Unfortunately, the case of WiFi is a bad one. Bandwidth in the electromagnetic spectrum is a finite resource; if no one owns the resource, and no price can therefore be placed on it, then it will not be used efficiently in the economic sense.
The basic principles underlying efficient markets are mathematical. The theorems require certain assumptions to be made, which consitute the axioms of the system, and these lead directly to logical conclusions.
Among these conclusions is that competitive markets lead to efficient allocation of resources, in the sense that no other allocation of resources would make all the market participants better off.
This is an amazing result, but it's not religious, it's mathematical.
From a policy point of view, this kind of legal restriction can prevent the conditions from being true, and therefore result in inefficient allocation of resources. Which might mean that *everyone* is worse off than in an efficient solution.
I don't claim to draw conclusions based on a Slashdot summary, but, if, for instance, internet providers were prohibited from charging for quality of service, whether latency or bandwidth, one could very easily end up in a situation where people who are perfectly willing to pay for improved service would be unable to, and instead, would have to deal with whatever the internet providers give out as their lowest common denominator, or whatever results from everybody downloading as much as they can from a system that has no incentive to invest in infrastructure.
It's natural to limit web storage to significantly less than e-mail storage.
- Only one user at a time is typically reading your e-mail. A web page can get hits from any number of people at once. - Most e-mail is plain text or simple HTML. Web pages routinely include large files like photos, and if enough space is available, videos. - Large attachments to e-mail get downloaded just once, typically. Popular photos or videos could be downloaded hundreds or thousands of times.
Unless, of course, you leave your e-mail inbox with video attachments open for any of the billion or so web users in the world to click on anytime they feel like it.
The failure mode of circular references defeating reference counting is more general than the toy example.
It can very easily occur as one builds up complex data structures to make more abstract representations possible, and then using the abstractions in straightforward ways. Imagine embedding a table in an object, used to efficiently find neighboring objects in a network. If that network has any cycle in it, boom. What do you do if the user is the one building the network? Tell them not to use your tool for networks with cycles?
This is analogous to the problems of manual memory management. In simple cases, i.e., simple enough to see the problem, you can always come up with a "simple" solution: i.e. "caller owns the resulting allocated object, and is responsible for deallocating." The problems develop when the call and ownership chains become complex, and the protocols built on top of protocols eventually fail to cover edge cases. Then, you've got a hell of a mess to crawl through to plug the leak.
Let's check your reading comprehension: your conclusion
Bill Gates,... valued his work year out at 13,333 bucks.
The actual quote
The value of the computer time we have used exceeds $40,000.
As a caveman, I remember back in those days, computer time was actually metered and paid for by the amount of CPU activity and I/O that the computer had to do on behalf of your process.
Unless Bill Gates thinks he is a computer, *his* labor has nothing to do with the quoted figure.
And is Apple anywhere near the goal of eliminating Windows marketshare? The Mac gossip sites still can't figure out which single-digit marketshare Apple will achieve this year.
I like Macs, and use them. I use PCs at work, but have *never* bought a PC with my own money.
Apple has the same problem as Linux; the PC marketshare they *currently* have is, in part, maintained only by the continued support MS gives for Word, Excel, and Powerpoint.
Note that Visio, Project, and Access are not on that list. Note that developing VBA applications on Macs is significantly less convenient (no Intellisense! no custom controls!) on Macs than on Windows.
People have just as much trouble switching to Macs as they do Linux, without (up till recently) the hope that something WINE-like could offer speedy emulation.
To switch to Macintosh, businesses would still have a substantial barrier to porting their apps AND they would have to give up the large-scale availability of PC hardware solutions for a single source.
It's obvious you have no ability to tell anything about me from my post; I've been using command lines for the last 25 years, without fear or fright. Actually, I find Windows from a user point of view to be a nightmarish, barely documented maze of DLL's with needlessly cryptic 8.3 names, impossible to reliably distinguish from viruses and malware in the Task Manager.
For the Nth time, this is not about *me*, it is about the commercial reality that there are apps in Windows that businesses depend on, whose functionality is *not* painlessly replacable by free software substitutes, or even proprietary substitutes that run on Linux. Microsoft knows this very well, and the Linux zealots need to recognize the same reality if they are to have any success at all at changing that reality, as opposed to pronouncing every year to be the "year of Linux on the desktop." My company is not going to spend hundreds of man years porting a Windows-locked system into Linux for no net gain, no matter how much UNIX/Linux skill I or my co-workers have, just as many people will not take the twenty painless twenty minutes it takes to switch to Firefox. There are plenty of people here who think that the choice of Windows was short-sighted, but those are the breaks.
