"I'm surprised this is being labelled censorship by some people -- it's complying with the law and ensuring that a very important trail isn't jeopordised."
Well, that I suppose is obviously the question; should the right to a fair trial trump the right to free expression?
IMO the answer is clearly "yes" (not least because the fundamental value of free expression is the ability to criticize government actors and government policy, not write anything about anyone). The harm to free expression is in any case minimal, because there is a temporal constraint on such restrictions; when the trial ends (or is abandoned) the reporting restrictions are for the most part lifted (restrictions on naming minors and so on notwithstanding).
... all of these benefit extensively from caps lock. As all have a de facto standard of writing certain kinds of thing (macros, constants) in all-caps, I find I use my caps lock rather frequently. Much quicker to hit caps lock and type normally than have to bounce my shifting from left to right.
That it uses change notifications (which are for some unfathomable reason performed in kernel-mode) is neither here nor there. It does nothing that separate apps cannot do; external apps can listen for changes just as well as kernel modules.
Windows' built-in indexing doesn't work in this way; it instead uses the NTFS usn journal. This allows it to track changes that Spotlight would miss (for example, if you dual boot, the usn journal still gets updated so the indexer still knows what needs indexing), and allows you to for example suspend indexing temporarily without having to start from scratch (because, again, the journal gets updated regardless of whether the indexer is indexing or not). Spotlight, however, has to depend on a background periodical scanning/indexing process to pick up anything that falls through the cracks (due to e.g. booting into an older OS X, connecting a removable disk to a non-10.4.x computer, disabling Spotlight)
"ROFL. Oh, yeah? Where is it? I love how Windows fanboys always refer to Microsoft's non-existent vaporware as if it's a shipping product. The most Microsoft has done is release that crappy MSN indexing toolbar." What the fuck are you talking about?
The Indexing Service has been around for ages. It's not fucking vapourware. It's real.
"You have to face facts here. Apple shipped with metadata indexing FIRST. Microsoft didn't. It doesn't matter if some "Option Pack" for NT4 had some lame indexing service that resembled Apple's Sherlock." It doesn't matter that something for NT 4 did what Apple are touting as remarkable innovation in OS X nearly a decade later?
"WinFS was dumped because it didn't work. It was never a good idea in the first place. It was marketing fluff to get people like you excited about vaporware so you didn't jump ship to superior alternatives like OS X, which actually ships its technologies instead of promising them for half a decade." The Spotlight technology is something that's been in Windows for years. Why would it cause anyone to jump ship?
"Have fun waiting another 12 months for Microsoft to squeeze out Vista onto a plate. Mac users already have Vista's features--since April of 2005, in fact. Microsoft has been following Apple's technical lead for over two decades. You can't change history, bub." What, Apple's technical lead like pre-emptive multitasking and protected memory?
Er... Spotlight is a direct rip-off of technology that MS have shipped since the NT 4 Option Pack, and which has been part of the default OS install since Windows 2000.
A Spotlight model where you write an importer that decides what metadata matters?
Do you mean an Indexing Service model where you write an importer that decides what metadata is available etc.?
An importer implementing the interface IFilter, the same interface that Vista will use?
IFilter which has been in use for more than half a decade, and several years prior to Spotlight?
MS have been there, done that. They wanted to move *beyond* the "Spotlight model" (that is to say, the model that they shipped long before Spotlight was ever even conceived) because they felt at the time it would afford greater flexibility and capabilities.
At this point in time it now very much appears that this WinFS-style indexing is for one reason or another unworkable (probably the biggest issue is that it's just not all that useful), so with Vista they're sticking with the old model. But they're not fucking copying Spotlight. Spotlight *is* the copy.
They moved GDI into the kernel. Other bits of Win32 remain user-mode, as does POSIX. Moving GDI in this way improved performance with no effective loss of stability because GDI was always deemed a "critical" subsystem; if it crashed in NT 3.51 the OS was still brought down. CSRSS still maintains this property.
In Vista, large tracts of the new graphics subsystem are moving back out into user-mode, as are various other odds and sods like sound drivers. Vista will also have a greater ability to restart crashed user-mode components, perhaps even the video subsytem.
They don't appear to make any arguments that hold up to any kind of scrutiny; all the points in favour of hardware patents are applicable to software patents.
The article already briefly outlines why they're the same; I don't feel any particular need to restate it, since we've all read it prior to posting comments. Right?
I'm not really sure what point you're trying to make. Just because groups pretend that there are meaningful differences does not mean that they are correct in their appraisal.
