If you persistently tell people they should be afraid -- they WILL be. It matters not at all whether they SHOULD be.
Witness that, lacking both better things to do and the ethics to do better things, our American news media plays up every negative incident as OMG the sky is falling, run for your lives!! Consequently, ask the average American (or any of our detractors) whether they think violent crime is out of control in the U.S., and they will uniformly declare that it is -- despite that the *actual* incidence of violent crime has been dropping steadily for almost two decades.
One of the reasons I like RaidMax cases is because the HD bays (and their midtowers have a total of 10 drive bays!) enforce half an inch of free space between drives, so even if you fill ALL the drive bays there is still room for airflow (and the side-mounted case fan swishes air through them well enough, too).
This is in direct contrast to the typical case (cheap or not!) that unless you only use half the drive bays, crams the HDs together cheek-by-jowl, literally touching one another. So they're baking in one another's heat, and said heat can't dissipate properly even if a fan is blowing directly on the HDs.
I remember this point from Google's survey of some 100,000 of their own HDs: HDs that run hot have an average lifespan of about 3 years, whereas HDs that run cool have a lifespan of 5 or more years. This exactly matches my own observation of typical system failure rates -- the typical OEM case is DESIGNED to retain heat, and if the OEM system's HD fails, it will do so at about 3 years. Conversely, HDs in clone machines (which usually are not so "efficiently engineered" and spread stuff around inside the case, and are typically MUCH better vented overall) usually survive 5 years or longer.
Don't believe that OEMs design 'em to run hot? I have a Dell P4 here, one of their more-expensive models ($4000 new -- it was given to me because the previous owner got sick of its problems). With stock cooling, which consisted of one case fan and a shroud aimed at the CPU, it ran so hot that it was unstable. I removed the shroud, and added a proper CPU heatsink/fan** and a case fan. Its running temperature dropped 40F DEGREES, and it stopped crashing.
** You can buy a standard HSF to fit the nonstandard Dell mounting arrangement for about $15 from tekgems.com
The plastic sheathing that's always been on most OEMs and is now popular on clone cases is another problem. Metal dissipates heat, and a large portion (sometimes the majority) of system cooling is heat transfer *through* the metal case. But plastic is an insulator, and the dead airspace between metal interior and plastic exterior is ALSO an insulator. Take a standard oldfashioned metal case, throw a towel over it (don't cover the front or back, only the sides and top) and watch the temperature skyrocket. Then consider whether plastic sheathing is such a wonderful idea after all.
HDs don't like running at temps below about 60F either, but that's not usually an issue, unless you're an Eskimo:)
I try to keep my system innards somewhere in the 35C range, and there's little doubt in my mind that this helps longevity. (I don't rush out and buy new shit all the time; I use old paid-for shit til it either dies or is no longer of any use for anything.) I'm writing this on a box whose major innards are almost 10 years old.:)
How do people take notes? They hear a speaker, they see a presentation, they scribble down content and context as best they can, but even under ideal circumstances, class notes are a summary and approximation, and are therefore always primarily in the student's own words -- exactly like what a reporter does in the field. There will be snatches of exact quotes (which do fall under Fair Use), but those will NOT make up the main body of the notes. Even someone proficient at shorthand can't keep up well enough to do a full quote of a lecturer.
Conversely, a Reader's Digest Condensed Book (I remember them well:) starts directly from a clean copy of the original work, and excises portions therefrom. It is NOT "notes"; it's simply a selection of the original text. I've read some in both RD and original form, and speaking as an experienced editor myself -- RD's technique is almost entirely to omit sections and excise references to the omitted sections. They do almost no rewriting. So yes, RDCBs are very different from class notes (and from Cliff's Notes).
Hmm... I've been using those "under the HD" fans for 10 years; they REALLY help when a HD runs a bit hot and there's no good way to mount a fan blowing directly across the HD. (Not every PC case was designed by someone who actually puts components inside it.:)
Even with one that developed Rattly Worn-out Bearings, I didn't observe any wobble (and the HD under it outlived two of those fans, and was just retired at age 8) -- but my HD mounts are solid and don't wiggle even if you bang on them. Some mounts are so cheapassed that they vibrate just from the HD running, and in those cases you're probably right.
