You will see, as responses, a number of statements made by people with one degree or another claiming that their education is what allows them to refute or substantiate these claims.
Not so much that; it's more like they learned a few tricks for testing physics hypotheses while there. Often (althoug not always) less entertaining than some of the other tricks also learned in those ivy halls (and under various bushes, tables, on roofs, etc!)
A degree is only proof that you have the economic power to attend a center of education and regurgitate what they want you to regurgitate.
Perhaps a bit more than that. For instance, it also means that you've been exposed to some of the classical theory testing tools such as Carnot's little engine, Michaelson and Morley's elegant lightshow, the Hall effect, and Einstein's test cases involving FTL and causation. It's not the answers, you see; it's all those pesky questions.
In addition it means that you've been exposed to handy concepts like, "if you don't see it happening in nature you can bet there's a reason why." Hydrogen fusion doesn't happen in all that seawater often enough to bother about because there's a huge activation energy. Mills' stuff, on the other hand, involves mixing common chemicals under conditions that are pretty common in the Universe -- which rather invites the question of why we haven't noticed those little hydrinos getting it on in e.g., fertilizer plants.
Think about this: if it were possible for hydrogen atoms to transition to a lower energy state, they would have already done so by now. Mother Nature likes to be in the lowest energy state possible. If she gould have squeezed more energy out of hydrogen atoms, she would have already done so.
She's stingy, all right. Look at the merry Hob that the transition between cis- and trans- states cause for LH2.
Background: the protons in a hydrogen atom can either align with parallel or antiparallel magnetic axes. The energy difference is miniscule, and at room temperature thermal exitation keeps them pretty much equally common. At liquid-hydrogen temperatures, though, the only stable state is antiparallel. Having a tank of LH2 drop from 50%/50% to ground state can be an exciting event!
So Mums is really good at finding lower energy states -- but somehow missed something in the ultraviolet range?
More accurately, "if hydrogen has a lower-energy state, that invalidates the fundamentals of quantum mechanics, which otherwise has a splendid track record of predicting phenomena like lasers, semiconductors, quantum optics, etc. etc. etc." Which in turn raises the interesting question: if Schroedinger's equation is wrong, then what is that scanner at the checkout counter doing?
overshoot, wondering what the orbit state below h_bar/2 might be, and what would happen if all that Solar hydrogen discovered that it had somewhere to go?
Interesting -- have they simply delegated the setting of DOD standards to Microsoft? When MS changes their file formats (for instance) do the old documents become noncompliant, or is the new software noncompliant, or does some O-5 somewhere draft a program to convert to the new standard, or what?
FWIW, I've been there a few times. In theory, Government RFQs are supposed to set out the needed characteristics; the order is required by law to go to the lowest bidder complying with the requirements. In theory this prevents corruption by leveling the playing field. In practice, bureaucrats get around it by writing the requirements to exactly match the (preselected) product.
For instance, I once saw an RFQ that specified the exact length of the power cord plus or minus a quarter inch. Amazingly, that was the length of IBM's cord, but not one commercially available. All of the other bidders had to have custom cords made (at extra cost.) The same pattern was repeated all through the RFQ.
In Corel's case, what appears to have happened was that the Labor Department insisted on several of Word's quirkier features (e.g., macro language) which almost nobody uses. Instant disqualification for everyone but Microsoft.
First step - the download: it was ok so far, but you have to be a community.borland member in order to get your license code. I already was, but newcomers should better be told that it's for free and doesn't come with any drawbacks.
Be sure to use a throwaway e-mail address, though, unless you want to get a LOT of spam. Inprise doesn't seem to have quite gotten the message yet and keeps flirting with the RBL.
Assume that MS gets diced into (my own favorite) three planks plus sawdust: Consumer OS (W98), business/server OS (WNT), and productivity apps (Office). Divest MSN, games, keyboards, etc.
Then along come at least 19 class-action suits plus the OEMs. I'm assuming that IBM and HP still have time to file, since they seem to have plenty of cause.
