while(!Slashdot.Readers.economic_knowledge) { printf(" I think a lot of confusion vis-a-vis viewing the economy as "Zero-Sum" comes into play if you only view money as the tried and true measure of all things economical. I get $X via my labor, and I pay $X for some widgets, and from the perspective of the market, there's been a zero sum. Or so Sengan would have you believe.
Even this is wrong, taking into account inflation, money demand, and the interest rate. Even if money is our only benchmark, the value of money changes. It's so wrong that there's an entire branch of macroeconomics studying money market fluctuations..
Knowing that economics is not* a zero-sum game if we just considered money the benchmark, let's move on to consider social cost pricing**. The intrinsic worth of a transaction is based on: 1.how much one values the object 2. the social impact of having it produced, consumed, and changing hands 3. and various X-factors which I don't feel like listing
In short -- Comparative Advantage is what matters in any given transaction, not Absolute Advantage. A common analogue would be: If I had all the gold and you had all the hamburgers, sure, the gold is more valuable than the hamburgers normally but King Midas has got to eat sometime!
This isn't even taking into account that technology, capital goods procurement, and an expanding labor market increase production as well. Open source fits here quite well, i posit. Nowhere do I have to pay you for your labor in creating Open Source Software. Presumably, the social impact of using this software pays for itself. (No Microsoft Tax on OEM hardware if Linux becomes mainstream, that sort of thing.) On the flip side of things, if the social impact of totally free software becomes cumbersome (net traffic goes to a standstill every time a kernel is released) then the market will attempt to seek a solution (people will fork out $5 for CD-ROM of Linux).***
It is true that no one can truly know everyone's preferences, or measure how much impact social impact is.**** That's why people are tempted to view economics solely in terms of money. Do not be lulled into this trap, however! Armed with basic axioms of consumer preference and an entire branch of statistical economics -- Econometrics -- we can massage cross-sectional and time-series data to infer preferences. Such preferences deny a zero-sum game.
Besides letting us use phrases like "running a white test on a heteroskedasticity-consistent variance-covariance matrix in order to run a robust first-order regression,"***** econometrical analysis allows us to measure the cost of objects in terms of their correlative effects on human behavior. Here's where the zero-sum idea breaks down...
A zero-sum game wouldn't even need a regression line. Because there is no net gain, the correlative impact of X on Y or Z is nil -- do whatever you want, you're going to end up screwed anyway. Given the fact that there is enough variation in human economic behavior to require probability and autocorrelation tests, it's fairly self evident that X has an effect on Y and Z and vice versa.
If X and Y and Z have correlative effects on each other, it means the following:
1. Said economic transaction is not a zero-sum game, because the elements are not independent. 2. Non-Independence means that the variables are covariant. 3. Covariant variables tend to follow one or more of the Gauss-Markov assumptions about spherical disturbances ******. This means that the changes are stimulated in part by "outside forces," known as "Error terms," "disturbances," etc. 4. A zero sum game has no disturbances because there is nothing to disturb. 5. Technology, capital gains, political climate, social change, etc. count as "Disturbances" if they are infrequent or weak enough, else they count as variables of their own. (in which case they are covariant as above) 6. I enjoy saying the phrase "heteroskedasticity-consistent variance-covariance matrix." =P 7. Q.E.D. (sorta)
Discuss amongst yourselves.:)
Whew. That was quite rant. \n"); } /* sorry folks, I needed to let out some steam =) */
*(and probably never will be, except in peer-to-peer oligopoly transactions) **(Campbell and Pereira, 1998) *** Will the software still be free? Yes. But general procurement will not be. Of course, I have yet to find a person with free Internet access for those downloads... **** Not 100% true. There is such a thing as experimental economics. Most people's preferences fit within Additive, Leontiff, or Cobb-Douglas utility functions (for substitutes, compliments, and not-quite-either goods respectively) ***** (Pindyck and Rubenfeld, 1997) ****** Even if they don't, they follow the assumptions if you do a robust regression. A robust regression replaces the linear approximation of trends with a taylor series expansion of an nth-degree polynomial fitting the data.
/* Imagine: Department of Computer Usage. Licensing test -- open a (Microsoft) word processor, compose and send an e-mail (using Outlook). Find the DCU's homepage using Internet Explorer. Three misclicks and out you go... And then: "computer usage is a privilege, not a right!". Two speeding convictions on the Internet and your license is suspended! */
Heh! Well, the United States tend to have low state standards for drivers licenses anyway! This would be very in line for us.
Seriously, thought, there is a line between the human urge to maximize utility now (luxury, simplicity, ease) and maximize utility later (freedom, liberty, knowledge). This is going to be very atypical for slashdot -- I don't know where to draw that line! (He admitted he does not know! Dear God!)