The argument isn't that I haven't happened to switch, and that Linux should change to suit my whims; just that there are real barriers in the Windows ecosystem that favor Microsoft's continued dominance, and a real lack of non-political, non-warm-fuzzy, actual software domination that makes Linux really inevitable on the desktop. My workplace environment has several critical (unfortunate) dependences on MS that would take something like 100 man years or more to port to Linux.
Yes, Apache, emacs, and various ultra-subsidized things like Eclipse make free software a useful and valuable proposition, but it isn't going to displace Windows unless something much *more* earth-shaking than a Google Linux distro happens, or MS really screws up and slowly pisses their advantage away because Bill G.'s focus drifts and his successors blow it.
I think you seriously overrate the limits on your practical freedom. Your "lead and explore" line is particularly arrogant. Is downloading the same distro as everyone else really "leading"? Is watching the results of apt-get or a./configure; make; make install really "exploring"? Are you single-handedly determining the direction Open Office is taking? Redefining what software can do to solve real problems? Or, as most slashdot denizens, are you pretty much waiting for the next point release along with all the other consumers? Is Linux really "exploring", or is it simply repaving the same roads that Solaris and other proprietary Unixes laid out?
I have just as much liberty to use Linux as you do. The reality is, however, that Free Software does *not* meet all the needs of all the people, and the warm fuzzies do not compensate for the real shortcomings in various open source applications that prevent them from replacing proprietary, commercial alternatives, whether for Windows or otherwise.
I'm one of the folks who shed a tear every time I see a glimpse of marvelous software environments like Symbolics Genera, and the huge potential that there was for software; and are brought back to Earth by the realization that UNIX is not a thousand times better than it was in 1985 when RMS wrote the GNU manifesto, although our computers are nearly a thousand times faster and more capacious. We still use xterms and emacs in our X Window system, still peer at man pages to look at command-line options for ls.
Whatever barrier you're pointing at, I don't see it.
3. Running all the applications, some of which are critical to business processes, which only run on Windows, like, um, MS Office, for one?
Admittedly, the barrier could be overcome, but only if EVERY significant Windows app had a replacement that ran with sufficient compatibility to allow pain-free migration of documents and custom applications, or even just some compelling real-world advantage that more than compensates for the pain of change.
That is a HUGE barrier. The rest of my comment was an unsuccessful attempt to forestall all the "Open Office is good enough (...except for those pesky VBA macros that make up custom applications...and Project...)" or "GIMP is just fine for me (except if I wanted the UI to be comfortable for Photoshop jockeys.)" I.e., NOT pain-free compatibility.
Yet even people like you are only willing to claim 80% on the "as good as" scale, where the barrier is something like 98% in FILE OR UI COMPATIBILITY. Not "oh, you'll get used to it, eventually", or "It's different because it is much, much better" or "no one really needs that feature, anyway," or "many open source volunteers given years of additional time will improve things eventually" because all of that is still PAIN, in exchange for which there is no NET GAIN.
The point of the Firefox example, is even when something IS much better, and has a virtually painless migration route, most people STILL DON'T MIGRATE.
What is with the lack of reading comprehension today?
I didn't say OSS was "doomed." GCC is still here, Linux is still here, none of this is going away. Perl, etc., etc.
The parent was claiming that two, and presumably only two, things were standing in the way of the Linux steamroller, and as soon as Google makes those two things go away, bye-bye Windows.
I point out that there is a third, crucially important barrier to widespread replacement of Windows, and suddenly I am an ignoramus, who should just try Gentoo or Ubuntu, and download all my software, or just switch to Open Office, spouting flamebait or bullshit.
If anybody wants Free Software (or Open Source) to actually replace anything, they should understand the reality behind the thing being replaced, not some geeky caricature of how things *might be*.
Keep in mind, however, that "demand" in this case means "people willing to cough up US$30+ per install, usable for only one year, and a hassle to use for more than one return."
Not "willing to download free, open-source knockoff".
I think the only way Linux folks will be seeing tax software any time soon is through their web browser; which probably counts as a win for Linux, anyway. Tax software folks probably prefer the web browser versions because it saves the cost of physical boxes and stock-keeping.
And excel macros work just fine in openoffice. If you're doing somehting particularly complicated with it then you might have to do it a different way, but it is still somehting you can do. Anyone who can learn one complicated macro can learn another.