Which pieces of MS get nailed with the residual liability of their monolythic ancestor? In other words, who's left to sue? I don't for a minute believe that the USSC would allow a conclusion forclosing IBM's pursuit of redress, especially since the harm to them wasn't at issue in the trial, but would there be some sort of shared membership in a liability pool or what?
First off, you will be better off with a non-x86 processor. Whether they suck in other ways or not, they absolutely suck juice. That said, there's a lot you can do.
Step one: forget the floppy and use a cheap CD-rom drive. They're made to shut down nicely and that way you won't be counting bytes.
Step two: get rid of the heat generation. Use an old-but-not-ancient CPU on a motherboard somwhat newer than it is, with a selection for supply voltages lower than the rated voltage for the CPU. Processor power is proportional to the square of supply voltage and proportional to the frequency, so a reduction from 2.3 to 2.1 volts and from 233 MHz to 200 will result in a 30% power savings. The lower voltage won't support full speed, but will support a derated speed. You'll need to experiment with power, temperature, and clock multipliers. This is like overclocking in reverse; take the speed down (let's say from 233 to 166) and find out where the processor fails at 70C ambient. Put the voltage up a notch and drop the speed to 133 or less. That kind of thing.
Finally, you'll want to slow down the fan. Notice I didn't say 'stop'. Fan noise is amazingly nonlinear. Drop the speed by half and you won't be able to hear it. You can drop the speed by using a lower power supply voltage than design; most fans seem to run on the +12v supply. Try the 5v supply and see if it starts reliably. Alternately, see if you can get a 24 v fan and run it at 12 v. Either way, the speed will be about half of normal, which translates into better than a 10 dB noise reduction. Put the fan behind a baffle and if it's not silent it'll be very close.
I'd be very nervous about shutting the fan down entirely unless the PS is made to work that way (Apple was seriously anti-fan for years) because even in idle there are parts in the power supply itself that dissipate quite a bit of power and depend on air to get rid of the heat. Maybe not a LOT of heat, but when you divide by zero airflow the result ain't pretty.
The Wing Lover flew in with: But Linux simply isn't ready for non-computer-geeks to be using all the time. It's propbably okay for smart non-computer-geeks, as long as they have a bit of support once in a while. But it's still not ready for Aunt Helga who wants to check her email once in a while and run a word processor.
Coulda fooled me. The SO is about as antitechnological as it gets (a can of paint is close to the limit; the kitchen is way too Buck Rogers) and the kids are -- well, kids. Oddly enough, though, it doesn't seem to slow them down any, and they seem to take it for granted that they can do Stupid Net Tricks but not access each others' data (which they used to do for mischief.)
Now all I have to do is keep the daughter from finding out that I could let her have some of the messaging S/W that she wants but that it doesn't work because I have it blocked at the firewall....
Otto asked: Provide me an address and a way to look up other addresses. I'm happy then. Unless there's something else I don't know about (which is probably true).
If you want NNTP, you're going to need an NNTP host. If you want DNS, you need a DNS service. Most small nodes benefit from SMTP hosting on redundant server farms; it's more reliable. Even large companies with fat pipes and server farms of their own are finding it beneficial to use external webhosting services. Hey, even NTP is best run with some structure.
The list goes on. Are these necessary services? Some, yes. Other, maybe. Many, probably not. So shop around. The diversity of Internet Service providers allows the market to sort these issues out rather than having them decided by one suit in New Jersey.
Paul Carver writes: If there were a legal way out of this contract AT&T would gladly let other ISPs deliver service over their cable lines since more content leads to more demand. AT&T markets phone service in partnership with a variety of other companies and organizations and would be glad to have dozens or even hundreds of special interest ISPs doing targeted marketing and paying AT&T for the infrastructure.
Help me here. If AT&T is so unhappy to be bound by the MediaOne contract, howcome they're fighting so hard against regulations that would free them from it? After all, contractual obligations are always subordinated and superceded by legislative and regulatory ones.
Robin Hood wrote: Say MediaOne starts blocking the ports used for IP telephony -- after all, that's a direct competitor to AT&T's primary business. Suddenly, millions of MediaOne customers are forced either to switch to another ISP or give up using IP telephony. And if they switch to another ISP, they're still paying for MediaOne! Don't want to pay for MediaOne? Sorry -- it comes with your DSL connection; if you don't want MediaOne, you're going to have to find another DSL service.