People care about more results, alas, then the quality and freedom of code. And, sadly enough, people would probably sell their fundamental liberties (warning: USA bias) to go watch TV.
I actually agree wholeheartedly. But most end users are going to be using the machine to do things like surf the world wide web, and word processing. Very rarely will they do much more, not because they are incapable intellectually, but because such fields are not yet in the public interest.
Personally, I'd like to create a licensing program for computer usage, just like drivers must be licensed. But it took many years of chaotic driving from unlicensed drivers about the turn of the century to convince people that it was worth licensing. This is social cost pricing. To implement an idiot-check for computer users now would be doomed to fail. As unproductive as widespread computer use is now, it has to get really downright annoying before people will submit to changes.
Sorry guys, but computers are a tool. A general purpose tool, yes, but a tool. A steep learning curve is counterintuitive to using a tool. The end user has paid for this computer + service, and expects to be able to use it.
To say that cluebies should be kept away from computers puts a nail in the "open source can still make money from technical support" argument. Have it one way or another. Or do you believe that you can support Linux easily by telling inexperienced users to go to hell?
A car is a Tool. It gets you from point A to point B. Nowhere do we specify which points these are. A car can take you to the store or across the country. You can store stuff in your car. If you are camping, or on a road trip, or homeless, or lazy, you can sleep in your car. You can listen to music in your car. If you have a carphone, you can carry on conversations in your car.
It's ridiculous to say that if you are too clueless to re-pack your engine bearings, you shouldn't be supported. The world does not operated on a roll-your-own system. The joy of Linux, *BSD, and the like is that it gives you the freedom to do so, if you need to. Living your entire life on a roll-your-own basis is not the way for everyone. Accept this.
A delicious point of irony to the people who wrote the feature in question: Perhaps the newbie they refuse to help with difficult (yet documented) computer problems is the tax lawyer who will refuse to help them with difficult (yet documented) tax problems.
"Oh, I'm sorry, you don't understand Title XI, Section 2, SubSection 3, Paragraph 3.5-4.2? RTFM! thomas.loc.gov!!!"
Heh. Sounds ridiculous on the other side of the fence, doesn't it? Sounds like a bit of geek rage rather than a logical assessment of the end user.
"They made fun of me in high school, but now Linux is popular,
gollum gollum, yesss... mustn't let the end user have an easy time, precious.... yesssss...."
I've been working Unix tech support for close on two years now and believe it or not -- it's been a pleasure. Some end users are idiots, some are irate, but some are people who don't have the time to solve documented but pain-in-the-ass problems. I help them, and occasionally, they help me. Welcome to the world of social interaction. =)
"You mean if we're
nice to them, they'll buy us stuff?" -- PCU
Yikes! I hit the "preview" button, fixed my problems, and lo and behold they still existed after submission! Sorry -- sinator
The econ/CS major's $0.02:
Sorry to rain on your parade here, but why exactly is this so great? The fundamental concept that everyone seems to be missing is that Windows sucks. I don't want to see different companies produce it. I don't want WINE to emulate it. I don't want to see anyone make, use, or sell it. EVER. I want it to die. Not because I hate Micro$oft (after all, this is about somebody else selling it), but because it's a bad idea. The concept of a brutally complex, single-user operating system - term used loosely - is a bad, bad idea. Better to kill it immediately than to prolong the pain, even if it is supposedly at Micro$oft's expense.
Such short sightedness. Remember, opening up the source lets people *fix* things. This same imposition was made regarding the magtape distributions of UNIX System III and V from AT&T. Opening up the source helps. If the source to Windows is opened (including Windows NT/9x/CE/etc), people can pluck out the bad parts and keep the (OLE anyone?) good parts of kernel and userland alike. (Let me respond to any free-source zealots who would point me to CORBA/ORB: No, it isn't the real deal... yet.)
Let's also make this clear. A lot of supposed *flaws* in a Microsoft system are actually flaws in the coding practice brought about by the closed source. For instance, you don't *need* to reboot an NT system after making changes and reloading the TCP/IP stack (for you point-and-clickies out there, allowing for a new network service or protocol). I've actually reloaded the stack on the fly and have had no problems. NT *can* handle it (who knows, this might even apply to peripheral configuration as well). But, because the code is NOT open source, programmers assume their peers are idiots and demand a reboot. Open source will help stop that.
Furthermore, selling the source (but forcing this to be done) shows that Open Source need not be Pay-Nothing Source. This is a situation that will make pseudo-libertarians and GNU enthusiasts alike happy, or at least content. I'm a big fan of Open Source, but I much prefer the BSD-style licensing, which imposes less in the way of ideological restrictions. I believe a laissez-faire attitude will pop in and weed out those who would attempt to corrupt the system. And before anyone tells me that the DOJ case reeks of government interventionism, let me respond by saying that someone elected the people who appointed these folks.