Oh, wait, you didn't. You just made an assertion "hardware patents good, software patents bad" without doing even the slightest thing to justify it.
All the arguments for hardware patents can be made for software patents. All the arguments against hardware patents can be made against software patents.
To object to one but not the other is inconsistent.
You have I presume noticed that the browser gets "www.google.com" stuck in its address bar by virtue of visiting www.google.com?
This isn't due to scripts putting arbitrary text in the address bar. It's caused by a race condition. If in quick succession you visit a shockwave site and then some other URL, the browser does the following:
1) start loading the evil shockwave site (showing its evil URL) 2) start loading the friendly site (showing its friendly URL) 3) finish loading the friendly site 4) display the friendly site 5) finish loading the evil shockwave site 6) display the evil shockwave site
As such, constraining the drawable area doesn't actually help--the exploit isn't writing to the address or status bar or anything like that. It's simply exploiting the behaviour of the shockwave plugin. If the shockwave plugin begins to load a site it's damn well going to finish loading it, even if the user (or a script) has navigated away from that site.
To be honest, it's not clear to me if this is a problem with IE or the plugin in question; does IE exhibit this behaviour for any plugin, or is it just shockwave? I don't know if IE tells its plugins "cancel what you're doing, the user has navigated away", for example, or whether it would have the ability to discard the output of the plugin.
"And would the divorce of IE7 from Vista's Windows Explorer help?"
Of course it fucking won't.
Vista does have things that will help (such as running IE as an unprivileged user) but separating IE from Explorer will not do a damn thing, because it's never been a cause of problems.
If I can exploit IE to run arbitrary code, or read or write files from places it shouldn't, the file manager/shell of the OS doesn't matter a jot. Consider, there is not one single IE exploit that would be mitigated by the use of a shell other than Explorer.
The whining about the "security" implications of the integration has never had any basis in fact.
The problem is, IT doesn't live in the world of business; the geeks live in a fantasy land. IT is responsible for ensuring that, for example, there is enough storage space on the fileserver. This is IT's responsibility. IT are the ones with visibility of the number and size of drives, the amount of free space, the partitioning, the number of free bays in the drive cage, etc.. Not the users. The users need the space, but IT are the ones who must provide it. If space is running low then it's up to IT to resolve this problem; they have the user requirement ("provide storage for our files") and they need to fulfil it.
The solution here is not for someone in IT to mutter to someone higher up "oh, we need a few more hard disks". Nor is it for IT to just decide implement some absurd restriction such as quotas. It's for IT to create a costed request for a storage upgrade. Depending on what infrastructure exists, this may be small and simple (just stick a few more disks into the rack) or it may be big and complex (to satisfy future growth, availability, backup requirements we need to phase out "file servers" with their own local storage and invest in NAS or SAN-based storage), or it may be somewhere in between. Part of this request will of course include the rationale ("department X produce lots of data"), and will possibly need some detail to justify the decisions. These then need to be given to management.
In other words, a proper business case needs to be put together. Management may then say "the benefit doesn't justify the cost, what else can we do?". It's at this point that alternatives may need to be devised; maybe archive off older files to tape with an HSM system; maybe implement some quotas; maybe backup workstations to permit local storage. Because these alternatives result in an inability to fulfil the original demand, they need to be worked through with the department in question to evaluate their impact. Again, it needs to be costed and business-oriented. Vague demands for "more disks" and quota diktats are not acceptable.
But IT don't get this, because they're antisocial geeks who for some unfathomable reason believe that they know better than everyone else.
It doesn't matter how much or how little the non-IT people know. There are basic things that should just work. I should always be able to access my mail. Server software upgrades should not happen during office hours. Backups should work. You don't need technical expertise to be able to demand these things. They're the bare minimum expectation. I don't want to hear about how hard you think it is, because this stuff should just work. It's not rocket science. And fobbing people off with whining is not acceptable. Yet it's endemic to the IT world.
ZombieMime asks: "The non-IT employees at my company (approximately some 5,000,000,000 people) are showing signs of incompetence, and have been ignoring knowledgeable technology input for about a year. Additionally, they haven't been able to accept needed changes to senior management.
Right, because the IT guy knows how to run the company. I suppose it could happen, but it's not likely. If he could, he'd be making more money running the damn company.
Unacceptable computer usage,
No such thing. If someone's computer usage (a) breaks no laws (b) does not negatively impact their productivity then it's not "unacceptable". And frankly, the idea of IT dictating how someone should and shouldn't use their computer to be productive is ludicrous.
maxed bandwidth usage,
Then get more.
and no common sense
O RLY? That's a problem I see far more from the IT side of the fence than the user side.