Much more to the point, you still have port 119 blocked. Your access to YAISP is through MediaOne, not around them. Their Terms of Service still apply, and so does their firewalling of port 119.
This isn't hypothetical; I tried to get Cox/@Home to connect me, and was even willing to pay extra for access to my old ISP. No such luck; their TOS forbids having any servers attached and I use NFS for the home machinery.
Someone explain (please!) why a Trojan payload couldn't just throw up a fake BSOD, fake reboot, and fake login screen? "Active content" of all kinds is supposed to do that kind of screen manipulation, right? The main exploit is that people take sudden BSOD for granted.
One reason to speed things up is that if MICROS~1 thinks that they are going to be broken up, they have an increased incentive to clear the field first. Anyone care to guess how many bodies they could pile up if they thought that they had nothing in the long run to lose?
(Slightly OT; I think I've mentioned this before.) At this year's Design Automation Conference in New Orleans, there were penguins everywhere. Most of the exhibitors seemed oblivious to the Linux connection, but I asked as many as possible about Linux versions.
The near-universal response was that they would have a Linux version of their software as soon as they got an order large enough to make it worthwhile. Says I, "I know that porting to Linux is easy." Replied they: "No, you don't understand. Our developers insist on doing Linux first, then porting to other systems. We just need an order."
Apparently Model Technology (BIG simulation company) got the order; they announced that starting in December they'll be shipping Linux versions along with all of their other platforms.
And he was definitely plowing new ground. Traditionally, companies have gotten into anti-trust trouble when their monopolies become so vast they monopolize products and goods, prevent competition and innovation, and unfairly control and drive up the price consumers pay for those products. That was the rationale behind one of the first landmark anti-trust rulings, the one that broke up Standard Oil, and behind the decision that dispersed AT&T.
Jon, before you lecture on the Standard Oil case (the archetypal antitrust example, to be sure) it's a good idea to learn a little bit about it.
Standard Oil did not jack up prices. Instead, like Microsoft, they used their monopoly power to guard and extend their monopoly. Under Standard Oil, prices actually fell as Rockefeller shared the economies of scale. (OK, so maybe he was a bit more public-spirited than Bill.) It was, however, impossible to survive as an independent oil company because Standard controlled the infrastructure and the channels and exerted pressure on equipment manufacturers while engaging in predatory pricing to make the independents unprofitable.
Sound familiar? Judge Jackson is on very familiar and well-tested legal ground here.
Focus on that. Thousands of people all working together, no giants. Maybe an aerial shot of a LOT of people in a field, collectively in the shape of a penguin.
Shouldn't be too hard to do with graphics composition, should it?
I find myself, surprisingly, against breaking up MICROS~1. In light of recent decisions levying honking fines agains major coprorations, it's likely that a fine of (let's say) 5-10% of their market cap would pass muster on appeal. Due and payable next year, but held (drawing interest) until final resolution.
Peanuts, you say? Perhaps, for many companies. In MS' case, though, that's several years' revenues even with their funny accounting. Also, it won't go down based on subsequent events, so if (as looks likely) they've pretty much hit their zenith then by the time the USSC slams the gavel a little matter of $5E10 might just finish them off.
OK, so I'm not the forgiving type. Deal.
PS: Ya gotta love Judge Jackson's sense of humor. How many of us knew that he read Mathematical Games, much less that he would actually find a chance to play a variant of the Unexpected Hanging?
The Company actually had a couple of Alpha machines that we were going to use for SPICE boxes, but DEC wanted several thousand $$$ per for a single-user license. Wasn't worth it. They sat around for a while and we loaded Alpha Linux on them and that was kool, but no executables for anything we really needed to do.
So, a couple of weeks ago, they were officially scrapped. I hauled them home hoping to load them with a current Linux distro (still no Alpha RH6.1, though -- wonder why?) So while getting around to it along comes Compaq with a rush of brains to the head and guess what?