Business and Politics are just two of many ways to get the masses to improve their situation. Together these ways make up economics. A lot of people confuse business with economics =P )
Now let's listen to your proposed solution:
Therefore my settlement proposal is as follows: "Microsoft may not produce or sell any operating system or operating system-like product for a period of 10 years, and must immediately destroy all copies of source and/or binaries for any of its current operating systems or operating system-like products." This gets rid of Windows permanently, and gives competitors a chance to do just that - compete. Tough but fair.
Neglecting for the minute the difficulty in achieiving this goal, let's look at history. Sure, keeping Microsoft from being an operating system OEM has been done. It's been done to AT&T actually. That's why they opened up the source of UNIX S3 to universities and the like. That's why bored university folk hacked at the system till, by God, it worked. But to destroy all the copies of Windows because of a personal grudge you have against it? Sounds like the Ch'in emperor burning all the books and building the Great Wall of China to keep out foreigners.
It didn't work. Neither will your plan. Opening the source will allow peers to see exactly what's going on behind the curtain, and thus more efficient coding practices will come into play (no more assuming that loading X will fuck up Y, if you can see how the loading process works), and who knows? Maybe compiling Windows with a better optimization flag helps! =)
Sorry to rain on your parade here, but why exactly is this so great? The fundamental concept that everyone seems to be missing is that Windows sucks. I don't want to see different companies produce it. I don't want WINE to emulate it. I don't want to see anyone make, use, or sell it. EVER. I want it to die. Not because I hate Micro$oft (after all, this is about somebody else selling it), but because it's a bad idea. The concept of a brutally complex, single-user operating system - term used loosely - is a bad, bad idea. Better to kill it immediately than to prolong the pain, even if it is supposedly at Micro$oft's expense.
Such short sightedness. Remember, opening up the source lets people *fix* things. This same imposition was made regarding the magtape distributions of UNIX System III and V from AT&T. Opening up the source helps. If the source to Windows is opened (including Windows NT/9x/CE/etc), people can pluck out the bad parts and keep the (OLE anyone?) good parts of kernel and userland alike. (Let me respond to any free-source zealots who would point me to CORBA/ORB: No, it isn't the real deal... yet.)
Let's also make this clear. A lot of supposed *flaws* in a Microsoft system are actually flaws in the coding practice brought about by the closed source. For instance, you don't *need* to reboot an NT system after making changes and reloading the TCP/IP stack (for you point-and-clickies out there, allowing for a new network service or protocol). I've actually reloaded the stack on the fly and have had no problems. NT *can* handle it (who knows, this might even apply to peripheral configuration as well). But, because the code is NOT open source, programmers assume their peers are idiots and demand a reboot. Open source will help stop that.
Furthermore, selling the source (but forcing this to be done) shows that Open Source need not be Pay-Nothing Source. This is a situation that will make pseudo-libertarians and GNU enthusiasts alike happy, or at least content. I'm a big fan of Open Source, but I much prefer the BSD-style licensing, which imposes less in the way of ideological restrictions. I believe a laissez-faire attitude will pop in and weed out those who would attempt to corrupt the system. And before anyone tells me that the DOJ case reeks of government interventionism, let me respond by saying that someone elected the people who appointed these folks.
Business and Politics are just two of many ways to get the masses to improve their situation. Together these ways make up economics. A lot of people confuse business with economics =P )
Now let's listen to your proposed solution:
Therefore my settlement proposal is as follows: "Microsoft may not produce or sell any operating system or operating system-like product for a period of 10 years, and must immediately destroy all copies of source and/or binaries for any of its current operating systems or operating system-like products." This gets rid of Windows permanently, and gives competitors a chance to do just that - compete. Tough but fair.
Neglecting for the minute the difficulty in achieiving this goal, let's look at history. Sure, keeping Microsoft from being an operating system OEM has been done. It's been done to AT&T actually. That's why they opened up the source of UNIX S3 to universities and the like. That's why bored university folk hacked at the system till, by God, it worked. But to destroy all the copies of Windows because of a personal grudge you have against it? Sounds like the Ch'in emperor burning all the books and building the Great Wall of China to keep out foreigners.
It didn't work. Neither will your plan. Opening the source will allow peers to see exactly what's going on behind the curtain, and thus more efficient coding practices will come into play (no more assuming that loading X will fuck up Y, if you can see how the loading process works), and who knows? Maybe compiling Windows with a better optimization flag helps! =)
/* So what we have is two seperate systems, built for completely disparate environmnents encroaching on each other's living space. (Imagine a whale being forced to walk on land and an elephant being forced to swim.) And at this point, they either evolve, or you end up with a lot of beached whales and drowned elephants. */
A little off topic, but elephants are actually very good swimmers. Their large surface area to allows them much buoyance, and their trunks can be used as snorkels.