I live where the power spikes and sags constantly. My machines are all on UPSs. And each PC has a decent quality PSU. And if a HD runs more than "pleasantly warm" to the touch, it gets its own dedicated fan. Consequently, I firmly believe all HDs are supposed to live A Long Time... the oldest of my 24/7 HDs right now is 10 years old, and has about 80,000 actual hours on it -- Like yourself, I think they're supposed to be worn out before being thrown out.:)
Of course, yonder is a large stack of backups, which also help increase HD longevity.;)
In 15 years with PCs, I've had 3 everyday-use HDs fail, but between 'em they were almost old enough to vote, and ran 24/7 most of their lives, and they all staggered along for a year or more after the first symptoms:
2GB W.D., got headcrashed when I moved and developed the creeping crud; 3 years later, at age 5, it finally got to where it lost the partition table and I had to replace it.
6GB W.D., at age 7 the thermal calibration function developed a habit of sticking on. No bad sectors, tho.
45GB W.D., retired at age 8 (with almost 55,000 actual hours on it) because the tail end of the drive had the creeping crud (tho my *real* motivation for replacing it was that I was out of disk space). Other than that last gig, it was still perfect. (Most likely it got headcrashed when I had the machine apart to insert the network card. I'm careful, but sometimes eggs still break.) This one ran a little hot by my standards (more than pleasantly-warm to the touch), so spent its life with its own dedicated fan, which I expect helped its longevity.
In clients' machines and donations to our user group, I seldom see sick or dead HDs, and when I do, it's usually in an OEM 1) with a really cheapassed power supply and 2) where the HD is crammed into an unventilated corner and is being cooked in its own juice.
Anyway, I agree with you... HDs' reputation for failure seems to be rather worse than the reality. Most of the so-called HD failures I've seen were actually fucked-up filesystems or trashed MBRs, nothing to do with the hardware itself.
Plugging in my own HDs (usually 6 HDs going 24/7) the annual failure number comes out to about 0.05, which is a little low, but not ridiculously so -- in the Real World, I have a HD failure about once every 5 years.
Hmm... actually, that's right on as an average. My first computer with a HD arrived in 1993. I just replaced the 3rd HD I've ever had get really sick. (Tho it was 8 years old.)
But when you take notes, you DO rewrite the presented facts in your own format -- that's the whole point here, that said notes are a new work based on the same facts, NOT a derivative work of the professor's notes.
A ruling in his favour would also eliminate products like Cliff's Notes.
Erm... if it's infringement for a 3rd party to reproduce notes taken in class, then it's equally infringement for students in the class to take notes and copy them among themselves. That's why this is so important -- as you say it has nothing to do with knowledge, but it impacts severely on whether one is allowed to redistribute knowledge, and the implications are a lot wider than this single case.
Copyright is NOT the right to sell; it's the right to distribute.
You're failing to see how this generalizes, or how far-reaching it could be:
What does a newspaper reporter DO?
He listens to someone else's words and takes notes, which the newspaper subsequently publishes and sells to various individuals who had nothing to do with the original speaker. (And sometimes the reporter is independent, not a direct employee of the newspaper.)
This is EXACTLY what this note-taking/publishing company is doing.
Just because the ultimate source is a professor, who plans to publish his own version of these notes, makes it no different than if a streetcorner preacher, whose words are summarized by a reporter and printed by a newspaper, had plans to publish his views.
In fact, a professor's words would have LESS protection, because FACTS are not copyrightable.
IF a court agrees with the professor, then ANY time ANYONE speaks where another might hear, take notes, and pass those notes on to anyone else (which is the essense of all news-journalism, including many blogs and forums like slashdot), becomes a copyright lawsuit waiting to happen. Remember that copyright is not the right to SELL; it is the right to REPRODUCE.
If someone stole the professor's original notes, photocopied, and published them, then he'd have a case.
But as others point out, when you go into certain professions, you may give up some rights. Frex, if you're a photographer working for NASA, all images taken for NASA are public domain; you cannot copyright them. If you're a professor, whose goal is to impart your knowledge to others, why on earth should said knowledge be restricted? From a standpoint of "information wants to be free", said knowledge should be public domain by definition. Sure, you have to pay to access the *distribution system* (ie. to attend the university) but once you acquire that knowledge, it should be yours without restrictions of any kind.
Another evil side effect of a ruling in this guy's favour could ultimately be more-general restrictions on what you are allowed to do with knowledge garnered through paid education (ie. what you learn in university classes). I leave the implications of that as an exercise to the reader.