I see, so I can go through all the risk and effort of switching, and I have, what only a 5%, 10% chance of my critical apps breaking? (Yes, I know Windows/Office upgrades have a similar risk, but I don't like those either.)
And, I gain, what, exactly? The $50 Dell probably charged my company (but won't give back if I chose Linux) to install Windows on my work laptop? The warm feeling from being part of the Linux jihad? Respect from my fellow cavemen? The ability to post on Slashdot with moral superiority? What?
I actually used Gentoo once, when I had a 64 MB Pentium II computer with nothing better to do.
Then, hours later, I discovered that swap with a slow hard drive is not fun, and that the graphics card only supported some incredibly low resolution, and that it wasn't actually much fun. So I did something else. Someday soon, I might cough up $200 for one of those bargain bin PC's and start bizarro Linux kernel hacking, but I haven't.
Listen, I don't buy software at Best Buy, etc., and never claimed to. I just check every so often when I'm in the store, and EVERY TIME, I see that nothing there works on anything but Windows.
Will you people stop suggesting that *I* simply need to try Ubuntu/Gentoo, whatever, then I will be convinced that Linux is "ready" to kick Bill Gates in the ass? I can find my own way to productivity and entertainment, and know the difference. But *I* and *people like me* is not equivalent to *enough people to make a dent in Windows popularity*.
I happen to recognize this. Every dweeb who says "$NEW_OSS_RELEASE is how Linux will kill Windows, Yay!" apparently has not.
Meanwhile, the 90% of computer users who aren't like you will continue to use Windows.
The barrier to Linux "killing" Windows is NOT you, and it's not me either, but thanks for your suggestion. I don't feel like spending my free time and my cash to get a computer to install Ubuntu on (so I can do what, again?); I use Windows at work and OS X at home, and have plenty of other things to do with my free time.
If you've spent any time in an Apple store, you've encountered the fact that even home users care about having Office so they can bring work home with them, or whatever. Or the games they want to play, or the educational software they imagine will make their kids the next Bill Gates, or whatever.
Go to Best Buy, or Target, or Office Depot, or any other retailer of boxed software for consumers, and see how much will run on anything other than Windows. Hint: NONE.
That's a real barrier, and it's not just in the business environment.
Because of simple inertia? I use Netscape 7 on Windows, and Safari on Mac OS X, unless some web site forces me to use Explorer on either. I've worked this way for a few years, and hey, if it ain't broke, don't mess with it.
I download lots of software recreationally, but it tends to be other kinds of open source projects; improving my web browsing experience doesn't seem worth the effort.
"Reverse engineering" is, get this, "engineering" in "reverse."
Engineers start with a design, more or less formally specified, and end up delivering a product, which may or may not be delivered with complete documentation, or conform to a publically available standard.
Reverse engineers start with the product, which may or may not have complete documentation, or conform to a public standard, and come up with a design.
Get it?
Your metioning that reverse engineers "might not have access to the design documents", and your earlier discussion that involved "reading the source" shows that you have a fuzzy notion about what reverse engineering is.
"By your logic, sir, Google might as well just take the concept of the automobile one step farther and create flying cars."
Agreed. Now when can I expect my promised flying car!!!
And when will I get to see Bill Gates homeless and shaking his fist in the air, shouting "Damn you Google, and your flying cars for destroying Microsoft!"
3. Running all the applications, some of which are critical to business processes, which only run on Windows, like, um, MS Office, for one?
If you believe Open Office, or some other knockoff, or half-assed open source replacement for any other Windows-only application I might think of, is enough to cause people to totally change their Windows-buying habits, you are out of touch with reality.
Firefox, from what I hear, is an eminently usable Web browser, and Internet Explorer is hardly critical to most Web applications. Yet, only a relatively small fraction of users make the effort to switch.
Take a case where something like Microsoft Excel VBA macros or Microsoft Project is playing a daily role in a company's actual function of making money, and how eager will people be to switch to something else, just for the warm feeling they can get from using the Linux distribution du jour?
Slashdot Mirror
Unfortunately, the case of WiFi is a bad one. Bandwidth in the electromagnetic spectrum is a finite resource; if no one owns the resource, and no price can therefore be placed on it, then it will not be used efficiently in the economic sense.
You mistake mathematical rigor for religion.
The basic principles underlying efficient markets are mathematical. The theorems require certain assumptions to be made, which consitute the axioms of the system, and these lead directly to logical conclusions.
Among these conclusions is that competitive markets lead to efficient allocation of resources, in the sense that no other allocation of resources would make all the market participants better off.