Rob:/* Why are the courts allowed to decide this? */
Simple. We the citizenry voluntarily give some of our power to these old guys to make sweeping decisions about all things, great and small.
If you don't agree, you can do something about it. These judges are appointed by elected individuals and you have the power to elect them based on issues like this.
/* Come to think of it, I wonder why nature hasn't developed a breed of critter that's outside the food chain. Something that tastes bad to everything and breeds like a porn show. Sure it would be bad for the world, but I see nature as a random force and not an intellectual one */
One of the facts that a lot of people neglect is, no matter how much flaming occurs, splinter O/S's _WILL_ be produced. Open source isn't just about free, or reliable software; a lot of it is to hack the good hack just because you want to!
People love to compare aspects of their favorite O/S's because, let's face it, each O/S has its strong points. We all wish we could have an O/S with the strengths of every other O/S with none of the weaknesses.
For instance, FreeBSD executes system calls directly from the stack (as opposed to Linux, which calls it from the stack to a register). Whether this is a good or bad thing is dependent on what you're looking for.
The speed of DMA for a syscall is a plus for speed junkies on x86. On the other hand, it means that you can't port it easily to RISC processors, which require that all calls be run from registers anyway (thank you patterson and hennessy). Linux would be a wee bit more portable then, at the risk of losing DMA. What is more valuable? Totally subjective.
The advantage to ports is when installation of a package can't be done due to a dependency failure, it automatically fetches the dependency as well. rpm, for example, will just tell us what's missing.
However, given that most ftp sites for any given rpm or deb will also have a dependency rpm/deb package, this may be a moot point.
This is so true:) But we're forgetting the biggest "flavor fight" yet... Linuxen versus the BSDs!
Both are open source.
Both are robust.
Both are splintered into a numbers of specialized distributions
Linux versus BSD arguments have been the worst because neither side has any real advantage over the other. Each system has advantages and disadvantages. (Perhaps a compromise? BSD/Linux? Running Linus' kernel as a microkernel over BSD would prove interesting, albeit with a lot more overhead...)
Oh, and some fuel for the fire: BSD has real threads support. Have at you, Linux dogs!:)
Actually, boxen is an example of the only true English way of pluralizing a word.
By 'true' I mean actual English, unpolluted by the Norman invasion of 1066.
Ox, Oxen. Child, Children.
Very few of these plural forms survive. AFAIK the above two are the only one I can think of offhand. (Except the ever-present but irregular [wo]man, [wo]men.)
S suffixes to denote plurality are a greco-latinate derivative.
OTOH I see no need for computer geeks to speak Ye Olde English. But then what would you call a Beowulf cluster?
Let's parrot the previous statement with a little bit of twist to show how ridiculous this is.
"UNIX is an O/S that _NEEDS_ to be killed. Look at the sniveling piece of garbage, it's been around since 1970 and has its roots in MULTICS. The only thing keeping us from moving to faster O/Ses with better fileystems and a more streamlined kernel is Linux! I know the [BeOS, Apple, Microsoft, etc] development whizzes will port UNIX functionality to their own O/Ses in a week!"
See how ridiculous it sounds? Things EVOLVE, people. And evolution does not neccesarily entail replacing things. If it ain't broke, don't fix it! UNIX/Unix clones have been in development since 1970. NT is supposedly (ripped-off VMS code aside) a "ground-up" system and I bet that XENIX is more scalable. (Tongue in cheek statement)
x86 in and of itself isnt bad. However ISA architecture *is* a problem, because the same evolution that took place in the processor did not take place in the bus, processor cache, et al.
I had four NeXT Cubes (Uncle was a VP for NeXT). Excellent machines, I had them for almost ten years. The alloy of which you speak was a cast magnesium alloy. The box weighed about 30 lb. One of them was broken (cracked NeXTBus), so we threw it off a second story balcony. Amazingly, the case survived (of course, we took care to remove the working Winchester drives beforehand. The stupid optical drives were screwed from day one.)
The problems with NeXTcubes were manifold. First of all, I don't know if you can fit an ATX motherboard in there, because there is an enormous shared bus acting as a backplane. Logic boards slide into the bus sideways and access it that way. The hard drives and power supply plug into the shared bus in different places as well. To put an ATX motherboard in there would probably mean removing the shared bus, but then you wouldn't have a locking mechanism for the various parts. You can't screw anything in easily, because you'd have to drill a hole in thick metal.
The shared bus means, however, that you can plug in multiple motherboards. With some configuration of the O/S, of course. (This was in 1989, mind you -- years before Linux SMP.) The air flow in that machine was terrible. The old Winchester full-heights got HOT... i heated a burrito on one after fscking it three times in a row. The optical drives were typically sandwiched between the hot hard drive and hot power supply at the bottom -- that's why NeXT optical drives were notoriously unreliable. Magneto-optical works at a temperature called the Curie point (when the transition metal alloy in the Optical Disks was susceptible to vector realignment by a magnetic field.) Damn right, those boxes were poorly designed.
