capitalism) is reduced which dramatically curtails inflation. While I can't currently prove it, I am under the belief that inflation is dramatically reduced when you reduce the number of monopolies that are paracitic to the general population. While some monopolies are enevitable (telephone / cable industry, etc), the fewer monopolies you have (assuming competative markets can replace them), the less excess each citizen must pay (eaten from economic welfare), and thus the less it costs to live. A reduction in the number of forces encouraging higher costs of living should constrain inflation.
I'd be interested to hear further arguments for or against a free-market economy (again, I'd like to reinforce that we don't live in one). I know full well that no politician would cut off his own neck in suggesting curtailing IP-laws, but it doesn't hurt to contemplate such a society.
-Michael
Open Source software is insanely susceptible to hackers/bugs/etc. Moreso than even M$.
Explain then how there have been few viruses that have had success against linux/BSD servers, while IIS + winX has a new security headline weekly. You could make the argument that the proliferation of Windows machines makes them more suseptible to an attack, and therefore if Linux was 99% of the market, we'd see few Windows viruses and lots of Linux viruses.. But I'm talking about servers where Apache has a >50% share (at least at one point in time). Apache is recognized as one of the most robust services out there, and it's obviously a critical piece of the cracker puzzle. OpenSSL + OpenApache + OpenLinux (with and open firewall design) are extremely robust (though not impenetrible). Most of the security holes that I'm aware of deal with user-level applications. If you're trying to hack a bank, you can be sure that user-level applications will not be a factor. Should a bank use winXP Pro + IIS then? (Sadly I know a few that do).
I just couldn't sit by with such an unsubstantiated comment went by.
-Michael
Once apon a time they had low prices. But not anymore, and your statement is becomming increasingly false. As their software actually gets close to companies like Oracle / IBM in terms of capability and robustness, their prices are reflecting it. Don't forget to take into account upgrading into the cost. A nice expensive SUN hardware work-station has been running the same version of Sybase for years at our company, and there hasn't been a compelling reason to upgrade; that's thanks to the mostly settled UNIX standard. MS APIs are ever-forward looking and thus obsoleting of previous standards (read interpolibility and network security methods).
I fully acknowledge that MS has contributed a lot, and provided a realistic alternative to UNIX / custom-solutions. Such innovations as MS Works, MS Access, and virus-friendly do-anything office-suites have contributed leaps and bounds to the industry. (note my synicism)
The problem is the same as if you had to choose which company to work for. Company A has low pay, marginal benifits, not particularly exciting projects, but have open and moral business practices. Company B is very radical, very growth-oriented; it has a lot of flash and excitement. It's already number one in two or three fields, and it has it's sights set on dominating other markets. The business men are ruthleess, smart, and know how to turn a buck.
Well, it would seem that most people would prefer Company B, if for nothing else than it's like a shiny toy to a child. But, risk-takers, statistically fail more often - including the probability that illicit legal affairs go on. Look at Enron, MS (by all rights, if Bush hadn't stepped in, their stock would be significantly less valuable today as they'd have been broken up like the Baby Bells), etc. Ruthlessness is a sure-sign of anything goes, which is another word for morally-indifferent. Such action works great for the here-and-now, but says nothing about one's future.
Thus the conservative employee would be best advised to choose company A, and do their best to make it competative. While in our Attention Deficit Disorder society (including stock-investors), it's entirely possible that good companies will be jettisend in favor of the Blitzkriek companies, you still have a good measure of reliability.
The same sort of comparison SHOULD go on when choosen business associates (including who's administrative products you purchase). You wouldn't purchase a database from a company that you've never heard of; least you eventually lose support. Likewise, dealing with a company who's made a publicly known habit of coersing it's customers (strong arming IBM, schools, etc via maliscious audit-threats) whenever they determine probable gain is a definate factor in the risk-analysis.
For many of us "heretical slashdotters", the risk-assesment is at the forefront of our minds. We stand to lose x,y,z as we chant our mantra over and over. That common businesses don't always take these factors into consideration is what we really fight. That they decided that the risk-cost isn't higher than other opportunity costs, is fine; so long as it is part of their calculation.