Not only that (I was going to make the same point), but if a court were to find otherwise, it has dire implications: Any streetcorner yammerer could claim that the mere act of *speaking* endowed their words with copyright, and they could then sue anyone who reported a summary of those words. Bye-bye newspapers, whose primary JOB is to summarize and publish other people's words!
I was just wondering something like that myself: at what point does this sort of datamining overstep privacy to the point that it revokes common carrier status??
Also, AOL's little fiasco proved that you CAN identify individuals through their searches... what's to prevent this from being used similarly?
I also have to wonder about what if the ISP is "clean" but their backbone is datamining??
[goes off, reads link] "64-bit versions of Windows do not include the 16-bit WoW subsystem or NTVDM and therefore cannot run 16-bit Windows or DOS applications."
Well, that kills 64bit Windows for me... there are still a few ancient apps (mostly DOS, but one Win16) that I can't live without and haven't found any reasonable replacement for.
Ah yes, I'd forgotten about the reverse engineering aspect... yeah, I suppose the best way to test whether WINE is getting it right is to put it into direct parallel with Windows, inside of Windows. But it's still funny:)
Vista was a kludge; and I'm sure M$ is likewise sick and tired of having to test against a bazillion possibilities, hence this concept. And I expect with modular-Windows you won't have to test near as much, because your app will only need to speak to the core OS, and won't care about any other crap. Fewer layers, less hassle. Think sideways rather than vertical.
But yeah, to make cranky old app speak to shiny new Windows, I can see using WINE instead of a VM... I'm not sure why, but maybe it'd work better for certain things. Anyone here know enough to comment intelligently?:)
I don't think M$ feels as threatened by apps being ported away as some folks think. All you have to do is look at Wine... yeah, it works, but is it up to what enterprise business (M$'s REAL customers) needs? Highly doubtful.
In fact, I think part of why M$ is looking at this scheme is because they know damn well that most businesses would rather buy a monolithic solution, and a modular OS makes it easier for M$ to customize that solution and thereby maximize profit.
To refute my own post [g] the main reason I'd put forth as to why M$ probably wouldn't go to an outright UNIX kernel, is because they have a known entity in Windows; why throw out two decades of expertise with their own product? And the Windows core is really a pretty good OS in itself; it goes wrong when too much crap gets shoveled in on top. If we no longer bury it in junk, it can better do its job.
Exactly. I don't see any downside for anyone. I see a lot *more* opportunity for money to flow in all directions, and not just toward M$.
I would hope that drivers would be developed at more than "their own pace", tho, because otherwise only a few popular components will have a chance. One of the reasons so many different types and brands of hardware are in the market today is because for most, drivers are available NOW, no waiting (at least for Windows. Linux, not so much.) Buy it, install it, works right now. Exceptions are not well-received in the market.
Small 3rd party developers could partner with OEMs or even with M$ and build up a repository of both free and pay-to-install software, rather akin to what Linspire originally did. This is an opportunity for apps that are presently shut out because there's just not a NEED for 'em when Windows already includes the kitchen sink. As to pricing, market forces will decide what's worth paying for. And if something gets pirated a lot... well, maybe it was overpriced.
As to updates, there's really no reason to complicate it by making into pay-per-level or pay-per-update -- why would I want an update that addresses software I didn't install in the first place? and unless an update is a wholesale replacement for a component, it's not going to install a complete app anyway, so it's of no use to someone trying to pirate a level of Windows they didn't pay for. Hell, updates could be unified and only install required packages per autodetect. -- This also prevents the disaster that's currently going on, where my non-WinXP system can't download updates for my WinXP customer whose machine is presently borkend.
Slashdot Mirror
And there's the question requiring a yes/no answer, for which *either* answer is an indictment.
Frex, "Have you quit beating your wife?"
If you answer "Yes," you've admitted that you used to beat your wife.
If you answer "No," you've admitted that you still beat your wife.
So, both YES and NO mean you're guilty. How is an innocent person supposed to answer such a question without experiencing stress?
(Some wag once postulated that the correct answer was "Which one?")
Which also ensures a monoculture and a uniform point of failure when (more likely than if) their custom setup is compromised.
Joseph Stalin, is that you??
If you persistently tell people they should be afraid -- they WILL be. It matters not at all whether they SHOULD be.