Oddly enough, I've heard a lot of complaints about the lack of expandability in cubes. I never found that a problem because I could always daisy-chain my SCSI devices externally. NeXTs used standard SCSI-1, I might add, nothing mutant like the old MicroVAX II...
Slashdot Mirror
while(!Slashdot.Readers.economic_knowledge)
:)
{
printf("
I think a lot of confusion vis-a-vis viewing the economy as "Zero-Sum" comes into play if you only view money as the tried and true measure of all things economical. I get $X via my labor, and I pay $X for some widgets, and from the perspective of the market, there's been a zero sum. Or so Sengan would have you believe.
Even this is wrong, taking into account inflation, money demand, and the interest rate. Even if money is our only benchmark, the value of money changes. It's so wrong that there's an entire branch of macroeconomics studying money market fluctuations..
Knowing that economics is not* a zero-sum game if we just considered money the benchmark, let's move on to consider social cost pricing**. The intrinsic worth of a transaction is based on:
1.how much one values the object
2. the social impact of having it produced, consumed, and changing hands
3. and various X-factors which I don't feel like listing
In short -- Comparative Advantage is what matters in any given transaction, not Absolute Advantage. A common analogue would be: If I had all the gold and you had all the hamburgers, sure, the gold is more valuable than the hamburgers normally but King Midas has got to eat sometime!
This isn't even taking into account that technology, capital goods procurement, and an expanding labor market increase production as well. Open source fits here quite well, i posit. Nowhere do I have to pay you for your labor in creating Open Source Software. Presumably, the social impact of using this software pays for itself. (No Microsoft Tax on OEM hardware if Linux becomes mainstream, that sort of thing.) On the flip side of things, if the social impact of totally free software becomes cumbersome (net traffic goes to a standstill every time a kernel is released) then the market will attempt to seek a solution (people will fork out $5 for CD-ROM of Linux).***
It is true that no one can truly know everyone's preferences, or measure how much impact social impact is.**** That's why people are tempted to view economics solely in terms of money. Do not be lulled into this trap, however! Armed with basic axioms of consumer preference and an entire branch of statistical economics -- Econometrics -- we can massage cross-sectional and time-series data to infer preferences. Such preferences deny a zero-sum game.
Besides letting us use phrases like "running a white test on a heteroskedasticity-consistent variance-covariance matrix in order to run a robust first-order regression,"***** econometrical analysis allows us to measure the cost of objects in terms of their correlative effects on human behavior. Here's where the zero-sum idea breaks down...
A zero-sum game wouldn't even need a regression line. Because there is no net gain, the correlative impact of X on Y or Z is nil -- do whatever you want, you're going to end up screwed anyway. Given the fact that there is enough variation in human economic behavior to require probability and autocorrelation tests, it's fairly self evident that X has an effect on Y and Z and vice versa.
If X and Y and Z have correlative effects on each other, it means the following:
1. Said economic transaction is not a zero-sum game, because the elements are not independent.
2. Non-Independence means that the variables are covariant.
3. Covariant variables tend to follow one or more of the Gauss-Markov assumptions about spherical disturbances ******. This means that the changes are stimulated in part by "outside forces," known as "Error terms," "disturbances," etc.
4. A zero sum game has no disturbances because there is nothing to disturb.
5. Technology, capital gains, political climate, social change, etc. count as "Disturbances" if they are infrequent or weak enough, else they count as variables of their own. (in which case they are covariant as above)
6. I enjoy saying the phrase "heteroskedasticity-consistent variance-covariance matrix." =P
7. Q.E.D. (sorta)
Discuss amongst yourselves.
Whew. That was quite rant.
\n");
}
/* sorry folks, I needed to let out some steam =) */
*(and probably never will be, except in peer-to-peer oligopoly transactions)
**(Campbell and Pereira, 1998)
*** Will the software still be free? Yes. But general procurement will not be. Of course, I have yet to find a person with free Internet access for those downloads...
**** Not 100% true. There is such a thing as experimental economics. Most people's preferences fit within Additive, Leontiff, or Cobb-Douglas utility functions (for substitutes, compliments, and not-quite-either goods respectively)
***** (Pindyck and Rubenfeld, 1997)
****** Even if they don't, they follow the assumptions if you do a robust regression. A robust regression replaces the linear approximation of trends with a taylor series expansion of an nth-degree polynomial fitting the data.
/* Imagine: Department of Computer Usage. Licensing test -- open a (Microsoft) word processor, compose and send an e-mail (using Outlook). Find the DCU's homepage using Internet Explorer. Three misclicks and out you go... And then: "computer usage is a privilege, not a right!". Two speeding convictions on the Internet and your license is suspended! */
:(
Heh! Well, the United States tend to have low state standards for drivers licenses anyway! This would be very in line for us.