-Michael
I assume you're talking about the velocity of money (where-apon a dollar is respent many times over, so long as it never sits still). While it's true that investing money in an institutions which immediately re-issues can produce a >1.0 velocity during average to strong economic times, when the economy eventually falters (which it didn't really do during our little American recession), that multiplier contracts and can go < 1.0. At which point, the artificially inflated money suddenly dissapears.
The larger the mulitplier, the more it hurts during a real recession.
As part of the multiplier, is the idea of financial float; checks are written that are immediately deposited by the federal reserve into bank A, but funds are not immediately withdrawn from bank B. Thus extra money temporarily exists. So long as the number of transactional cash-flow remains constant, this temporarily created money sticks around. Once a recession hits, purchasing slows/stops, and checks clear...
Since recessions are cyclical, anything that artificially heightens an economic boom will effectively over-stear the economy and make the ensuing correction more powerful.
I personally believe the 90's represented an incredible growth of efficiency in our economy. Banks were allowed greater freedoms, computers reduced transactional costs, demand shifted from expensive goods to high-profit-margin-goods (like software solutions, and raw experienced human labor such as IT and effective middle-management). Since people were generally required to do more, their higher pay was justified; we simply did more work per day than we did in the 80's as a whole.
I believe this growth allowed us to avoid a catastrophic recession / depression on our economic correction. Aside from another socio-technological revolution, we can't sustain such growth as in the 90's, and thus such augmented multipliers are not ultimately to our benifit.
Disclaimer: I've taken several undergraduate economics courses, but I don't claim to be an expert, so comment-accordingly.
-Michael
Damnit.. Forgot to append the best link in the world that describes this in detail.
Britney Spearse Guide to semiconductors
Well, I'm an undergraduate Electrical Engineer, so I only have superficial understandings of how semi-conductors interact with light, but it doesn't seem too great a stretch of the imagination.
First, semi-conductors work based on the principle of the band-gap (which they even mentioned) (correct me if I'm wrong with any of this, I'm doing it straight from rusty memory).
A little background:
The outer 8 electrons held by an atom are the most important (the valence) - They are responsible for the bonding of other atoms. The configuration of all the electron orbitals in free space is nicely geometric; the first two electrons form a spherical shell (s-shell), the second 6 form dumb-bells in each of three axis's (p-shell). These types of configurations affect the geometries of the connection of the atoms. Configurations get more complex as the number of electrons grow (which is somewhat independent of the atomic number (number of protons), but such ionized atoms are unstable; especially when the number of electrons differs dramatically from the #protons). The important thing to understand here is that each additional electron takes more energy. Instead of worrying about the geometries, you can plot each electron orbital at a different (successively higher) energy level. Different atoms (characterized by atomic-number and even, to a small degree, the number of neutrons present), have differing characteristic energy-levels. The discrete nature of atoms includes the probabilistic nature wherein electrons have an extremely high probability of occupying the exact energy levels (which can be thought of as the distance away from the center of the nucleus). There is a chance that an electron will pass through any point around the shell of an atom, but it's highly unlikely that it will deviate from its characteristic point.
But, since different atoms have different characteristic levels, warping an atom will warp its points. Warping can occur by simply placing two atoms near each other (such as in an ionic or covalent bond). As it happens, when you squeeze atoms closer and closer together, the discrete lines that represent the energy levels start to merge together. Eventually the 8 outer valence bands merge into one continuous band... As you squeeze them even closer together, this band breaks into two continuous pieces. As you get even closer together, these pieces get further and further apart (I would presume that eventually one of these bands starts to merge with preceding energy levels, but that's not relevant here). This gap of continuous energy levels is called the band-gap.
As it turns out, in perfectly bonded atoms (those where every electron in the valence layer are bonded, and each atom has exactly 8 outer electrons; such as carbon, Silicon, etc) we have a total of 4 electrons that fill the inner continuous shell and 4 electrons that are void in the outer continuous shell. BUT, that outer shell is looped across neighboring atoms. When a diamond-lattice is organized (which is as close as you can possible get multiple atoms to sit next to each other), you have the greatest band-gap you can get for that particular element. Different elements (or even molecules) that can form the diamond-lattice will have differing characteristic band-gaps. What we have here are 4 electrons that are tightly tied to a core atom, and 4 potentially absorbed electrons that can freely be shared across every single atom in the entire crystalline lattice. In semi-conductor crystals, the problem is that every electron is accounted for so there are no free electrons to put into the outer band (which could roam free as current through an almost zero-resistance substrate; due mostly to quantum effects). Impurities are therefore inserted into the crystalline lattice which act as ionic donators of electrons or ionic acceptors of electrons (namely atoms not in the 4-column of the periodic table). Thermal excitation (heat) causes an electron to be ripped from donor atoms and those which are then quickly swept up in the outer-most continuous band.