Hrunting wisely observed: Some people are saying that this only counts authenticated NT users, but this statement would contradict that. They are squarely saying that if you're going to have multiple people connecting to your system, you're going to pay for it. This is similar to news outsourcing where a company pays for a number of concurrent connections. They don't pay for all their customers, just the ones they'll expect to be connected at once. Now come the interesting questions. What if you use Apache as your web server and someone tries to connect. Do you have to pay for it? Can you charge for connections to an operating system? This doesn't sound like a feasible pricing scheme to me unless Microsoft is going to implement some sort of connection limiting scheme in its software (highly unlikely, although, like I said, they have a twisted sense of humor).
Since IIS is a kernel service (!) connection limiting would be easy. OTOH, since the issue is license enforcement there's no reason that they couldn't set up a bot to interrogate sites now and then about their peak connection load. If it's over the wire, an invoice arrives. Don't pay the invoice, and thanks to some effective lobbying MICROS~1 now has the legal right to remotely shut the server down -- anyone want to bet that they didn't add the hooks to do it? They fought HARD for that clause, after all.
... there was a major computer manufacturer. The manufacturer had historically produced high-powered machinery with full-blown operating systems, but in recent years had seen some erosion of its sales at the low end from less-capable but much less expensive brands. The Big Company decided to get in on the new trend and introduced its own line of low-end computers.
Since its customers needed more than hardware, the Big Company offered their machines with a choice of operating software: users could have the industry standard or a new product made specifically for the new machines. The standard software naturally commanded a premium price: $40 extra.
Now according to your theory, users would have overwhelmingly preferred the standard, premium operating software over the freebie included with the machine. That's not what happened, though. Anyone recognize the story?
Slashdot Mirror
Hmmmm... sort of like a physics version of Quackwatch?
You will see, as responses, a number of statements made by people with one degree or another claiming that their education is what allows them to refute or substantiate these claims.
Not so much that; it's more like they learned a few tricks for testing physics hypotheses while there. Often (althoug not always) less entertaining than some of the other tricks also learned in those ivy halls (and under various bushes, tables, on roofs, etc!)
A degree is only proof that you have the economic power to attend a center of education and regurgitate what they want you to regurgitate.
Perhaps a bit more than that. For instance, it also means that you've been exposed to some of the classical theory testing tools such as Carnot's little engine, Michaelson and Morley's elegant lightshow, the Hall effect, and Einstein's test cases involving FTL and causation. It's not the answers, you see; it's all those pesky questions.
In addition it means that you've been exposed to handy concepts like, "if you don't see it happening in nature you can bet there's a reason why." Hydrogen fusion doesn't happen in all that seawater often enough to bother about because there's a huge activation energy. Mills' stuff, on the other hand, involves mixing common chemicals under conditions that are pretty common in the Universe -- which rather invites the question of why we haven't noticed those little hydrinos getting it on in e.g., fertilizer plants.
Think about this: if it were possible for hydrogen atoms to transition to a lower energy state, they would have already done so by now. Mother Nature likes to be in the lowest energy state possible. If she gould have squeezed more energy out of hydrogen atoms, she would have already done so.
She's stingy, all right. Look at the merry Hob that the transition between cis- and trans- states cause for LH2.
Background: the protons in a hydrogen atom can either align with parallel or antiparallel magnetic axes. The energy difference is miniscule, and at room temperature thermal exitation keeps them pretty much equally common. At liquid-hydrogen temperatures, though, the only stable state is antiparallel. Having a tank of LH2 drop from 50%/50% to ground state can be an exciting event!
So Mums is really good at finding lower energy states -- but somehow missed something in the ultraviolet range?
More accurately, "if hydrogen has a lower-energy state, that invalidates the fundamentals of quantum mechanics, which otherwise has a splendid track record of predicting phenomena like lasers, semiconductors, quantum optics, etc. etc. etc." Which in turn raises the interesting question: if Schroedinger's equation is wrong, then what is that scanner at the checkout counter doing?
overshoot, wondering what the orbit state below h_bar/2 might be, and what would happen if all that Solar hydrogen discovered that it had somewhere to go?