Seriously, thought, there is a line between the human urge to maximize utility now (luxury, simplicity, ease) and maximize utility later (freedom, liberty, knowledge). This is going to be very atypical for slashdot -- I don't know where to draw that line! (He admitted he does not know! Dear God!)
People care about more results, alas, then the quality and freedom of code. And, sadly enough, people would probably sell their fundamental liberties (warning: USA bias) to go watch TV.
Wait, most of them already have.
I actually agree wholeheartedly. But most end users are going to be using the machine to do things like surf the world wide web, and word processing. Very rarely will they do much more, not because they are incapable intellectually, but because such fields are not yet in the public interest.
Personally, I'd like to create a licensing program for computer usage, just like drivers must be licensed. But it took many years of chaotic driving from unlicensed drivers about the turn of the century to convince people that it was worth licensing. This is social cost pricing. To implement an idiot-check for computer users now would be doomed to fail. As unproductive as widespread computer use is now, it has to get really downright annoying before people will submit to changes.
Such is the nature of humanity.
To say that cluebies should be kept away from computers puts a nail in the "open source can still make money from technical support" argument. Have it one way or another. Or do you believe that you can support Linux easily by telling inexperienced users to go to hell?
A car is a Tool. It gets you from point A to point B. Nowhere do we specify which points these are. A car can take you to the store or across the country. You can store stuff in your car. If you are camping, or on a road trip, or homeless, or lazy, you can sleep in your car. You can listen to music in your car. If you have a carphone, you can carry on conversations in your car.
It's ridiculous to say that if you are too clueless to re-pack your engine bearings, you shouldn't be supported. The world does not operated on a roll-your-own system. The joy of Linux, *BSD, and the like is that it gives you the freedom to do so, if you need to. Living your entire life on a roll-your-own basis is not the way for everyone. Accept this.
A delicious point of irony to the people who wrote the feature in question: Perhaps the newbie they refuse to help with difficult (yet documented) computer problems is the tax lawyer who will refuse to help them with difficult (yet documented) tax problems.
Heh. Sounds ridiculous on the other side of the fence, doesn't it? Sounds like a bit of geek rage rather than a logical assessment of the end user.
I've been working Unix tech support for close on two years now and believe it or not -- it's been a pleasure. Some end users are idiots, some are irate, but some are people who don't have the time to solve documented but pain-in-the-ass problems. I help them, and occasionally, they help me. Welcome to the world of social interaction. =)
sinator
Filing tax returns happily
The econ/CS major's $0.02:
Sorry to rain on your parade here, but why exactly is this so great? The fundamental concept that everyone seems to be missing is that Windows sucks. I don't want to see different companies produce it. I don't want WINE to emulate it. I don't want to see anyone make, use, or sell it. EVER. I want it to die. Not because I hate Micro$oft (after all, this is about somebody else selling it), but because it's a bad idea. The concept of a brutally complex, single-user operating system - term used loosely - is a bad, bad idea. Better to kill it immediately than to prolong the pain, even if it is supposedly at Micro$oft's expense.
Such short sightedness. Remember, opening up the source lets people *fix* things. This same imposition was made regarding the magtape distributions of UNIX System III and V from AT&T. Opening up the source helps. If the source to Windows is opened (including Windows NT/9x/CE/etc), people can pluck out the bad parts and keep the (OLE anyone?) good parts of kernel and userland alike. (Let me respond to any free-source zealots who would point me to CORBA/ORB: No, it isn't the real deal... yet.)
Let's also make this clear. A lot of supposed *flaws* in a Microsoft system are actually flaws in the coding practice brought about by the closed source. For instance, you don't *need* to reboot an NT system after making changes and reloading the TCP/IP stack (for you point-and-clickies out there, allowing for a new network service or protocol). I've actually reloaded the stack on the fly and have had no problems. NT *can* handle it (who knows, this might even apply to peripheral configuration as well). But, because the code is NOT open source, programmers assume their peers are idiots and demand a reboot. Open source will help stop that.
Furthermore, selling the source (but forcing this to be done) shows that Open Source need not be Pay-Nothing Source. This is a situation that will make pseudo-libertarians and GNU enthusiasts alike happy, or at least content. I'm a big fan of Open Source, but I much prefer the BSD-style licensing, which imposes less in the way of ideological restrictions. I believe a laissez-faire attitude will pop in and weed out those who would attempt to corrupt the system. And before anyone tells me that the DOJ case reeks of government interventionism, let me respond by saying that someone elected the people who appointed these folks.