Normally, electrons must have a precise energy-value in order to live in an atomic orbital. When an atom absorbs an electron, it gives off a photon of the remainder of the energy. To change orbital-levels, it has to accept a photon of exactly the correct amount of energy. It can accept a larger energy photon, but it will again give off the remainder of energy. Eventually that excited electron will fall back to its lower energy level, giving off another photon which will have the exact energy as the distance between the two energy levels.
In the continuous region of these silicon atoms, excitation between energy levels isn't apparent, since an electron can have any value within the region. The only difference is that separating the band gap... An electron from the inside can jump into the outer band if it's given at least enough energy to make the jump... This gap is usually enormous for semi-conductors. I believe its 1.2 electron Volts for Silicon, and 10 electron Volts for Carbon. The 1.2V is within the range of thermal excitation. That means that heat (in the form of vibrating atoms in the crystal) is enough to shake an electron free; e.g. jump the gap (like water successfully spitting to the lid of a boiling pot). In carbon, however, room-temperature heat is no where near enough to make the jump. This property (along with others) is why we don't use carbon-based semi-conductors. Germanium and silicon are much more practical in our particular earth climate.
There is another aspect to the band-gap that is relevant to our discussion. Each electron has not only an associated energy, but a quantum-form of momentum. You must not only have conservation of energy, but conservation of momentum. I'm a little fuzzy on this topic, but this momentum is represented by the letter k, and we can plot energy verses k for different things. For semi-conductors, we get parabolas, and inverted parabolas, but with discrete points. This says that while we have a continuous set of energy levels within a region, we have only a certain set of allowable energy+momentum values for the electrons. And like the discrete energy levels of atomic orbitals, you can only have one electron occupying a given state. In a rather unfulfilling way, I'll stop talking about things that I don't fully understand and simply say that this multitude of characteristic parabolas says that in order to have an electron jump, you have to not only have a precise amount of energy absorbed or emitted, but you have to be able to transition your momentum somehow. Energy transition occurs through photons, and momentum transition occurs through phonons - which is energy present in lattice vibrations (e.g. packets of heat).
Gallium arsenide is an example where the lowest point of the upper parabola and the highest point of the lower inverted parabola are aligned with respect to momentum. This means that the smallest amount of energy needed to make an electron jump the gap requires zero change in momentum. Because of this, gallium arsenide crystals easily will easily absorb or emit light with no dependence on the heat of a lattice. For this and other reasons, GaAs is great for laser diodes. Silicon, on the other hand requires a momentum change for its lowest energy transfer. Thus lots of heat is generated and absorbed; (Not to mention that silicon doesn't conduct heat as well as some other semi-conductors).
Given this superficial description, what I get out of this is that heat of a certain resonant point (in the form of vibrating atoms in the crystal) could provide the proper momentum shift needed for efficient electron excitation. You'd still need to provide photonic energy for the transition, but you'd have a perfect combination of heat + light absorption. Eventually (due to statistical decay), the electrons would fall back to their lower-energy-level states. But they'd give off light of specific frequencies.
Putting all this together, my initial impression from the article was that that the tungsten injection into a silicon substrate change the characteristic e-k curves enough to absorb the phonon-heat generated by IR light. The result is a 60% efficient absorption of the heat + light (e.g. nearly perfect efficiency). That energy is retransmitted as diode light (e.g. an exact energy level transition, producing a constant level of energy photons, which requires an equally constant frequency of light).
What I don't know at the moment is if this is actually emitting mono-chromatic light, or if a multitude of frequencies (e.g. white-light) permeates. The only way I could see white-light emitting is if the standard tungsten light-bulb is making it, and the Tungsten semi-conductor is amplifying a particular frequency.
My understanding is that they said that IR-frequencies are synonymous with "heat". They frequently used the term black-body radiation. I remember IR-HEAT being associated with green-house effects; the angle of refraction is low for IR and glass, for example. So when sunlight enters your car (at a direct angle), it bounces off things but hits the glass on the inside at too great of an angle, and thus bounces back inwards, amplifying the total heat.