All too often IT believe their purpose is to be as great an impediment as possible in the smooth running of the company and the productivity of its employees. IT services should be like janitorial services. They should not get in people's way. They should not tell people how to do their jobs. They should simply be enablers. They do the dirty work to let other people earn the company money.
"are remotely exploitable problems with windows, especially if the boxes are unpatched, "
No. Not "especially" if they are unpatched. Only if they are unpatched. If they were patched (and in both cases the patches were available long before the exploits were) then they were not vulnerable.
So that means an admin has to do his job. Guess what? That requirement ain't OS-dependent.
"These systems cannot run anti-virus software at the same time they record "
Rubbish.
AV software should impose zero overhead on the recording process, because AV software should not be scanning the (presumably large) data files produced by the recording. All online scanners I'm aware of let you choose between scanning only executables, all files, or user-specified extensions. Use this feature ffs.
And viruses don't just get onto a computer. Viruses get onto computers through improper user behaviour (namely, users running viruses). An online scanner can be effective against this (user education even more so, as well as telling people to stop running arbitrary programs on the recording computers...). Worms can get in through open ports, but the answer to that (on/any/ OS) is to patch (and you can probably help out by e.g. ensuring that you use processors with non-exec page protection). Switching to some other platform will not remove the need to patch.
Sorry, but this is bollocks. This kind of moral relativism is one of the big things wrong with the world. Oh, the Chinese government isn't/bad/ because it suppresses speech, because it imprisons arbitrarily, because it tortures, because it executes willy nilly. It's just different!
These things are simply/bad/. They're not just different. They're/bad/. They're worse than what western governments do. The Chinese government is an abusive and tyrannical regime.
Christ, next thing you know you'll be telling us that the fucking holocaust was perfectly "valid", it's simply that certain high-ranking Nazis had different "values" from the rest of us, and we should have respected those values.
It's hilarious that you criticize others for not looking at anything other than the "politically correct" view. Because saying "our culture is better than yours" is not "politically correct". On the contrary, it's YOUR position--the one where you defend monstrous regimes as simply having different but valid values that we should respect--that is the politically correct position. It's complete bullshit. People like you will justify any act, no matter how horrendous.
If American companies should ignore local law when operating in China, does this mean that Dutch companies should now be encouraged to sell pot in the US?
"You're right, you should make backups." Right. And so why do you give a shit about the OS and apps? The OS is on a silver disk with the PC. Backing up means that your important shit is on a gold disk. The copy of the OS on the hard drive at this point becomes practically worthless. You can do whatever you want to it and it really doesn't matter. And if you don't have a backup you're fucked anyway, regardless of the state of the OS.
It's the data that matters, and most OSes are pisspoor at protecting data, because the person who created the data can also destroy it. Perhaps a capability-based system would offer an improvement in this area, but there are no widely used capability-based systems around, so at this point in time it's academic; it's not even obvious that the system would be usable by untrained and unskilled users (that is to say, it's completely unacceptable for the computer to tell the user "Access Denied" when they try to do something they want to do; DAC is bad enough, and I'm suspicious that MAC and capabilities systems will be even worse).
Protecting the system is more important in multi-user systems, but the typical desktop PC isn't multi-user, so it's not an important consideration. I'm not the least bit convinced that Unix is any better at protecting the system in such systems. But for the purposes of this argument it doesn't actually matter.
Unfortunately, that isn't what happened. The patent was upheld, because the software claimed to be prior art was deemed to be different; the differences between the prior art and the patent were negligble or non-existant, and the patent was certainly obvious, but the courts don't care.
You must remember that the courts have fuck all notion of what it is to be "obvious" (or not). To a software developer it would be argued that, for example, the Eolas patent was an obvious and natural refinement of existing technologies (using in-process dynamically linked libraries instead of external processes) and thus should have been invalidated by the copious prior art (which covered approximately 99.9% of the same material). If nothing else, you'd think that independent discovery (which is not held to be a protection against charges of patent violation) would demonstrate obviousness.
But not in the eyes of the courts or the patent office.
It would frankly not surprise me if the same were true in this case.
Slashdot Mirror
Indeed.
Or the Daemonette (Ceren Ercen).
"I'm surprised this is being labelled censorship by some people -- it's complying with the law and ensuring that a very important trail isn't jeopordised."
Well, that I suppose is obviously the question; should the right to a fair trial trump the right to free expression?
