The main difference between nVidia's 'pro' and 'consumer' (Quadro and Geforce) lines is the drivers.
I believe the Quadro chips also undergo more stringent testing - they are optimised for display quality, not performance.
Since Apple write their own drivers for OS X (based on the nVidia reference drivers), it is not entirely a fair comparison.
Nevertheless, it can't be validly claimed that a Mac Pro is massively cheaper than an equivalent dell when the video card difference alone accounts for ca. AU$1400 - $1500.
(This is not to detract from the excellent value the Mac Pro represents, although it's a shame the default video card is so weak and aftermarket options are still limited).
I honestly have never understood this idea that Macs would suddenly get more interest from Virus writers if they had market share.
There are a few important reasons for this (which may or may not all apply)
1. Return on Investment: Where is the benefit in gaining access to a machine only one out of every 100 people (roughly) uses and which is even more uncommon in business environments ?
2. Infection rates: Any "virus" infection is going to spread far, far more slowly on Macs than PCs. Heck, there's a pretty good chance it wouldn't even hit critical mass, because of Macs' relative scarcity, and never make it outside the initial infection zone.
3. Containment: Mac users are a much smaller and tighter-knit community and subsequently communication lines between them are relatively good. Put simply, its more likely someone will publically make note of any virus outbreaks and more likely other Mac users will find out about them.
To put it simply:
By targeting Windows, a cracker will - at a minimum, all else being equal - capture about two orders of magnitude more machines, will have access to machines much more likely to be "interesting" and will benefit from vastly higher propagation rate than they will targeting Macs.
Just consider it like a biological virus... Which one will spread faster, cause more damage and have a higher profile, the virus that only infects one in every 100 people or the virus that infects ~95 in every 100 people ?
Also, don't forget to add Apple Care to the Mac Pro (Dell includes a 3-year On-site Economy Plan).
Similarly, don't forget that the Precision 490 (at least here in Australia) comes by default with a professional 3D graphics card (Quadra FX 3450) that's worth ca. AU$1600 on its own (for comparison, the Mac Pro's pitiful 7300GT is worth ca. AU$120).
The Mac Pro is certainly a (truly amazingly, considering it's Apple) well priced machine (and it's a shame they don't have a less expensive version for us mere mortals), but the difference is not as large as the original poster would have you think. As close as I can get them, which requires adding Applecare to the Mac Pro and bumping both machines up to the Quadra FX 4500, puts the Dell at about AU$300 more.
The advantage Dell has is a model (and price) range that actually covers the massive gap between a 1.66Ghz Core Duo based machine and a quad-core Xeon based machine.
It's true that it's (usually) a poorly designed piece of software that needs to run with administrative privileges but it does stem from Windows' history.
Well, that's true in a kind of academic sense, but Microsoft has been telling people to write non-Admin-friendly software for a good eight years now and had the OS infrastructure to support it.
I'd have to say applications that only ran in Administrator accounts ceased being something that could be reasonably blamed on Microsoft back around 2000.
Also, I guess you haven't seen the various applications that I've seen that won't run with "Run As".
No, actually, I haven't. I sure do hear about them a lot on Slashdot, but I can't say I've ever actually seen one.
Also, giving an office user the ability to use "Run As" kind of defeats the purpose, [...]
Indeed it does, but "office users" are not "home desktop users". "Office users" should be in a managed environment and have an IT department capable of setting the appropriate file and Registry ACLs so that those broken applications work *without* having to hand out access to an Administrator account.
True these are called, sane defaults (like no access to the e-mail address book unless you're the pre-installed e-mail client) [...]
So no mail merge then ?
[...] and asking the user in plain English and with a good UI when an aberration occurs.
The problem is they'll still do it.
"Computers are too stupid" is a cop out for programmers don't want to do it right because it is hard and we have a monopoly so giving customers a good product does not matter.
No, "computers are too stupid" is the *truth*. The reason they're too stupid is because it's pretty much impossible to programmatically detect the difference between "good" and "bad" with any sort of reliability.
They know my gender, how old I am, where I live, and what my income is... they also know what games I play and how often I play them... and this data is 100% accurate (considering my personal info is tied to my credit card, and the machine tracks my gaming habits). So with Xbox Live, Steam, Playstation Network, and Nintendo's online system... exactly why do we still need to rely on easily inaccurate surveys.
Because if they actually used that info for anything like that, the Slashdot headline would be in YRO, have a little Borg icon beside it and say "Microsoft uses XBox 360 to read your email, defraud your credit card and kick your dog !".
It is and will remain hand-craft. Live performance. The production of a movie. The writing of a book. It is trivially easy to make the copy, it is hard to produce the master work.
Then why is there so much of it ? Why was there so much of it before copyright was even an idea, let alone a law ?
If people are interested in spending money on something that I produce, I should be the one getting the money, not some no-talent dork who can only copy my cd's and sell them a little cheaper than me.
But why should you be paid for the one bit of work more than once ? No-one else (except others riding the copyright gravy train) is.