Business and Politics are just two of many ways to get the masses to improve their situation. Together these ways make up economics. A lot of people confuse business with economics =P )
Now let's listen to your proposed solution:
Therefore my settlement proposal is as follows: "Microsoft may not produce or sell any operating system or operating system-like product for a period of 10 years, and must immediately destroy all copies of source and/or binaries for any of its current operating systems or operating system-like products." This gets rid of Windows permanently, and gives competitors a chance to do just that - compete. Tough but fair.
Neglecting for the minute the difficulty in achieiving this goal, let's look at history. Sure, keeping Microsoft from being an operating system OEM has been done. It's been done to AT&T actually. That's why they opened up the source of UNIX S3 to universities and the like. That's why bored university folk hacked at the system till, by God, it worked. But to destroy all the copies of Windows because of a personal grudge you have against it? Sounds like the Ch'in emperor burning all the books and building the Great Wall of China to keep out foreigners.It didn't work. Neither will your plan. Opening the source will allow peers to see exactly what's going on behind the curtain, and thus more efficient coding practices will come into play (no more assuming that loading X will fuck up Y, if you can see how the loading process works), and who knows? Maybe compiling Windows with a better optimization flag helps! =)
Ok. i know I was begging the question. Sue me. :)
Sorry to rain on your parade here, but why exactly is this so great? The fundamental concept that everyone seems to be missing is that Windows sucks. I don't want to see different companies produce it. I don't want WINE to emulate it. I don't want to see anyone make, use, or sell it. EVER. I want it to die. Not because I hate Micro$oft (after all, this is about somebody else selling it), but because it's a bad idea. The concept of a brutally complex, single-user operating system - term used loosely - is a bad, bad idea. Better to kill it immediately than to prolong the pain, even if it is supposedly at Micro$oft's expense.
Such short sightedness. Remember, opening up the source lets people *fix* things. This same imposition was made regarding the magtape distributions of UNIX System III and V from AT&T. Opening up the source helps. If the source to Windows is opened (including Windows NT/9x/CE/etc), people can pluck out the bad parts and keep the (OLE anyone?) good parts of kernel and userland alike. (Let me respond to any free-source zealots who would point me to CORBA/ORB: No, it isn't the real deal... yet.)
Let's also make this clear. A lot of supposed *flaws* in a Microsoft system are actually flaws in the coding practice brought about by the closed source. For instance, you don't *need* to reboot an NT system after making changes and reloading the TCP/IP stack (for you point-and-clickies out there, allowing for a new network service or protocol). I've actually reloaded the stack on the fly and have had no problems. NT *can* handle it (who knows, this might even apply to peripheral configuration as well). But, because the code is NOT open source, programmers assume their peers are idiots and demand a reboot. Open source will help stop that.
Furthermore, selling the source (but forcing this to be done) shows that Open Source need not be Pay-Nothing Source. This is a situation that will make pseudo-libertarians and GNU enthusiasts alike happy, or at least content. I'm a big fan of Open Source, but I much prefer the BSD-style licensing, which imposes less in the way of ideological restrictions. I believe a laissez-faire attitude will pop in and weed out those who would attempt to corrupt the system. And before anyone tells me that the DOJ case reeks of government interventionism, let me respond by saying that someone elected the people who appointed these folks.
Business and Politics are just two of many ways to get the masses to improve their situation. Together these ways make up economics. A lot of people confuse business with economics =P )
Now let's listen to your proposed solution:
Therefore my settlement proposal is as follows: "Microsoft may not produce or sell any operating system or operating system-like product for a period of 10 years, and must immediately destroy all copies of source and/or binaries for any of its current operating systems or operating system-like products." This gets rid of Windows permanently, and gives competitors a chance to do just that - compete. Tough but fair.
Neglecting for the minute the difficulty in achieiving this goal, let's look at history. Sure, keeping Microsoft from being an operating system OEM has been done. It's been done to AT&T actually. That's why they opened up the source of UNIX S3 to universities and the like. That's why bored university folk hacked at the system till, by God, it worked. But to destroy all the copies of Windows because of a personal grudge you have against it? Sounds like the Ch'in emperor burning all the books and building the Great Wall of China to keep out foreigners.It didn't work. Neither will your plan. Opening the source will allow peers to see exactly what's going on behind the curtain, and thus more efficient coding practices will come into play (no more assuming that loading X will fuck up Y, if you can see how the loading process works), and who knows? Maybe compiling Windows with a better optimization flag helps! =)
Ok. i know I was begging the question. Sue me. :)
/* So what we have is two seperate systems, built for completely disparate environmnents encroaching on each other's living space. (Imagine a whale being forced to walk on land and an elephant being forced to swim.) And at this point, they either evolve, or you end up with a lot of beached whales and drowned elephants. */
A little off topic, but elephants are actually very good swimmers. Their large surface area to allows them much buoyance, and their trunks can be used as snorkels.
Just a little useless fact.
Simple. We the citizenry voluntarily give some of our power to these old guys to make sweeping decisions about all things, great and small.