IMO the answer is clearly "yes" (not least because the fundamental value of free expression is the ability to criticize government actors and government policy, not write anything about anyone). The harm to free expression is in any case minimal, because there is a temporal constraint on such restrictions; when the trial ends (or is abandoned) the reporting restrictions are for the most part lifted (restrictions on naming minors and so on notwithstanding).
... all of these benefit extensively from caps lock. As all have a de facto standard of writing certain kinds of thing (macros, constants) in all-caps, I find I use my caps lock rather frequently. Much quicker to hit caps lock and type normally than have to bounce my shifting from left to right.
That it uses change notifications (which are for some unfathomable reason performed in kernel-mode) is neither here nor there. It does nothing that separate apps cannot do; external apps can listen for changes just as well as kernel modules.
Windows' built-in indexing doesn't work in this way; it instead uses the NTFS usn journal. This allows it to track changes that Spotlight would miss (for example, if you dual boot, the usn journal still gets updated so the indexer still knows what needs indexing), and allows you to for example suspend indexing temporarily without having to start from scratch (because, again, the journal gets updated regardless of whether the indexer is indexing or not). Spotlight, however, has to depend on a background periodical scanning/indexing process to pick up anything that falls through the cracks (due to e.g. booting into an older OS X, connecting a removable disk to a non-10.4.x computer, disabling Spotlight)
"ROFL. Oh, yeah? Where is it? I love how Windows fanboys always refer to Microsoft's non-existent vaporware as if it's a shipping product. The most Microsoft has done is release that crappy MSN indexing toolbar."
What the fuck are you talking about?
The Indexing Service has been around for ages. It's not fucking vapourware. It's real.
"You have to face facts here. Apple shipped with metadata indexing FIRST. Microsoft didn't. It doesn't matter if some "Option Pack" for NT4 had some lame indexing service that resembled Apple's Sherlock."
It doesn't matter that something for NT 4 did what Apple are touting as remarkable innovation in OS X nearly a decade later?
"WinFS was dumped because it didn't work. It was never a good idea in the first place. It was marketing fluff to get people like you excited about vaporware so you didn't jump ship to superior alternatives like OS X, which actually ships its technologies instead of promising them for half a decade."
The Spotlight technology is something that's been in Windows for years. Why would it cause anyone to jump ship?
"Have fun waiting another 12 months for Microsoft to squeeze out Vista onto a plate. Mac users already have Vista's features--since April of 2005, in fact. Microsoft has been following Apple's technical lead for over two decades. You can't change history, bub."
What, Apple's technical lead like pre-emptive multitasking and protected memory?
Oh, wait.
Er... Spotlight is a direct rip-off of technology that MS have shipped since the NT 4 Option Pack, and which has been part of the default OS install since Windows 2000.
A Spotlight model where you write an importer that decides what metadata matters?
Do you mean an Indexing Service model where you write an importer that decides what metadata is available etc.?
An importer implementing the interface IFilter, the same interface that Vista will use?
IFilter which has been in use for more than half a decade, and several years prior to Spotlight?
MS have been there, done that. They wanted to move *beyond* the "Spotlight model" (that is to say, the model that they shipped long before Spotlight was ever even conceived) because they felt at the time it would afford greater flexibility and capabilities.
At this point in time it now very much appears that this WinFS-style indexing is for one reason or another unworkable (probably the biggest issue is that it's just not all that useful), so with Vista they're sticking with the old model. But they're not fucking copying Spotlight. Spotlight *is* the copy.
They moved GDI into the kernel. Other bits of Win32 remain user-mode, as does POSIX. Moving GDI in this way improved performance with no effective loss of stability because GDI was always deemed a "critical" subsystem; if it crashed in NT 3.51 the OS was still brought down. CSRSS still maintains this property.
In Vista, large tracts of the new graphics subsystem are moving back out into user-mode, as are various other odds and sods like sound drivers. Vista will also have a greater ability to restart crashed user-mode components, perhaps even the video subsytem.
They don't appear to make any arguments that hold up to any kind of scrutiny; all the points in favour of hardware patents are applicable to software patents.
The article already briefly outlines why they're the same; I don't feel any particular need to restate it, since we've all read it prior to posting comments. Right?
I'm not really sure what point you're trying to make. Just because groups pretend that there are meaningful differences does not mean that they are correct in their appraisal.
I like the well-reasoned argument you give.
Oh, wait, you didn't. You just made an assertion "hardware patents good, software patents bad" without doing even the slightest thing to justify it.
