I don't think it'll go on much longer. If this gets fast-tracked to the Supremes (and it will), they may elect to hear it in special session, or more likely in next year's docket. In any case they don't fool around with a lot of witnesses. That should put the end no further than 2003 as far as court rulings go. After that it's just shouting and throwing executives in jail for contempt.
Not as much as you think, if Judge J. rules before January. If this is fast-tracked (and it will be) to the Supreme Court, they'll probably throw out any whining that's not either a point of judicial error in procedure or an interesting point of law. That is, "too severe" probably won't be in their lexicon -- and they may refuse to hear the case if Judge J. doesn't slip up, and MS doesn't find a juicy precedent to dangle before them. Once their gavel falls, it's over.
[T]he financial community seems to agree on what the applications part is worth, but estimates of the value of the OS part are all over the place.
Agreed, but as long as all stockholders get a share of all new companies for each share of Microsoft, this works out -- the market will determine the price of the new shares in short order, and everyone gets to place their bets or stand pat. The only sticky part is when the officers divest -- how many shares of A do they trade for a share of B, and what does Bill do with his extra shares? This can be solved by permitting the companies to operate for 6 months or so under restrictions prior to the officers' divestiture -- let the market sort it out.
Let the MS Internet Company license (for fee) the browser to the MS OS company and the MS Apps company. The MSOS company can distribute it, and MSApps could integrate it, but they could decide to incorporate Mozilla instead...
Actually, you've compressed about a decade. DOJ entered a consent decree in 1985, complained around 1988, and spent the next 6 years winning and losing on appeal, and building the current case.
When a child whines and says "I'll be good! I'll be good", a parent is inclined to give them a chance to be good. When the child misbehaves again, it's time for punishment. Often the child is astonished that they're being punished, after having a second chance.
DOJ is getting ready to spank Microsoft good. Previously MS appealed saying "the agreement doesn't prohibit what we did" and the court agreed. This time there's no agreement, and they have to show why the judge misinterpereted the law. That'll be much harder to win.
Actually, he will. The court will first order shares in the new company issued to all stockholders 1 for 1. Then the court will order Bill (and all the other officers of the companies, and their boards) to sell (or swap with Steve) all his stock in all but one of the resulting companies. People with less than some percent of the companies can keep their shares in both.
Probably the current officers will just split up amongst the companies and swap stocks, to prevent crashing their value, so Steve & co. (the board et al) gets Bill's share of MS#1, and Bill gets Steve & co's share of MS#2, then Bill offers to swap the public for the remainder of his shares...
in reality there's another group that WE can't do without - the artists
Bingo. So support your local artists and help them not have to sign with a label to get their music out. Respect copyright and vote with your feet. I've yet to meet the artist or song that I couldn't do without, or pay for.
Encryption is a problem. Sadly simple open source encryption/players don't work - they always allow access to the raw bits at some level - I suspect what may end up working is some form of hardware decryption/playback where the bits are lost inside the electronics somewhere before they become analog.... (mp3 decryption/playback hardware anyone?).
Nah -- you just hook your digitizer to the speaker leads...
You're still part of the demand curve. If everyone who copies Metallica et al stopped listening, calling radio stations, etc, maybe you'd be taken seriously (I don't do commercial music anymore). By consuming the products (even if you don't pay) you're validating the RIAA's stand on "lost sales". The only way to prove you wouldn't buy it is not to own it (like me).
Oh, so if I have one gyroscope here in my town and I get you to start one turning in the opposite direction over in your town, and we both simultaneously turn them upside down, angular momentum adds to zero and there's no gyroscopic effect for either of us?
You're correct (the above example works), with reference to the system, from a point of view outside the system. But each of the counterrotating parts creates its own separate angular momentum, and the axle or frame that you do the adding with (in this example, our hands and bodies connected by the earth) has to be one strong MF to handle the kind of changes in angular momentum that go with rotating the axis of flywheels that can run a car... And you have the problem of coming up with bearings that take that stress without burning up or wobbling.
Sorry, I remain skeptical that such a system is practical for those kinds of flywheels.