Not that I'm satisfactorily answering your question, but throwing out some food for thought.
-Michael
Yes, I believe so. Foreign keys are implemented with triggers.
I know that you can embed various languages into postgres (like perl), but I've rather avoided most of it (due to proprietary nature).
-Michael
Even better, perl has several SQL-emulators that can work on flat files or berkley DBs. Using this approach, you could apply portable complex queries / relational constraints easily to legacy data.
-Michael
Well, that assumes that the sound-cards themselves don't hold keys like DVD players and thus the audio format of the furture isn't encrypted with audio cards exclusively being able to decrypt them.
But, I like how you think.
-Michael
Nothing magical about 8bits, except that we like multiples of 8. An 8bit add requires two 8bit numbers = 16input bits, and 8 output bits. Ignoring optimization and implementation complexities, that's on the order of 65 thousand transistors for a two-level adder. That's definately do-able in today's technology.
Two 16bit numbers on the other hand, would likewise require on the order of 4 billion transistors, which it not currently possible with today's technology (my guess is that by the time it is, we'll be using 128bit ALU's).
My point was to find the most practical pure combinational logic unit. Thus to suggest a 1-bit machine implies that we miscommunicated.
An alpha has a nice little byte-swapper which relatively efficiently extracts a byte out of a multi-byte word. It's computationally slower to work with than just testing the whole word, but when you take into account bandwidth / cache-space considerations, it's a boon. It can't be easy for a compiler to optimize away:
struct {
int f_flag1, f_flag2;
}
Can it drop the flags down to 1 byte?
While it's not always true for each architecture, you should find higher performance swapping out the int's with char's.
I'm not familiar enough with the state-of-the-art in compiler technology, so perhaps this can be determined. I was just generalizing on this type of point, nothing more.
The points you mentioned here would be included in the sort of tricks that have been devised over the years to avoid upping the word. It just furthers the idea that larger word-sizes are expensive and worth side-stepping.
-Michael
Very simple. Lack of competition.. They held a monopoly on ALL OS / motherboards. The only real competition that I'm aware of were the 3'rd parties that sold the various chips / expansion cards.
When you have a vertical monopoly like that, you can coordinate an architectural shift. If Intel decided to start a new CPU line, and it turned out to not provide the best bang-for-the-buck, then AMD/Cyrix competitors could supply legacy and current MS-products a better alternative. Intel would have lost all that money. From the other side, MS is spread so thin that they don't have the time to rework their core to be optimal on multiple platforms (look at the death of NT on any non x86 platform). The lack of a compelling reason for someone to purcase the alternative platform says that MS shouldn't devote too many resources in that direction, which of course kills it off. Hense platform architects are at the mercy of software people, who must provide killer apps for that platform.. If any major killer app isn't immediately available, then a domino effect of lost support will occur; and more importantly, business people understand this a priori.
This is actually a lot more exciting then it might first appear. On the one hand, you have a controlling hand-of-God who enforces their will. So long as they can project a bottom line that benifits their customer, they can make radical changes (shedding virtually all of it's former self). On the other hand, we have multiple independant organizations, who each act in their own best interest. In monopolized environments, changes are swift and clean (but not always in the best interests of everyone). In independent environments, no organization can squander or otherwise take too great a risk. Efficiency is upheld, since only rational decisions can be made (involving the mutual benifit of progress). The side effects are a slowing of evolution, and an accumulation of "useless appendages". On the other hand, it provides an incredible level of trust on everbody's part that the architecture has staying power; that it'll weather the storm of change, instead of flippantly changing with the current mood; throwing 3'rd party interests aside when it's convinient (read Apple's resinding licences over the years).
Personally, I think an architecture that has "grown" over the years is more remarkable then one that simple borrows the best ideas that come out of universities.
Note that I'm not really advocating x86's. I'm just admiring it's successful history. The proliferation of UNIX and the general ability to recompile source could theoretically alleviate a "better" platforms' lack of killer apps, and thus perpetuate a radical acceleration of architectural designs.. Go Linux!!