All the arguments for hardware patents can be made for software patents.
All the arguments against hardware patents can be made against software patents.
To object to one but not the other is inconsistent.
You have I presume noticed that the browser gets "www.google.com" stuck in its address bar by virtue of visiting www.google.com?
This isn't due to scripts putting arbitrary text in the address bar. It's caused by a race condition. If in quick succession you visit a shockwave site and then some other URL, the browser does the following:
1) start loading the evil shockwave site (showing its evil URL)
2) start loading the friendly site (showing its friendly URL)
3) finish loading the friendly site
4) display the friendly site
5) finish loading the evil shockwave site
6) display the evil shockwave site
As such, constraining the drawable area doesn't actually help--the exploit isn't writing to the address or status bar or anything like that. It's simply exploiting the behaviour of the shockwave plugin. If the shockwave plugin begins to load a site it's damn well going to finish loading it, even if the user (or a script) has navigated away from that site.
To be honest, it's not clear to me if this is a problem with IE or the plugin in question; does IE exhibit this behaviour for any plugin, or is it just shockwave? I don't know if IE tells its plugins "cancel what you're doing, the user has navigated away", for example, or whether it would have the ability to discard the output of the plugin.
"And would the divorce of IE7 from Vista's Windows Explorer help?"
Of course it fucking won't.
Vista does have things that will help (such as running IE as an unprivileged user) but separating IE from Explorer will not do a damn thing, because it's never been a cause of problems.
If I can exploit IE to run arbitrary code, or read or write files from places it shouldn't, the file manager/shell of the OS doesn't matter a jot. Consider, there is not one single IE exploit that would be mitigated by the use of a shell other than Explorer.
The whining about the "security" implications of the integration has never had any basis in fact.
The problem is, IT doesn't live in the world of business; the geeks live in a fantasy land. IT is responsible for ensuring that, for example, there is enough storage space on the fileserver. This is IT's responsibility. IT are the ones with visibility of the number and size of drives, the amount of free space, the partitioning, the number of free bays in the drive cage, etc.. Not the users. The users need the space, but IT are the ones who must provide it. If space is running low then it's up to IT to resolve this problem; they have the user requirement ("provide storage for our files") and they need to fulfil it.
The solution here is not for someone in IT to mutter to someone higher up "oh, we need a few more hard disks". Nor is it for IT to just decide implement some absurd restriction such as quotas. It's for IT to create a costed request for a storage upgrade. Depending on what infrastructure exists, this may be small and simple (just stick a few more disks into the rack) or it may be big and complex (to satisfy future growth, availability, backup requirements we need to phase out "file servers" with their own local storage and invest in NAS or SAN-based storage), or it may be somewhere in between. Part of this request will of course include the rationale ("department X produce lots of data"), and will possibly need some detail to justify the decisions. These then need to be given to management.
In other words, a proper business case needs to be put together. Management may then say "the benefit doesn't justify the cost, what else can we do?". It's at this point that alternatives may need to be devised; maybe archive off older files to tape with an HSM system; maybe implement some quotas; maybe backup workstations to permit local storage. Because these alternatives result in an inability to fulfil the original demand, they need to be worked through with the department in question to evaluate their impact. Again, it needs to be costed and business-oriented. Vague demands for "more disks" and quota diktats are not acceptable.
But IT don't get this, because they're antisocial geeks who for some unfathomable reason believe that they know better than everyone else.
It doesn't matter how much or how little the non-IT people know. There are basic things that should just work. I should always be able to access my mail. Server software upgrades should not happen during office hours. Backups should work. You don't need technical expertise to be able to demand these things. They're the bare minimum expectation. I don't want to hear about how hard you think it is, because this stuff should just work. It's not rocket science. And fobbing people off with whining is not acceptable. Yet it's endemic to the IT world.
ZombieMime asks: "The non-IT employees at my company (approximately some 5,000,000,000 people) are showing signs of incompetence, and have been ignoring knowledgeable technology input for about a year. Additionally, they haven't been able to accept needed changes to senior management.
Right, because the IT guy knows how to run the company. I suppose it could happen, but it's not likely. If he could, he'd be making more money running the damn company.
Unacceptable computer usage,
No such thing. If someone's computer usage (a) breaks no laws (b) does not negatively impact their productivity then it's not "unacceptable". And frankly, the idea of IT dictating how someone should and shouldn't use their computer to be productive is ludicrous.
maxed bandwidth usage,
Then get more.
and no common sense
O RLY? That's a problem I see far more from the IT side of the fence than the user side.