1)You can't win
2)You can't try
3)You'll always loose.
She probably got it third-hand from an earlier version. The one I learned in the 60s from reading John Campbell was:
Blacker and better physics.
always worked well, and involved a lot less reentering of the wheel than xf86config. YMMV.
Interesting -- have they simply delegated the setting of DOD standards to Microsoft? When MS changes their file formats (for instance) do the old documents become noncompliant, or is the new software noncompliant, or does some O-5 somewhere draft a program to convert to the new standard, or what?
FWIW, I've been there a few times. In theory, Government RFQs are supposed to set out the needed characteristics; the order is required by law to go to the lowest bidder complying with the requirements. In theory this prevents corruption by leveling the playing field. In practice, bureaucrats get around it by writing the requirements to exactly match the (preselected) product.
For instance, I once saw an RFQ that specified the exact length of the power cord plus or minus a quarter inch. Amazingly, that was the length of IBM's cord, but not one commercially available. All of the other bidders had to have custom cords made (at extra cost.) The same pattern was repeated all through the RFQ.
In Corel's case, what appears to have happened was that the Labor Department insisted on several of Word's quirkier features (e.g., macro language) which almost nobody uses. Instant disqualification for everyone but Microsoft.
First step - the download:
it was ok so far, but you have to be a community.borland member in order to get your license code. I already was, but newcomers should better be told that it's for free and doesn't come with any drawbacks.
Be sure to use a throwaway e-mail address, though, unless you want to get a LOT of spam. Inprise doesn't seem to have quite gotten the message yet and keeps flirting with the RBL.
Assume that MS gets diced into (my own favorite) three planks plus sawdust: Consumer OS (W98), business/server OS (WNT), and productivity apps (Office). Divest MSN, games, keyboards, etc.
Then along come at least 19 class-action suits plus the OEMs. I'm assuming that IBM and HP still have time to file, since they seem to have plenty of cause.
Which pieces of MS get nailed with the residual liability of their monolythic ancestor? In other words, who's left to sue? I don't for a minute believe that the USSC would allow a conclusion forclosing IBM's pursuit of redress, especially since the harm to them wasn't at issue in the trial, but would there be some sort of shared membership in a liability pool or what?
First off, you will be better off with a non-x86 processor. Whether they suck in other ways or not, they absolutely suck juice. That said, there's a lot you can do.
Step one: forget the floppy and use a cheap CD-rom drive. They're made to shut down nicely and that way you won't be counting bytes.
Step two: get rid of the heat generation. Use an old-but-not-ancient CPU on a motherboard somwhat newer than it is, with a selection for supply voltages lower than the rated voltage for the CPU. Processor power is proportional to the square of supply voltage and proportional to the frequency, so a reduction from 2.3 to 2.1 volts and from 233 MHz to 200 will result in a 30% power savings. The lower voltage won't support full speed, but will support a derated speed. You'll need to experiment with power, temperature, and clock multipliers. This is like overclocking in reverse; take the speed down (let's say from 233 to 166) and find out where the processor fails at 70C ambient. Put the voltage up a notch and drop the speed to 133 or less. That kind of thing.
Finally, you'll want to slow down the fan. Notice I didn't say 'stop'. Fan noise is amazingly nonlinear. Drop the speed by half and you won't be able to hear it. You can drop the speed by using a lower power supply voltage than design; most fans seem to run on the +12v supply. Try the 5v supply and see if it starts reliably. Alternately, see if you can get a 24 v fan and run it at 12 v. Either way, the speed will be about half of normal, which translates into better than a 10 dB noise reduction. Put the fan behind a baffle and if it's not silent it'll be very close.
I'd be very nervous about shutting the fan down entirely unless the PS is made to work that way (Apple was seriously anti-fan for years) because even in idle there are parts in the power supply itself that dissipate quite a bit of power and depend on air to get rid of the heat. Maybe not a LOT of heat, but when you divide by zero airflow the result ain't pretty.
cr0sh asked:
Anybody know what the -5V is used for in a PC?