-Michael
I disagree from an architectural standpoint. In an ideal world, we'd all have 8-bit machines. All our arithmatic would be insanely fast; we'd be able to use combinational logic to allow two probagation levels for ANY operation (add, sub, mul, div, sqrt, log, etc). That's because it's cost effective to do so; a minimal set of possible outcomes. I'm not completely sure, but I'll speculate that it's possible to arbitrarily generate an arbitrarily sized number from just these 8 bits; though most likely it would be programatically (even if done via micro-code), and thus would be non-optimal for larger than 8-bit data-sets. So obviously, as we've been able to, we've increased the data-length throughout history as we've demonstrated a need.
Contrary to the impression that's given in these posts, a larger word size fundamentally is slower in calculating smaller values. Sticking with higher performance two-stage combinational logic requires an exponentially increasing number of transistors. Breaking the logic up into tiers allows designers to trade the number of transistors for the number of probagation delays. The more delays, the slower the clock; the more transistors, the less practical the design (due to heat, cost, and feasibility of fabrication). Pipelining somewhat helps aleviate the issue of extreme probagation delay, but it's impossible to achieve 100% efficiency, and thus you're practically garunteed slower operation for deeper pipelines. What's more, pipelining requires additional probagation layers for buffering, so you take an immediate performance hit; speculating that you'll achieve greater over-all performance.
In an ideal architecture, you'd minimize the probagation delays for each instructional unit, but practical measures say you must group most, if not all, of the CPU such that the slowest part drives the system. (P4's are nice in that they sub-divide the clock for the simpler integer units).
Combining the two ends, we can better appretiate the trade-off.. If we're performing large-valued arithmetic which is slow programatically (emulated 64bit), then it's worth the extra cost (towards the speed of each operation, and in terms of the number of transistors). In other words, one hardware 64bit add is most certainly faster than than several assembly language instructions that piece together 32bit values. BUT, now all your 32bit arithmetic is slower (unless you have separate 32bit/64bit logic cores).
It's possible to design 32/64bit cores that only take as many clock-ticks to complete as necessary, and thus 32bit arith isn't horribly slower, but there are definately additional probagation delays. The augmentation to 64bit can never increase the speed of a 32bit operation. (Any speed ups must be due to over-all advances in computational efficiency, which should benifit a pure 32bit core even more).
The trade-off must then be a statistical one. We cost out the largest word size that provides benifit. You're going to have arbitrarily large ALU operations (just look at encryption), so choose a cutoff where a certain percentage of all operations occur at that high of a word size. This is how we moved from 8 to 16 bit, and then the painful shift from 16 to 32 bits. And for server-targeted machines, the shift has already been cost-out to adopt 64bits. The desktop has not yet made sufficient requirements to adopt 64bits, though the underlying x86 CPUs are being shared in server-space which is nudging 64bit's acceptance.
Another important factor (which is presumably obvious in concept) is that a higher word-size has a greater probability of wasted space. A 1-bit boolean, for example, wasts 63bits.. Booleans are very common, and though they can easily be consolidated in c-struct's, such is rarely the case, since there are memory alignment issues (and flat-out laziness on the part of programmers). The wasted word-space also affects the instructions. Rarely do you actually see 64bit aligned CPU-instructions (except in VLIW or in places that the data-word-size was irrelevant). Such a situation would have massive implications towards performance. But one serious consideration is that the population of 64bit constants using a 32bit instructional word is expensive. Now you have to perform at least 3 (probably 4 or 5) instructions just to load a constant. Suddenly "a++" starts to look scary (at least when non-optimal compilers are used). In all cases sub-word-size'd instructional arguments are permissable to the delight of compiler designers, but there are still classes of problems that thwart this.. Namely memory addressing...
Memory addressing is arguably the strongest supporter of 64bit architectures. The 4GB limit is already apon us on desk-top machines (I have half a gig on all my home machines, and I don't need it). When you add swap-space, it's entirely possible for modern desk-tops to run enough apps to desire 4+Gig of memory. (Especially considering that large chunks of the address space are wasted). Aside from the various tricks designers have employed over the years to avoid augmenting the address space (8086's segment-registers, 80386's segment-selectors, OS's swapping out apps completely from memory, etc), it's arguably slower to emulate larger address spaces.