Witness that, lacking both better things to do and the ethics to do better things, our American news media plays up every negative incident as OMG the sky is falling, run for your lives!! Consequently, ask the average American (or any of our detractors) whether they think violent crime is out of control in the U.S., and they will uniformly declare that it is -- despite that the *actual* incidence of violent crime has been dropping steadily for almost two decades.
See stats at http://www.ojp.usdoj.gov/bjs/gvc.htm
[gazing at Statue of Liberty, hip-deep in the sands of time]
You were right. They ruined it. We've become the nation of Soylent Green.
And since they all do so incessently, there is SO much shit to quote and paste, the work never ends!!
"Bloggers have learned all they know from reading one another."
(Apologies to Jack Vance... the original was about critics, but bloggers are much the same thing.)
Which makes it sound sortof like spammers' similar complaints, except in this case, readers seek the spam rather than having it thrust upon them!
One of the reasons I like RaidMax cases is because the HD bays (and their midtowers have a total of 10 drive bays!) enforce half an inch of free space between drives, so even if you fill ALL the drive bays there is still room for airflow (and the side-mounted case fan swishes air through them well enough, too).
:)
:)
This is in direct contrast to the typical case (cheap or not!) that unless you only use half the drive bays, crams the HDs together cheek-by-jowl, literally touching one another. So they're baking in one another's heat, and said heat can't dissipate properly even if a fan is blowing directly on the HDs.
I remember this point from Google's survey of some 100,000 of their own HDs: HDs that run hot have an average lifespan of about 3 years, whereas HDs that run cool have a lifespan of 5 or more years. This exactly matches my own observation of typical system failure rates -- the typical OEM case is DESIGNED to retain heat, and if the OEM system's HD fails, it will do so at about 3 years. Conversely, HDs in clone machines (which usually are not so "efficiently engineered" and spread stuff around inside the case, and are typically MUCH better vented overall) usually survive 5 years or longer.
Don't believe that OEMs design 'em to run hot? I have a Dell P4 here, one of their more-expensive models ($4000 new -- it was given to me because the previous owner got sick of its problems). With stock cooling, which consisted of one case fan and a shroud aimed at the CPU, it ran so hot that it was unstable. I removed the shroud, and added a proper CPU heatsink/fan** and a case fan. Its running temperature dropped 40F DEGREES, and it stopped crashing.
** You can buy a standard HSF to fit the nonstandard Dell mounting arrangement for about $15 from tekgems.com
The plastic sheathing that's always been on most OEMs and is now popular on clone cases is another problem. Metal dissipates heat, and a large portion (sometimes the majority) of system cooling is heat transfer *through* the metal case. But plastic is an insulator, and the dead airspace between metal interior and plastic exterior is ALSO an insulator. Take a standard oldfashioned metal case, throw a towel over it (don't cover the front or back, only the sides and top) and watch the temperature skyrocket. Then consider whether plastic sheathing is such a wonderful idea after all.
HDs don't like running at temps below about 60F either, but that's not usually an issue, unless you're an Eskimo
I try to keep my system innards somewhere in the 35C range, and there's little doubt in my mind that this helps longevity. (I don't rush out and buy new shit all the time; I use old paid-for shit til it either dies or is no longer of any use for anything.) I'm writing this on a box whose major innards are almost 10 years old.
How do people take notes? They hear a speaker, they see a presentation, they scribble down content and context as best they can, but even under ideal circumstances, class notes are a summary and approximation, and are therefore always primarily in the student's own words -- exactly like what a reporter does in the field. There will be snatches of exact quotes (which do fall under Fair Use), but those will NOT make up the main body of the notes. Even someone proficient at shorthand can't keep up well enough to do a full quote of a lecturer.
:) starts directly from a clean copy of the original work, and excises portions therefrom. It is NOT "notes"; it's simply a selection of the original text. I've read some in both RD and original form, and speaking as an experienced editor myself -- RD's technique is almost entirely to omit sections and excise references to the omitted sections. They do almost no rewriting. So yes, RDCBs are very different from class notes (and from Cliff's Notes).