No, the article mentions using counterrotating flywheels to control torsional reactions when spinning them up or down. I watched carefully, but no convincing mechanism was given for handling the axis problem.
Another source (EV World) suggest counterrotating cylinders will have "less angular momentum". Correct me if I'm wrong (and provide calculations) but doesn't that mean "less energy"? As I read it, they're just moving the mechanical problem to the flywheel body and axle, but if you store enough energy in the darn thing, you'll still wreck the car if you turn too fast:-).
I suspect gimbals will be an inescapable complication for real world mobile systems, if the high-RPM route is taken.
That sounds like security to me. Why should Microsoft allow authentication to wobbly resources that could come from anywhere?
Ignoring the hyperbole that follows that quote -- the cryptography prevents the authentication from coming from "just anywhere". Microsoft apps should roll over and trust the KDC -- that's what it's there for: why you set it up. The bit the MS server wants is your ACL, which is about permission, rather than authentication. The MS server should get it from a trusted source, if it's not going to hold onto it itself, rather than depend on an optional field. By making the field non-optional for MS servers, MS breaks the standard. DCE does too (IIRC), but they admit it. Breaking the standard is not wrong, but claiming you didn't, when you did, is. I'll sputter all day about that (apparently:-).
The key to this is that Ford doesn't weld the hood shut and forbid you from looking under it, or adding a turbocharger, or using your own paint to put racing stripes on it or taking it apart and building a race car from the parts. And if your Pinto explodes, you can sue.
A software license agreement that is not "free" in the sense of Mr. Stallman or Mr. Raymond (they do have common ground) forbids you to alter the product and give it to someone else, even if you give them the sole copy for which you purchased a license. Furthermore, you cannot easily inspect the product to determine if its failure to perform is due to an error or deliberate design.
Personally, I think software liability should be enforced upon those who deliver software without the privilege of changing it. It's ok to make a living from your software, if you want, but you should take responsibility for its quality and manufacture, if you intend to represent it as a product that has value. If you don't so represent it, you shouldn't charge for it...
Perversely, I also think it's perfectly fair to restrict copying, if you follow those guidelines. The immorality I see is the one you find in RMS's printer story -- representation of value in a product (the printer) without responsibility to repair or replace a defective part (software), and provide for its repair and replacement should you yourself become unable. Many companies that create real products make agreements with other companies to provide those services for those products should they go out of business.
To me, it seems doing so would break MS's propietary version and place them in a situation where they must either drop their own, propietary extension, or lock access to only Microsoft products.
Well, that's kind of what's going on. MS has found a way to make Kerberos clients accept their authority, without having to accept the authority of a Kerberos KDC (the master key server). So you have to have a Microsoft KDC that clues their boxes in, in order to use Kerberos in the mixed environment. They're wedging themselves into the controlling seat.
That way they can come out with a new more-non-standard service that pulls more of the non-MS servers out. As it is, the marketing line will be "hey, you're implementing a Microsoft Active Directory anyway, why not replace those old servers while you're at it".
To me, the MS lock-in is ok, if you sell it that way. It's lying about compatibility (or misleading about interoperability) that sets me off. MS is pitching to folks who, as you say, don't understand the protocol. They say "see? We support Kerberos. We'll play nice with your secure setup." But only if you let them be King of the Mountain.
Only if MS's OS and apps are split will it be in the best interests of the OS company to publish API's. If there's competition on the OS front, as the amicus breif from the economists suggested, then the OS company will be kept honest by its peers (nothing like your competition watching you to keep your pants up). Competition in the OS market would keep the Apps company from wanting to use non-standard API's -- heck, the shifting sands drive me crazy and I only have my employer for a customer.
No, actually it's not. It walks like Kerberos, but it doesn't quack like Kerberos. It permits Kerberos clients to authenticate using a Microsoft KDC, but it does not permit use of Microsoft services if you use another KDC. If it were Kerberos, it would.
If there's only one datafield used by MS to do the authentication thing, wouldn't this be quite easy to hack / reverse engeneer?