If you don't agree, you can do something about it. These judges are appointed by elected individuals and you have the power to elect them based on issues like this.
Welcome to the world of being politically aware.
Homo Sapiens Sapiens.
One of the facts that a lot of people neglect is, no matter how much flaming occurs, splinter O/S's _WILL_ be produced. Open source isn't just about free, or reliable software; a lot of it is to hack the good hack just because you want to!
People love to compare aspects of their favorite O/S's because, let's face it, each O/S has its strong points. We all wish we could have an O/S with the strengths of every other O/S with none of the weaknesses.
For instance, FreeBSD executes system calls directly from the stack (as opposed to Linux, which calls it from the stack to a register). Whether this is a good or bad thing is dependent on what you're looking for.
The speed of DMA for a syscall is a plus for speed junkies on x86. On the other hand, it means that you can't port it easily to RISC processors, which require that all calls be run from registers anyway (thank you patterson and hennessy). Linux would be a wee bit more portable then, at the risk of losing DMA. What is more valuable? Totally subjective.
oh well...
The advantage to ports is when installation of a package can't be done due to a dependency failure, it automatically fetches the dependency as well. rpm, for example, will just tell us what's missing.
However, given that most ftp sites for any given rpm or deb will also have a dependency rpm/deb package, this may be a moot point.
Linux versus BSD arguments have been the worst because neither side has any real advantage over the other. Each system has advantages and disadvantages. (Perhaps a compromise? BSD/Linux? Running Linus' kernel as a microkernel over BSD would prove interesting, albeit with a lot more overhead...)
Oh, and some fuel for the fire: BSD has real threads support. Have at you, Linux dogs! :)
Actually, boxen is an example of the only true English way of pluralizing a word.
By 'true' I mean actual English, unpolluted by the Norman invasion of 1066.
Ox, Oxen.
Child, Children.
Very few of these plural forms survive. AFAIK the above two are the only one I can think of offhand. (Except the ever-present but irregular [wo]man, [wo]men.)
S suffixes to denote plurality are a greco-latinate derivative.
OTOH I see no need for computer geeks to speak Ye Olde English. But then what would you call a Beowulf cluster?
Let's parrot the previous statement with a little bit of twist to show how ridiculous this is.
"UNIX is an O/S that _NEEDS_ to be killed. Look at the sniveling piece of garbage, it's been around since 1970 and has its roots in MULTICS. The only thing keeping us from moving to faster O/Ses with better fileystems and a more streamlined kernel is Linux! I know the [BeOS, Apple, Microsoft, etc] development whizzes will port UNIX functionality to their own O/Ses in a week!"
See how ridiculous it sounds? Things EVOLVE, people. And evolution does not neccesarily entail replacing things. If it ain't broke, don't fix it! UNIX/Unix clones have been in development since 1970. NT is supposedly (ripped-off VMS code aside) a "ground-up" system and I bet that XENIX is more scalable. (Tongue in cheek statement)
x86 in and of itself isnt bad. However ISA architecture *is* a problem, because the same evolution that took place in the processor did not take place in the bus, processor cache, et al.
I gave you a dollar. Where's my $0.98?
I had four NeXT Cubes (Uncle was a VP for NeXT). Excellent machines, I had them for almost ten years. The alloy of which you speak was a cast magnesium alloy. The box weighed about 30 lb. One of them was broken (cracked NeXTBus), so we threw it off a second story balcony. Amazingly, the case survived (of course, we took care to remove the working Winchester drives beforehand. The stupid optical drives were screwed from day one.)
... i heated a burrito on one after fscking it three times in a row. The optical drives were typically sandwiched between the hot hard drive and hot power supply at the bottom -- that's why NeXT optical drives were notoriously unreliable. Magneto-optical works at a temperature called the Curie point (when the transition metal alloy in the Optical Disks was susceptible to vector realignment by a magnetic field.) Damn right, those boxes were poorly designed.
The problems with NeXTcubes were manifold. First of all, I don't know if you can fit an ATX motherboard in there, because there is an enormous shared bus acting as a backplane. Logic boards slide into the bus sideways and access it that way. The hard drives and power supply plug into the shared bus in different places as well. To put an ATX motherboard in there would probably mean removing the shared bus, but then you wouldn't have a locking mechanism for the various parts. You can't screw anything in easily, because you'd have to drill a hole in thick metal.
The shared bus means, however, that you can plug in multiple motherboards. With some configuration of the O/S, of course. (This was in 1989, mind you -- years before Linux SMP.) The air flow in that machine was terrible. The old Winchester full-heights got HOT
Oddly enough, I've heard a lot of complaints about the lack of expandability in cubes. I never found that a problem because I could always daisy-chain my SCSI devices externally. NeXTs used standard SCSI-1, I might add, nothing mutant like the old MicroVAX II...
Ah, those were the days.