Conversely, a Reader's Digest Condensed Book (I remember them well
Hmm... I've been using those "under the HD" fans for 10 years; they REALLY help when a HD runs a bit hot and there's no good way to mount a fan blowing directly across the HD. (Not every PC case was designed by someone who actually puts components inside it. :)
Even with one that developed Rattly Worn-out Bearings, I didn't observe any wobble (and the HD under it outlived two of those fans, and was just retired at age 8) -- but my HD mounts are solid and don't wiggle even if you bang on them. Some mounts are so cheapassed that they vibrate just from the HD running, and in those cases you're probably right.
I live where the power spikes and sags constantly. My machines are all on UPSs. And each PC has a decent quality PSU. And if a HD runs more than "pleasantly warm" to the touch, it gets its own dedicated fan. Consequently, I firmly believe all HDs are supposed to live A Long Time... the oldest of my 24/7 HDs right now is 10 years old, and has about 80,000 actual hours on it -- Like yourself, I think they're supposed to be worn out before being thrown out. :)
;)
Of course, yonder is a large stack of backups, which also help increase HD longevity.
In 15 years with PCs, I've had 3 everyday-use HDs fail, but between 'em they were almost old enough to vote, and ran 24/7 most of their lives, and they all staggered along for a year or more after the first symptoms:
2GB W.D., got headcrashed when I moved and developed the creeping crud; 3 years later, at age 5, it finally got to where it lost the partition table and I had to replace it.
6GB W.D., at age 7 the thermal calibration function developed a habit of sticking on. No bad sectors, tho.
45GB W.D., retired at age 8 (with almost 55,000 actual hours on it) because the tail end of the drive had the creeping crud (tho my *real* motivation for replacing it was that I was out of disk space). Other than that last gig, it was still perfect. (Most likely it got headcrashed when I had the machine apart to insert the network card. I'm careful, but sometimes eggs still break.) This one ran a little hot by my standards (more than pleasantly-warm to the touch), so spent its life with its own dedicated fan, which I expect helped its longevity.
In clients' machines and donations to our user group, I seldom see sick or dead HDs, and when I do, it's usually in an OEM 1) with a really cheapassed power supply and 2) where the HD is crammed into an unventilated corner and is being cooked in its own juice.
Anyway, I agree with you... HDs' reputation for failure seems to be rather worse than the reality. Most of the so-called HD failures I've seen were actually fucked-up filesystems or trashed MBRs, nothing to do with the hardware itself.
Thanks, that explains it nicely.
Plugging in my own HDs (usually 6 HDs going 24/7) the annual failure number comes out to about 0.05, which is a little low, but not ridiculously so -- in the Real World, I have a HD failure about once every 5 years.
Hmm... actually, that's right on as an average. My first computer with a HD arrived in 1993. I just replaced the 3rd HD I've ever had get really sick. (Tho it was 8 years old.)
But when you take notes, you DO rewrite the presented facts in your own format -- that's the whole point here, that said notes are a new work based on the same facts, NOT a derivative work of the professor's notes.
A ruling in his favour would also eliminate products like Cliff's Notes.
Well, they'd just have to fix that sad deficiency to be marketable, eh?
Erm... if it's infringement for a 3rd party to reproduce notes taken in class, then it's equally infringement for students in the class to take notes and copy them among themselves. That's why this is so important -- as you say it has nothing to do with knowledge, but it impacts severely on whether one is allowed to redistribute knowledge, and the implications are a lot wider than this single case.
:)
Copyright is NOT the right to sell; it's the right to distribute.
(Where is Cpt. Kangarooski when we need him?
You're failing to see how this generalizes, or how far-reaching it could be:
What does a newspaper reporter DO?
He listens to someone else's words and takes notes, which the newspaper subsequently publishes and sells to various individuals who had nothing to do with the original speaker. (And sometimes the reporter is independent, not a direct employee of the newspaper.)
This is EXACTLY what this note-taking/publishing company is doing.
Just because the ultimate source is a professor, who plans to publish his own version of these notes, makes it no different than if a streetcorner preacher, whose words are summarized by a reporter and printed by a newspaper, had plans to publish his views.
In fact, a professor's words would have LESS protection, because FACTS are not copyrightable.
IF a court agrees with the professor, then ANY time ANYONE speaks where another might hear, take notes, and pass those notes on to anyone else (which is the essense of all news-journalism, including many blogs and forums like slashdot), becomes a copyright lawsuit waiting to happen. Remember that copyright is not the right to SELL; it is the right to REPRODUCE.
If someone stole the professor's original notes, photocopied, and published them, then he'd have a case.