In addition to the above arguments against larger address spaces, there is massive cache polution; doubling the word-length, literrally halves the usefulness of a cache-line-load, unless you were previously emulating a larger word-size. You can only load 4 words on a pentium-class cache-line-fill instead of 8. Your bandwidth requirements litterally double (unless you don't standardize at one word-length).
Now in contrast, there are a few advantages. If your minimal word-size is larger, then the number of address pins that you need are reduced. But this is really independent of the core word-size. Pentiums have long required 64bits for their external bus, and use even larger cache-line sizes. Thus most of the advantages attributed to this argument are moot.
Theoretically, an architecture can be designed to split an ALU such that it acts as either 1 64bit unit or 2 32bit units. This is especially true for vector-cores (which are already up to 128bits for main-stream processors). In general, however, there is still the trade-off here, since additional logic-probagations are required which slow down the general case of only a single But comparing the X-box and the Nintendo 64, which were released many years apart won't buy you much of a conclusion other than current processors are generally faster than older processors.
I'd like to address the nature of large bit-sizes with respect to graphics. While this isn't my expertise to the extent of the above, this primarily affects the bus width. In graphics, you commonly have multi-integer structures (red, blue, green, alpha (opacity), Z-depth (the depth into the screen the geometrical object that drew this dot is), stencil, etc). The entire structure is usually just 16bits, 32bits, 64bits, etc. The larger the structure, both the more features you can pack into it, and the more accurate each individual number can be. Thus saying that an architecture is 128bits purely based on this is very misleading. What's even worse are labeling the bus-width numbers (e.g. 128, 256). That's like calling the Pentium I a 64bit CPU, just because it has a 64bit bus (used purely for cache-line burst fills). Yes it makes it go faster, but so does shortening the length of each wire; it's not really innovative. I'll throw this in, but I'm starting to get in over my head; Graphics units (especially the filtering parts) make heavy use of hard-wired combinational logic units. The number of bits going into these units is really meaningless (how many thousands of wires go into the control logic portion of a CPU?). Thus the ability of a custom pieces of hardware to utilize larger bit-depths is unimpressive. What would be impressive would be to say that a Graphics unit does it's integer / floating arithmetic in 128bits so as to minimize error (even though the input/output might only be 8 or 16bits per atomic unit).
-Michael
Course congressmen know full well that they'd lose their barganing chips; the I'll scratch your back if you scratch mine. The cornerstone of US democracy - compromise.
I remember cheering when the president got line-item veto power, and thus could signify the death of pork. Unfortunately this gave the president too much power, and was overturned.
Still, somehow a campaign finance bill was recently passed so anything could happen.
-Michael
So does that mean it has 3.6 Terabytes of swap space?
Cute, but swap is optional as far as I know.. When you have enough memory, or when you want real-time (at least in solaris or qnx), then simply leave it out.
-Michael
Thank you for being professional.
There was once a project to embed a RAMBUS controller into an intel CPU. I was hesitent to post the above because I wasn't 100% sure that the P4 indeed had it. Unfortunately I don't have time to research and find out what article (from over a year ago) prompted this.
While it's blatently obvious that embedding motherboard components will make that particular implementation faster, it does little to allow future advancesments. In what-ever article I did read about the RAMBUS controller, it didn't say whether the feature could be bypassed so as to utilize a more modern controller somewhere down the line. I remember feeling frustrated over that matter. What if CPU's came out with AGP 2x controllers in them when AGP first started. It would have slowed progression of AGP 4 and now 8x, simply because most cards would be sold to existing 2x-based systems.
I don't have a problem with embedding technology, so long as it can be upgraded (see the nForce chipset). Likewise with MS's IE, MediaPlayer or MSIM. I have no problem with them putting it on their OS, so long as I can 100% disable it if/when I choose Mozilla / winamp / AIM. Unfortunately I don't know how well you can shut down media player in XP; especially given XP's alledged taddle-tale role. Not to mention the insult of having MSIM reactivate when viewing hotmail.
What's more, I'm sure that such goodies in the windows line encourage them to jack the price up. This is the problem.. We're paying for a bundle of goods that we can't fully use, and we're not given the option of purchasing a streamed line version at a lower rate. (e.g. OEMs could ship winamp + Mozilla + Macaffe and shave $10 - $50 bucks off the OS).
-Michael
You must have missed the part where he said "ftp.exe". e.g. the context was inexperienced "windows" users.
-Michael