This is an amazing result, but it's not religious, it's mathematical.
From a policy point of view, this kind of legal restriction can prevent the conditions from being true, and therefore result in inefficient allocation of resources. Which might mean that *everyone* is worse off than in an efficient solution.
I don't claim to draw conclusions based on a Slashdot summary, but, if, for instance, internet providers were prohibited from charging for quality of service, whether latency or bandwidth, one could very easily end up in a situation where people who are perfectly willing to pay for improved service would be unable to, and instead, would have to deal with whatever the internet providers give out as their lowest common denominator, or whatever results from everybody downloading as much as they can from a system that has no incentive to invest in infrastructure.
The electron has a factual basis.
I'm not sure about that- has anybody ever seen one?
Hmm. Are you looking at a CRT screen when you read this? Or an LCD?
Because neither of those things work too well if you don't obey the rules of "electron theory" when you make them.
There are several ways different texts can agree with one another.
The most important way is that one text is copied or translated from another.
Ever heard of a Xerox machine? It means that documents found at different times and places can be *exact* copies of one another! Oh my!
Is that evidence that xerography was divinely inspired?
But only for the oxygen-free download over a gold-plated DSL connection with the bits degaussed to remove any residual polarization.
It's natural to limit web storage to significantly less than e-mail storage.
- Only one user at a time is typically reading your e-mail. A web page can get hits from any number of people at once.
- Most e-mail is plain text or simple HTML. Web pages routinely include large files like photos, and if enough space is available, videos.
- Large attachments to e-mail get downloaded just once, typically. Popular photos or videos could be downloaded hundreds or thousands of times.
Unless, of course, you leave your e-mail inbox with video attachments open for any of the billion or so web users in the world to click on anytime they feel like it.
Yes, and you can tell that very few readers got far enough to get the
The application is meant to primarily run on Linux, but should be portable to Windows without much difficulty.
The failure mode of circular references defeating reference counting is more general than the toy example.
It can very easily occur as one builds up complex data structures to make more abstract representations possible, and then using the abstractions in straightforward ways. Imagine embedding a table in an object, used to efficiently find neighboring objects in a network. If that network has any cycle in it, boom. What do you do if the user is the one building the network? Tell them not to use your tool for networks with cycles?
This is analogous to the problems of manual memory management. In simple cases, i.e., simple enough to see the problem, you can always come up with a "simple" solution: i.e. "caller owns the resulting allocated object, and is responsible for deallocating." The problems develop when the call and ownership chains become complex, and the protocols built on top of protocols eventually fail to cover edge cases. Then, you've got a hell of a mess to crawl through to plug the leak.
Let's check your reading comprehension: your conclusion
... valued his work year out at 13,333 bucks.
Bill Gates,
The actual quote
The value of the computer time we have used exceeds $40,000.
As a caveman, I remember back in those days, computer time was actually metered and paid for by the amount of CPU activity and I/O that the computer had to do on behalf of your process.
Unless Bill Gates thinks he is a computer, *his* labor has nothing to do with the quoted figure.
And is Apple anywhere near the goal of eliminating Windows marketshare? The Mac gossip sites still can't figure out which single-digit marketshare Apple will achieve this year.
I like Macs, and use them. I use PCs at work, but have *never* bought a PC with my own money.
Apple has the same problem as Linux; the PC marketshare they *currently* have is, in part, maintained only by the continued support MS gives for Word, Excel, and Powerpoint.
Note that Visio, Project, and Access are not on that list. Note that developing VBA applications on Macs is significantly less convenient (no Intellisense! no custom controls!) on Macs than on Windows.
People have just as much trouble switching to Macs as they do Linux, without (up till recently) the hope that something WINE-like could offer speedy emulation.
To switch to Macintosh, businesses would still have a substantial barrier to porting their apps AND they would have to give up the large-scale availability of PC hardware solutions for a single source.
It's obvious you have no ability to tell anything about me from my post; I've been using command lines for the last 25 years, without fear or fright. Actually, I find Windows from a user point of view to be a nightmarish, barely documented maze of DLL's with needlessly cryptic 8.3 names, impossible to reliably distinguish from viruses and malware in the Task Manager.