Well, no. See, the common use of Kerberos (and the original standard) is for authentication only -- knowing that you're you. MS, in order to let you use their services, is asking the question "Ok, you're you, but what can you do?". They're embedding the answer (your ACL) in the Kerberos header, but that means that you can't use any other Kerberos servers to talk to a MS box, since none of them can generate that info in the first place. A "proper" implementation would issue a ticket from the MS data server with the ACLs embedded, or better yet handle it as a separate datum, rather than forcing you to use an MS Keyserver.
Just because people disagree with you, it doesn't mean they're wrong or ignorant.
Well, if they disagree with me about my hair color, they are.
Open API's, published specs and open sourcing the current versions of software won't prevent things like the Kerberos maneuver in the future. Making a competitive situation will. I happen to agree with the 4-company plan, but even the 2-way split will make a competetive application company and a regulated monopoly OS company. That's a much more acceptable situation than an unregulated company with a monopoly...
Let's face it, even with an open source version of Windows 2000, all it takes is an Office bugfix that only works on the proprietary Windows 2002 to get us back to court.
I am referring here to the IBM anti-trust case (which was eventually dismissed),[...]
It's worth noting that it was dismissed with a consent decree that bound IBM to stop certain business practises, and to obey other restrictions (such as not discussing products until they were ready to ship) for a certain time. IBM and a lot of companies have knuckled under once it was clear that they would lose an anti-trust action. Part of MS's stupidity (or brilliance, leave that to history) has been their willingness to break such agreements and behave just as they please, even while the watchdogs are nipping at their heels.
Take it from a software geek. Lack of modular design in anything as complex as today's hit games is Major Overhead.
Since you pretty much have to split out render, sound and gameplay engines anyway, making them cross platform may be easier than you think. Multiple platforms can also point up platform-specific bugs, shortening debug time (i.e., is it me or DirectDraw? Well, the Linux port is working so it must be either in this module or DirectDraw....). Look at the developer logs at Station Terminus -- they ported to Linux and Mac in a couple of weeks (on a 3 year project - see 7/2). Pennies indeed.
First, if anyone has pointers to heavier articles, please put 'em up. I found a couple of discussions, but I lost 'em...
It seems that what Prof. Yilmaz is suggesting is that the potential energy of gravitation be included in the energy side of the equation (yes, adding an extra term). To my non-mathematical visual (i.e. probably wrong) conception this implies that a very strong gravitational field bends space more under GR-Yilmaz than standard GR, meaning there might be more space at the bottom of the well, and matter might not have to compress so small to appear to occupy such a small space. According to those who have done the math, it already predicts 10xSolar mass neutron stars, so a "black hole" would have more mass to get started with...
It would have been nice to be able to open the folder I was in the CLI in the GUI, and vice versa
>gmc `pwd`
All things are possible with a good desktop interface, but your point is taken. If we could point and type w/o hands reorienting all the time that would be a lot more powerful.
Slashdot Mirror
I don't think it'll go on much longer. If this gets fast-tracked to the Supremes (and it will), they may elect to hear it in special session, or more likely in next year's docket. In any case they don't fool around with a lot of witnesses. That should put the end no further than 2003 as far as court rulings go. After that it's just shouting and throwing executives in jail for contempt.
Not as much as you think, if Judge J. rules before January. If this is fast-tracked (and it will be) to the Supreme Court, they'll probably throw out any whining that's not either a point of judicial error in procedure or an interesting point of law. That is, "too severe" probably won't be in their lexicon -- and they may refuse to hear the case if Judge J. doesn't slip up, and MS doesn't find a juicy precedent to dangle before them. Once their gavel falls, it's over.
Agreed, but as long as all stockholders get a share of all new companies for each share of Microsoft, this works out -- the market will determine the price of the new shares in short order, and everyone gets to place their bets or stand pat. The only sticky part is when the officers divest -- how many shares of A do they trade for a share of B, and what does Bill do with his extra shares? This can be solved by permitting the companies to operate for 6 months or so under restrictions prior to the officers' divestiture -- let the market sort it out.
Let the MS Internet Company license (for fee) the browser to the MS OS company and the MS Apps company. The MSOS company can distribute it, and MSApps could integrate it, but they could decide to incorporate Mozilla instead...