All too often IT believe their purpose is to be as great an impediment as possible in the smooth running of the company and the productivity of its employees. IT services should be like janitorial services. They should not get in people's way. They should not tell people how to do their jobs. They should simply be enablers. They do the dirty work to let other people earn the company money.
"Then Pelco will tell you to eat shit when a warranty claim comes up."
So don't fucking break anything!
They can claim it's an "appliance" all day long. Doesn't make it true.
Tell the supplier to eat shit and install the patches yourself.
This is not rocket science.
"are remotely exploitable problems with windows, especially if the boxes are unpatched, "
No. Not "especially" if they are unpatched. Only if they are unpatched. If they were patched (and in both cases the patches were available long before the exploits were) then they were not vulnerable.
So that means an admin has to do his job. Guess what? That requirement ain't OS-dependent.
"These systems cannot run anti-virus software at the same time they record "
/any/ OS) is to patch (and you can probably help out by e.g. ensuring that you use processors with non-exec page protection). Switching to some other platform will not remove the need to patch.
Rubbish.
AV software should impose zero overhead on the recording process, because AV software should not be scanning the (presumably large) data files produced by the recording. All online scanners I'm aware of let you choose between scanning only executables, all files, or user-specified extensions. Use this feature ffs.
And viruses don't just get onto a computer. Viruses get onto computers through improper user behaviour (namely, users running viruses). An online scanner can be effective against this (user education even more so, as well as telling people to stop running arbitrary programs on the recording computers...). Worms can get in through open ports, but the answer to that (on
IOW, do your job and stop blaming the OS.
Sorry, but this is bollocks. This kind of moral relativism is one of the big things wrong with the world. Oh, the Chinese government isn't /bad/ because it suppresses speech, because it imprisons arbitrarily, because it tortures, because it executes willy nilly. It's just different!
/bad/. They're not just different. They're /bad/. They're worse than what western governments do. The Chinese government is an abusive and tyrannical regime.
These things are simply
Christ, next thing you know you'll be telling us that the fucking holocaust was perfectly "valid", it's simply that certain high-ranking Nazis had different "values" from the rest of us, and we should have respected those values.
It's hilarious that you criticize others for not looking at anything other than the "politically correct" view. Because saying "our culture is better than yours" is not "politically correct". On the contrary, it's YOUR position--the one where you defend monstrous regimes as simply having different but valid values that we should respect--that is the politically correct position. It's complete bullshit. People like you will justify any act, no matter how horrendous.
If American companies should ignore local law when operating in China, does this mean that Dutch companies should now be encouraged to sell pot in the US?
"You're right, you should make backups."
Right. And so why do you give a shit about the OS and apps? The OS is on a silver disk with the PC. Backing up means that your important shit is on a gold disk. The copy of the OS on the hard drive at this point becomes practically worthless. You can do whatever you want to it and it really doesn't matter. And if you don't have a backup you're fucked anyway, regardless of the state of the OS.
It's the data that matters, and most OSes are pisspoor at protecting data, because the person who created the data can also destroy it. Perhaps a capability-based system would offer an improvement in this area, but there are no widely used capability-based systems around, so at this point in time it's academic; it's not even obvious that the system would be usable by untrained and unskilled users (that is to say, it's completely unacceptable for the computer to tell the user "Access Denied" when they try to do something they want to do; DAC is bad enough, and I'm suspicious that MAC and capabilities systems will be even worse).
Protecting the system is more important in multi-user systems, but the typical desktop PC isn't multi-user, so it's not an important consideration. I'm not the least bit convinced that Unix is any better at protecting the system in such systems. But for the purposes of this argument it doesn't actually matter.
Unfortunately, that isn't what happened. The patent was upheld, because the software claimed to be prior art was deemed to be different; the differences between the prior art and the patent were negligble or non-existant, and the patent was certainly obvious, but the courts don't care.
Maybe they intend to appeal and don't want to be seen conceding the point?
You must remember that the courts have fuck all notion of what it is to be "obvious" (or not). To a software developer it would be argued that, for example, the Eolas patent was an obvious and natural refinement of existing technologies (using in-process dynamically linked libraries instead of external processes) and thus should have been invalidated by the copious prior art (which covered approximately 99.9% of the same material). If nothing else, you'd think that independent discovery (which is not held to be a protection against charges of patent violation) would demonstrate obviousness.
But not in the eyes of the courts or the patent office.
It would frankly not surprise me if the same were true in this case.