For the Nth time, this is not about *me*, it is about the commercial reality that there are apps in Windows that businesses depend on, whose functionality is *not* painlessly replacable by free software substitutes, or even proprietary substitutes that run on Linux. Microsoft knows this very well, and the Linux zealots need to recognize the same reality if they are to have any success at all at changing that reality, as opposed to pronouncing every year to be the "year of Linux on the desktop." My company is not going to spend hundreds of man years porting a Windows-locked system into Linux for no net gain, no matter how much UNIX/Linux skill I or my co-workers have, just as many people will not take the twenty painless twenty minutes it takes to switch to Firefox. There are plenty of people here who think that the choice of Windows was short-sighted, but those are the breaks.
The argument isn't that I haven't happened to switch, and that Linux should change to suit my whims; just that there are real barriers in the Windows ecosystem that favor Microsoft's continued dominance, and a real lack of non-political, non-warm-fuzzy, actual software domination that makes Linux really inevitable on the desktop. My workplace environment has several critical (unfortunate) dependences on MS that would take something like 100 man years or more to port to Linux.
Yes, Apache, emacs, and various ultra-subsidized things like Eclipse make free software a useful and valuable proposition, but it isn't going to displace Windows unless something much *more* earth-shaking than a Google Linux distro happens, or MS really screws up and slowly pisses their advantage away because Bill G.'s focus drifts and his successors blow it.
I think you seriously overrate the limits on your practical freedom. Your "lead and explore" line is particularly arrogant. Is downloading the same distro as everyone else really "leading"? Is watching the results of apt-get or a ./configure; make; make install really "exploring"? Are you single-handedly determining the direction Open Office is taking? Redefining what software can do to solve real problems? Or, as most slashdot denizens, are you pretty much waiting for the next point release along with all the other consumers? Is Linux really "exploring", or is it simply repaving the same roads that Solaris and other proprietary Unixes laid out?
I have just as much liberty to use Linux as you do. The reality is, however, that Free Software does *not* meet all the needs of all the people, and the warm fuzzies do not compensate for the real shortcomings in various open source applications that prevent them from replacing proprietary, commercial alternatives, whether for Windows or otherwise.
I'm one of the folks who shed a tear every time I see a glimpse of marvelous software environments like Symbolics Genera, and the huge potential that there was for software; and are brought back to Earth by the realization that UNIX is not a thousand times better than it was in 1985 when RMS wrote the GNU manifesto, although our computers are nearly a thousand times faster and more capacious. We still use xterms and emacs in our X Window system, still peer at man pages to look at command-line options for ls.
Whatever barrier you're pointing at, I don't see it.
3. Running all the applications, some of which are critical to business processes, which only run on Windows, like, um, MS Office, for one?
Admittedly, the barrier could be overcome, but only if EVERY significant Windows app had a replacement that ran with sufficient compatibility to allow pain-free migration of documents and custom applications, or even just some compelling real-world advantage that more than compensates for the pain of change.
That is a HUGE barrier. The rest of my comment was an unsuccessful attempt to forestall all the "Open Office is good enough (...except for those pesky VBA macros that make up custom applications...and Project...)" or "GIMP is just fine for me (except if I wanted the UI to be comfortable for Photoshop jockeys.)" I.e., NOT pain-free compatibility.
Yet even people like you are only willing to claim 80% on the "as good as" scale, where the barrier is something like 98% in FILE OR UI COMPATIBILITY. Not "oh, you'll get used to it, eventually", or "It's different because it is much, much better" or "no one really needs that feature, anyway," or "many open source volunteers given years of additional time will improve things eventually" because all of that is still PAIN, in exchange for which there is no NET GAIN.
The point of the Firefox example, is even when something IS much better, and has a virtually painless migration route, most people STILL DON'T MIGRATE.
What is with the lack of reading comprehension today?
I didn't say OSS was "doomed." GCC is still here, Linux is still here, none of this is going away. Perl, etc., etc.
The parent was claiming that two, and presumably only two, things were standing in the way of the Linux steamroller, and as soon as Google makes those two things go away, bye-bye Windows.
I point out that there is a third, crucially important barrier to widespread replacement of Windows, and suddenly I am an ignoramus, who should just try Gentoo or Ubuntu, and download all my software, or just switch to Open Office, spouting flamebait or bullshit.
If anybody wants Free Software (or Open Source) to actually replace anything, they should understand the reality behind the thing being replaced, not some geeky caricature of how things *might be*.
Is realism too much to ask for?
http://googlefight.com/index.php?lang=en_GB&word1= goobuntu&word2=googlix
Googlix wins! (for now...)
Keep in mind, however, that "demand" in this case means "people willing to cough up US$30+ per install, usable for only one year, and a hassle to use for more than one return."
Not "willing to download free, open-source knockoff".