When a child whines and says "I'll be good! I'll be good", a parent is inclined to give them a chance to be good. When the child misbehaves again, it's time for punishment. Often the child is astonished that they're being punished, after having a second chance.
DOJ is getting ready to spank Microsoft good. Previously MS appealed saying "the agreement doesn't prohibit what we did" and the court agreed. This time there's no agreement, and they have to show why the judge misinterpereted the law. That'll be much harder to win.
Actually, he will. The court will first order shares in the new company issued to all stockholders 1 for 1. Then the court will order Bill (and all the other officers of the companies, and their boards) to sell (or swap with Steve) all his stock in all but one of the resulting companies. People with less than some percent of the companies can keep their shares in both.
Probably the current officers will just split up amongst the companies and swap stocks, to prevent crashing their value, so Steve & co. (the board et al) gets Bill's share of MS#1, and Bill gets Steve & co's share of MS#2, then Bill offers to swap the public for the remainder of his shares...
Bingo. So support your local artists and help them not have to sign with a label to get their music out. Respect copyright and vote with your feet. I've yet to meet the artist or song that I couldn't do without, or pay for.
Encryption is a problem. Sadly simple open source encryption/players don't work - they always allow access to the raw bits at some level - I suspect what may end up working is some form of hardware decryption/playback where the bits are lost inside the electronics somewhere before they become analog .... (mp3 decryption/playback hardware anyone?).
Nah -- you just hook your digitizer to the speaker leads...
You're still part of the demand curve. If everyone who copies Metallica et al stopped listening, calling radio stations, etc, maybe you'd be taken seriously (I don't do commercial music anymore). By consuming the products (even if you don't pay) you're validating the RIAA's stand on "lost sales". The only way to prove you wouldn't buy it is not to own it (like me).
You're correct (the above example works), with reference to the system, from a point of view outside the system. But each of the counterrotating parts creates its own separate angular momentum, and the axle or frame that you do the adding with (in this example, our hands and bodies connected by the earth) has to be one strong MF to handle the kind of changes in angular momentum that go with rotating the axis of flywheels that can run a car... And you have the problem of coming up with bearings that take that stress without burning up or wobbling.
Sorry, I remain skeptical that such a system is practical for those kinds of flywheels.
Another source (EV World) suggest counterrotating cylinders will have "less angular momentum". Correct me if I'm wrong (and provide calculations) but doesn't that mean "less energy"? As I read it, they're just moving the mechanical problem to the flywheel body and axle, but if you store enough energy in the darn thing, you'll still wreck the car if you turn too fast :-).
I suspect gimbals will be an inescapable complication for real world mobile systems, if the high-RPM route is taken.
Just don't take any of the pins out of it, whatever you do...
Zot! You got me. Still, how many Gameboy programs work on those calculators? There's gotta be some architectural differences there...
Ignoring the hyperbole that follows that quote -- the cryptography prevents the authentication from coming from "just anywhere". Microsoft apps should roll over and trust the KDC -- that's what it's there for: why you set it up. The bit the MS server wants is your ACL, which is about permission, rather than authentication. The MS server should get it from a trusted source, if it's not going to hold onto it itself, rather than depend on an optional field. By making the field non-optional for MS servers, MS breaks the standard. DCE does too (IIRC), but they admit it. Breaking the standard is not wrong, but claiming you didn't, when you did, is. I'll sputter all day about that (apparently :-).
A software license agreement that is not "free" in the sense of Mr. Stallman or Mr. Raymond (they do have common ground) forbids you to alter the product and give it to someone else, even if you give them the sole copy for which you purchased a license. Furthermore, you cannot easily inspect the product to determine if its failure to perform is due to an error or deliberate design.
Personally, I think software liability should be enforced upon those who deliver software without the privilege of changing it. It's ok to make a living from your software, if you want, but you should take responsibility for its quality and manufacture, if you intend to represent it as a product that has value. If you don't so represent it, you shouldn't charge for it...
Perversely, I also think it's perfectly fair to restrict copying, if you follow those guidelines. The immorality I see is the one you find in RMS's printer story -- representation of value in a product (the printer) without responsibility to repair or replace a defective part (software), and provide for its repair and replacement should you yourself become unable. Many companies that create real products make agreements with other companies to provide those services for those products should they go out of business.