Sure. It's for the substrate bias on triple-supply (+12v, +5v, -5v) 16 kb DRAM chips. No, that's not a mistake; I really mean sixteen kilobits.
The Wing Lover flew in with:
But Linux simply isn't ready for non-computer-geeks to be using all the time. It's propbably okay for smart non-computer-geeks, as long as they have a bit of support once in a while. But it's still not ready for Aunt Helga who wants to check her email once in a while and run a word processor.
Coulda fooled me. The SO is about as antitechnological as it gets (a can of paint is close to the limit; the kitchen is way too Buck Rogers) and the kids are -- well, kids. Oddly enough, though, it doesn't seem to slow them down any, and they seem to take it for granted that they can do Stupid Net Tricks but not access each others' data (which they used to do for mischief.)
Now all I have to do is keep the daughter from finding out that I could let her have some of the messaging S/W that she wants but that it doesn't work because I have it blocked at the firewall....
Otto asked:
Provide me an address and a way to look up other addresses. I'm happy then. Unless there's something else I don't know about (which is probably true).
If you want NNTP, you're going to need an NNTP host. If you want DNS, you need a DNS service. Most small nodes benefit from SMTP hosting on redundant server farms; it's more reliable. Even large companies with fat pipes and server farms of their own are finding it beneficial to use external webhosting services. Hey, even NTP is best run with some structure.
The list goes on. Are these necessary services? Some, yes. Other, maybe. Many, probably not. So shop around. The diversity of Internet Service providers allows the market to sort these issues out rather than having them decided by one suit in New Jersey.
Paul Carver writes:
If there were a legal way out of this contract AT&T would gladly let other ISPs deliver service over their cable lines since more content leads to more demand. AT&T markets phone service in partnership with a variety of other companies and organizations and would be glad to have dozens or even hundreds of special interest ISPs doing targeted marketing and paying AT&T for the infrastructure.
Help me here. If AT&T is so unhappy to be bound by the MediaOne contract, howcome they're fighting so hard against regulations that would free them from it? After all, contractual obligations are always subordinated and superceded by legislative and regulatory ones.
Robin Hood wrote:
Say MediaOne starts blocking the ports used for IP telephony -- after all, that's a direct competitor to AT&T's primary business. Suddenly, millions of MediaOne customers are forced either to switch to another ISP or give up using IP telephony. And if they switch to another ISP, they're still paying for MediaOne! Don't want to pay for MediaOne? Sorry -- it comes with your DSL connection; if you don't want MediaOne, you're going to have to find another DSL service.
Much more to the point, you still have port 119 blocked. Your access to YAISP is through MediaOne, not around them. Their Terms of Service still apply, and so does their firewalling of port 119.
This isn't hypothetical; I tried to get Cox/@Home to connect me, and was even willing to pay extra for access to my old ISP. No such luck; their TOS forbids having any servers attached and I use NFS for the home machinery.
Someone explain (please!) why a Trojan payload couldn't just throw up a fake BSOD, fake reboot, and fake login screen? "Active content" of all kinds is supposed to do that kind of screen manipulation, right? The main exploit is that people take sudden BSOD for granted.
One reason to speed things up is that if MICROS~1 thinks that they are going to be broken up, they have an increased incentive to clear the field first. Anyone care to guess how many bodies they could pile up if they thought that they had nothing in the long run to lose?
(Slightly OT; I think I've mentioned this before.)
At this year's Design Automation Conference in New Orleans, there were penguins everywhere. Most of the exhibitors seemed oblivious to the Linux connection, but I asked as many as possible about Linux versions.
The near-universal response was that they would have a Linux version of their software as soon as they got an order large enough to make it worthwhile. Says I, "I know that porting to Linux is easy." Replied they: "No, you don't understand. Our developers insist on doing Linux first, then porting to other systems. We just need an order."
Apparently Model Technology (BIG simulation company) got the order; they announced that starting in December they'll be shipping Linux versions along with all of their other platforms.