I think the only way Linux folks will be seeing tax software any time soon is through their web browser; which probably counts as a win for Linux, anyway. Tax software folks probably prefer the web browser versions because it saves the cost of physical boxes and stock-keeping.
And excel macros work just fine in openoffice. If you're doing somehting particularly complicated with it then you might have to do it a different way, but it is still somehting you can do. Anyone who can learn one complicated macro can learn another.
I see, so I can go through all the risk and effort of switching, and I have, what only a 5%, 10% chance of my critical apps breaking? (Yes, I know Windows/Office upgrades have a similar risk, but I don't like those either.)
And, I gain, what, exactly? The $50 Dell probably charged my company (but won't give back if I chose Linux) to install Windows on my work laptop? The warm feeling from being part of the Linux jihad? Respect from my fellow cavemen? The ability to post on Slashdot with moral superiority? What?
I actually used Gentoo once, when I had a 64 MB Pentium II computer with nothing better to do.
Then, hours later, I discovered that swap with a slow hard drive is not fun, and that the graphics card only supported some incredibly low resolution, and that it wasn't actually much fun. So I did something else. Someday soon, I might cough up $200 for one of those bargain bin PC's and start bizarro Linux kernel hacking, but I haven't.
Listen, I don't buy software at Best Buy, etc., and never claimed to. I just check every so often when I'm in the store, and EVERY TIME, I see that nothing there works on anything but Windows.
Will you people stop suggesting that *I* simply need to try Ubuntu/Gentoo, whatever, then I will be convinced that Linux is "ready" to kick Bill Gates in the ass? I can find my own way to productivity and entertainment, and know the difference. But *I* and *people like me* is not equivalent to *enough people to make a dent in Windows popularity*.
I happen to recognize this. Every dweeb who says "$NEW_OSS_RELEASE is how Linux will kill Windows, Yay!" apparently has not.
Good for you.
Meanwhile, the 90% of computer users who aren't like you will continue to use Windows.
The barrier to Linux "killing" Windows is NOT you, and it's not me either, but thanks for your suggestion. I don't feel like spending my free time and my cash to get a computer to install Ubuntu on (so I can do what, again?); I use Windows at work and OS X at home, and have plenty of other things to do with my free time.
If you've spent any time in an Apple store, you've encountered the fact that even home users care about having Office so they can bring work home with them, or whatever. Or the games they want to play, or the educational software they imagine will make their kids the next Bill Gates, or whatever.
Go to Best Buy, or Target, or Office Depot, or any other retailer of boxed software for consumers, and see how much will run on anything other than Windows. Hint: NONE.
That's a real barrier, and it's not just in the business environment.
Why wouldn't you try [Firefox] for yourself?
Because of simple inertia? I use Netscape 7 on Windows, and Safari on Mac OS X, unless some web site forces me to use Explorer on either. I've worked this way for a few years, and hey, if it ain't broke, don't mess with it.
I download lots of software recreationally, but it tends to be other kinds of open source projects; improving my web browsing experience doesn't seem worth the effort.
"Reverse engineering" is, get this, "engineering" in "reverse."
Engineers start with a design, more or less formally specified, and end up delivering a product, which may or may not be delivered with complete documentation, or conform to a publically available standard.
Reverse engineers start with the product, which may or may not have complete documentation, or conform to a public standard, and come up with a design.
Get it?
Your metioning that reverse engineers "might not have access to the design documents", and your earlier discussion that involved "reading the source" shows that you have a fuzzy notion about what reverse engineering is.
"By your logic, sir, Google might as well just take the concept of the automobile one step farther and create flying cars."
Agreed. Now when can I expect my promised flying car!!!
And when will I get to see Bill Gates homeless and shaking his fist in the air, shouting "Damn you Google, and your flying cars for destroying Microsoft!"
Um, I think you forgot
3. Running all the applications, some of which are critical to business processes, which only run on Windows, like, um, MS Office, for one?
If you believe Open Office, or some other knockoff, or half-assed open source replacement for any other Windows-only application I might think of, is enough to cause people to totally change their Windows-buying habits, you are out of touch with reality.
Firefox, from what I hear, is an eminently usable Web browser, and Internet Explorer is hardly critical to most Web applications. Yet, only a relatively small fraction of users make the effort to switch.
Take a case where something like Microsoft Excel VBA macros or Microsoft Project is playing a daily role in a company's actual function of making money, and how eager will people be to switch to something else, just for the warm feeling they can get from using the Linux distribution du jour?