Well, that's kind of what's going on. MS has found a way to make Kerberos clients accept their authority, without having to accept the authority of a Kerberos KDC (the master key server). So you have to have a Microsoft KDC that clues their boxes in, in order to use Kerberos in the mixed environment. They're wedging themselves into the controlling seat.
That way they can come out with a new more-non-standard service that pulls more of the non-MS servers out. As it is, the marketing line will be "hey, you're implementing a Microsoft Active Directory anyway, why not replace those old servers while you're at it".
To me, the MS lock-in is ok, if you sell it that way. It's lying about compatibility (or misleading about interoperability) that sets me off. MS is pitching to folks who, as you say, don't understand the protocol. They say "see? We support Kerberos. We'll play nice with your secure setup." But only if you let them be King of the Mountain.
Only if MS's OS and apps are split will it be in the best interests of the OS company to publish API's. If there's competition on the OS front, as the amicus breif from the economists suggested, then the OS company will be kept honest by its peers (nothing like your competition watching you to keep your pants up). Competition in the OS market would keep the Apps company from wanting to use non-standard API's -- heck, the shifting sands drive me crazy and I only have my employer for a customer.
Don't forget marketing....
No, actually it's not. It walks like Kerberos, but it doesn't quack like Kerberos. It permits Kerberos clients to authenticate using a Microsoft KDC, but it does not permit use of Microsoft services if you use another KDC. If it were Kerberos, it would.
Well, no. See, the common use of Kerberos (and the original standard) is for authentication only -- knowing that you're you. MS, in order to let you use their services, is asking the question "Ok, you're you, but what can you do?". They're embedding the answer (your ACL) in the Kerberos header, but that means that you can't use any other Kerberos servers to talk to a MS box, since none of them can generate that info in the first place. A "proper" implementation would issue a ticket from the MS data server with the ACLs embedded, or better yet handle it as a separate datum, rather than forcing you to use an MS Keyserver.
Well, if they disagree with me about my hair color, they are.
Open API's, published specs and open sourcing the current versions of software won't prevent things like the Kerberos maneuver in the future. Making a competitive situation will. I happen to agree with the 4-company plan, but even the 2-way split will make a competetive application company and a regulated monopoly OS company. That's a much more acceptable situation than an unregulated company with a monopoly...
Let's face it, even with an open source version of Windows 2000, all it takes is an Office bugfix that only works on the proprietary Windows 2002 to get us back to court.
It's worth noting that it was dismissed with a consent decree that bound IBM to stop certain business practises, and to obey other restrictions (such as not discussing products until they were ready to ship) for a certain time. IBM and a lot of companies have knuckled under once it was clear that they would lose an anti-trust action. Part of MS's stupidity (or brilliance, leave that to history) has been their willingness to break such agreements and behave just as they please, even while the watchdogs are nipping at their heels.
Actually, they just proposed that the judge might order them not to do it in the future, depending on what "future" means.
Since you pretty much have to split out render, sound and gameplay engines anyway, making them cross platform may be easier than you think. Multiple platforms can also point up platform-specific bugs, shortening debug time (i.e., is it me or DirectDraw? Well, the Linux port is working so it must be either in this module or DirectDraw....). Look at the developer logs at Station Terminus -- they ported to Linux and Mac in a couple of weeks (on a 3 year project - see 7/2). Pennies indeed.
It seems that what Prof. Yilmaz is suggesting is that the potential energy of gravitation be included in the energy side of the equation (yes, adding an extra term). To my non-mathematical visual (i.e. probably wrong) conception this implies that a very strong gravitational field bends space more under GR-Yilmaz than standard GR, meaning there might be more space at the bottom of the well, and matter might not have to compress so small to appear to occupy such a small space. According to those who have done the math, it already predicts 10xSolar mass neutron stars, so a "black hole" would have more mass to get started with...
>gmc `pwd`
All things are possible with a good desktop interface, but your point is taken. If we could point and type w/o hands reorienting all the time that would be a lot more powerful.