But as others point out, when you go into certain professions, you may give up some rights. Frex, if you're a photographer working for NASA, all images taken for NASA are public domain; you cannot copyright them. If you're a professor, whose goal is to impart your knowledge to others, why on earth should said knowledge be restricted? From a standpoint of "information wants to be free", said knowledge should be public domain by definition. Sure, you have to pay to access the *distribution system* (ie. to attend the university) but once you acquire that knowledge, it should be yours without restrictions of any kind.
Another evil side effect of a ruling in this guy's favour could ultimately be more-general restrictions on what you are allowed to do with knowledge garnered through paid education (ie. what you learn in university classes). I leave the implications of that as an exercise to the reader.
Not only that (I was going to make the same point), but if a court were to find otherwise, it has dire implications: Any streetcorner yammerer could claim that the mere act of *speaking* endowed their words with copyright, and they could then sue anyone who reported a summary of those words. Bye-bye newspapers, whose primary JOB is to summarize and publish other people's words!
I was just wondering something like that myself: at what point does this sort of datamining overstep privacy to the point that it revokes common carrier status??
Also, AOL's little fiasco proved that you CAN identify individuals through their searches... what's to prevent this from being used similarly?
I also have to wonder about what if the ISP is "clean" but their backbone is datamining??
[goes off, reads link]
:)
"64-bit versions of Windows do not include the 16-bit WoW subsystem or NTVDM and therefore cannot run 16-bit Windows or DOS applications."
Well, that kills 64bit Windows for me... there are still a few ancient apps (mostly DOS, but one Win16) that I can't live without and haven't found any reasonable replacement for.
Ah yes, I'd forgotten about the reverse engineering aspect... yeah, I suppose the best way to test whether WINE is getting it right is to put it into direct parallel with Windows, inside of Windows. But it's still funny
Vista was a kludge; and I'm sure M$ is likewise sick and tired of having to test against a bazillion possibilities, hence this concept. And I expect with modular-Windows you won't have to test near as much, because your app will only need to speak to the core OS, and won't care about any other crap. Fewer layers, less hassle. Think sideways rather than vertical.
WINE for Windows ... very strange!!
:)
But yeah, to make cranky old app speak to shiny new Windows, I can see using WINE instead of a VM... I'm not sure why, but maybe it'd work better for certain things. Anyone here know enough to comment intelligently?
I don't think M$ feels as threatened by apps being ported away as some folks think. All you have to do is look at Wine... yeah, it works, but is it up to what enterprise business (M$'s REAL customers) needs? Highly doubtful.
In fact, I think part of why M$ is looking at this scheme is because they know damn well that most businesses would rather buy a monolithic solution, and a modular OS makes it easier for M$ to customize that solution and thereby maximize profit.
To refute my own post [g] the main reason I'd put forth as to why M$ probably wouldn't go to an outright UNIX kernel, is because they have a known entity in Windows; why throw out two decades of expertise with their own product? And the Windows core is really a pretty good OS in itself; it goes wrong when too much crap gets shoveled in on top. If we no longer bury it in junk, it can better do its job.
Exactly. I don't see any downside for anyone. I see a lot *more* opportunity for money to flow in all directions, and not just toward M$.
I would hope that drivers would be developed at more than "their own pace", tho, because otherwise only a few popular components will have a chance. One of the reasons so many different types and brands of hardware are in the market today is because for most, drivers are available NOW, no waiting (at least for Windows. Linux, not so much.) Buy it, install it, works right now. Exceptions are not well-received in the market.
Small 3rd party developers could partner with OEMs or even with M$ and build up a repository of both free and pay-to-install software, rather akin to what Linspire originally did. This is an opportunity for apps that are presently shut out because there's just not a NEED for 'em when Windows already includes the kitchen sink. As to pricing, market forces will decide what's worth paying for. And if something gets pirated a lot... well, maybe it was overpriced.
As to updates, there's really no reason to complicate it by making into pay-per-level or pay-per-update -- why would I want an update that addresses software I didn't install in the first place? and unless an update is a wholesale replacement for a component, it's not going to install a complete app anyway, so it's of no use to someone trying to pirate a level of Windows they didn't pay for. Hell, updates could be unified and only install required packages per autodetect. -- This also prevents the disaster that's currently going on, where my non-WinXP system can't download updates for my WinXP customer whose machine is presently borkend.