And he was definitely plowing new ground. Traditionally, companies have gotten into anti-trust trouble when their monopolies become so vast they monopolize products and goods, prevent competition and innovation, and unfairly control and drive up the price consumers pay for those products. That was the rationale behind one of the first landmark anti-trust rulings, the one that broke up Standard Oil, and behind the decision that dispersed AT&T.
Jon, before you lecture on the Standard Oil case (the archetypal antitrust example, to be sure) it's a good idea to learn a little bit about it.
Standard Oil did not jack up prices. Instead, like Microsoft, they used their monopoly power to guard and extend their monopoly. Under Standard Oil, prices actually fell as Rockefeller shared the economies of scale. (OK, so maybe he was a bit more public-spirited than Bill.) It was, however, impossible to survive as an independent oil company because Standard controlled the infrastructure and the channels and exerted pressure on equipment manufacturers while engaging in predatory pricing to make the independents unprofitable.
Sound familiar? Judge Jackson is on very familiar and well-tested legal ground here.
Focus on that. Thousands of people all working together, no giants. Maybe an aerial shot of a LOT of people in a field, collectively in the shape of a penguin.
Shouldn't be too hard to do with graphics composition, should it?
I find myself, surprisingly, against breaking up MICROS~1. In light of recent decisions levying honking fines agains major coprorations, it's likely that a fine of (let's say) 5-10% of their market cap would pass muster on appeal. Due and payable next year, but held (drawing interest) until final resolution.
Peanuts, you say? Perhaps, for many companies. In MS' case, though, that's several years' revenues even with their funny accounting. Also, it won't go down based on subsequent events, so if (as looks likely) they've pretty much hit their zenith then by the time the USSC slams the gavel a little matter of $5E10 might just finish them off.
OK, so I'm not the forgiving type. Deal.
PS: Ya gotta love Judge Jackson's sense of humor. How many of us knew that he read Mathematical Games, much less that he would actually find a chance to play a variant of the Unexpected Hanging?
The Company actually had a couple of Alpha machines that we were going to use for SPICE boxes, but DEC wanted several thousand $$$ per for a single-user license. Wasn't worth it. They sat around for a while and we loaded Alpha Linux on them and that was kool, but no executables for anything we really needed to do.
So, a couple of weeks ago, they were officially scrapped. I hauled them home hoping to load them with a current Linux distro (still no Alpha RH6.1, though -- wonder why?) So while getting around to it along comes Compaq with a rush of brains to the head and guess what?
ROTFLMAO!
Hrunting wisely observed: Some people are saying that this only counts authenticated NT users, but this statement would contradict that. They are squarely saying that if you're going to have multiple people connecting to your system, you're going to pay for it. This is similar to news outsourcing where a company pays for a number of concurrent connections. They don't pay for all their customers, just the ones they'll expect to be connected at once. Now come the interesting questions. What if you use Apache as your web server and someone tries to connect. Do you have to pay for it? Can you charge for connections to an operating system? This doesn't sound like a feasible pricing scheme to me unless Microsoft is going to implement some sort of connection limiting scheme in its software (highly unlikely, although, like I said, they have a twisted sense of humor).
Since IIS is a kernel service (!) connection limiting would be easy. OTOH, since the issue is license enforcement there's no reason that they couldn't set up a bot to interrogate sites now and then about their peak connection load. If it's over the wire, an invoice arrives. Don't pay the invoice, and thanks to some effective lobbying MICROS~1 now has the legal right to remotely shut the server down -- anyone want to bet that they didn't add the hooks to do it? They fought HARD for that clause, after all.
... there was a major computer manufacturer. The manufacturer had historically produced high-powered machinery with full-blown operating systems, but in recent years had seen some erosion of its sales at the low end from less-capable but much less expensive brands. The Big Company decided to get in on the new trend and introduced its own line of low-end computers.
Since its customers needed more than hardware, the Big Company offered their machines with a choice of operating software: users could have the industry standard or a new product made specifically for the new machines. The standard software naturally commanded a premium price: $40 extra.
Now according to your theory, users would have overwhelmingly preferred the standard, premium operating software over the freebie included with the machine. That's not what happened, though. Anyone recognize the story?
Hint: Monterey, California