But that's exactly why we need the protections offered by copyright. If the Foo Fighters have a great song that everyone wants, they should have the right to make the money from the song.
Abolishing copyright does not remove their "right" to make money from the song. All it does is remove the ease with which they can make money from the song by not allowing them to sell it over and over and over and over again without actually having to produce it more than once.
However, if anyone can copy and give away or sell the song on any media, then the band loses money, and will lose the incentive.
Not making money != losing money.
Just a nit-pick here, but copyright is not just a "legislated" thing, it's a constitutional thing.
Note that the "constitutional thing" doesn't say anything about the right to make money, or even control distribution. Simply ensuring that no one person could ever claim credit for another's work, would meet the constitutional "requirements".
Copyright - by every meaningful definition of the word - is most certainly a "legislative thing".
If I can't be guarenteed that I would be the sole owner of something that I create, I would have little profit incentive to create it.
Note that (legally) being considered the "sole owner" of something and also allowing other people to copy it are *not* mutually exclusive.
Ideas and information are not physical items. Any system which tries to treat them as such, and denies their fundamental properties, is broken by design.
However, it also costs $5.00 to sit down and have a cup of coffee. Everything is relative. My dad went on a business trip to switzerland, and was amazed at the price of even simple things. Thank diety his company paid for it all. You need to be making $300K in that country just to make ends meet.
I've just spent a week in our Zurich office, since I'm considering a move there for a year or two and I found the price comparisons interesting...
(Bear in mind this is comparing to Australian prices. It was actually very easy, since a Swiss Franc and an Australian Dollar are worth almost the same).
I found the most noticably higher expenses were eating (and drinking) out. You're luck to get away with a relatively simple meal for two and a couple of beers, for less than $80 - $100. Considering $80-$100 in Australia will easily buy you a nice meal *and* a bottle of wine, even in a Sydney restaurant, that's damn expensive.
Food & drink from supermarkets, OTOH, was only marginally more expensive. Except for some types of meats, which were very expensive.
Luxury items - PSPs, iPods, etc - seemed to cost the same as they would in Australia.
Public transport (well, I only used the trams) was quite reasonably priced and worked well.
I couldn't really compare rents, but from what I understand they're relatively high and bargains are pretty much nonexistant (ie: no matter what you look at in a certain price bracket, they're all going to basically the same). Buying property - at least in Zurich - is practically unheard of for anyone who isn't extremely wealthy.
All in all, it was a very nice place. The most amazing thing, to me, was that the river and lake in Zurich is so clean you can see the bottom in shallower areas, and people happily swim in it. All I need to do now is learn to speak German:).
Have you ever done this on a windows machine for an extended period of time?
Yes. For about ten years now.
Some programs don't even run unless you are administrator.
Which is why you have "Run As".
Now if we are talking about a work enviornment then sure, give everyone in the building (except engineering) non-admin accounts, but I would never recommend doing it to someone who didn't have a high level of computer knowledge and patience or an equivalant IT staff on hand to help out with any issues.
True to an extent, but this is the application's fault, not Windows'.
It is properly reacting to the file extension, it just is quite stupidly not displaying it. I disable that on every computer I use, but most people do not even know about that option. Personally, I do not know why such an option would exist.
Because to most people, file extensions are meaningless gibberish at best and confusing computer stuff at worst.
User clicks on.JPG file. Operating system (no names, please) looks at file, says "Oh, that's really an.EXE file, I'll just execute it without asking...".
I guess you must mean OS/2, because Windows sure as hell doesn't do this.
Active Desktop was IIRC just a way of putting HTML or Animations on the desktop - they weren't programs per-se, that could tell you useful stuff.
Active Desktop let you embed ActiveX controls in your Desktop (hence the name...). Since ActiveX controls - as everyone who loves to hate them knows - are just win32 programs, this effectively let you embed whatever the hell you wanted into your desktop.
However, given the specs of the average PC back in 1997, for obvious reasons it never really took off.
**perhaps if you are not informed on this, it's because I and probably that author reported this behaviour to apple and got the response, Whereas you did not.
No, the author stated the behaviour was "proper" based on the documentation he linked to, when the documentation quite clearly - in the second row of "Table 1" - states that it is not.
**Reality intrudes here: on Linux and Windows ubiquitously the applications need root, or whatever you want to call it, so often that everyone does run and install as root level user. Your reply is really pretty strange and weasly lawyer speak
My point was (pretty clearly) that this is not a *Windows* problem, it is an *application* problem.
**Oh come on...this is stupid. Have you ever tried to do large numbers of package installs this way and not basically break the usability of installed libraries for other users, or maintain any consistency between package mangers or had to hand edit other make files unaware of your non-standard installs??? This is completely disingenuous or you are not considering multi-user systems
Er, if you're installing for other users then you *should* require higher than normal privileges. That's kind of the point of a multiuser system.
I'll add that if the average Linux system came with the same (flawed, IMHO) default permissions for its equivalents of/Applications and friends, "admin" users on Linux could also do what they do on OS X, and "install" applications without needing to use sudo.
**errr no it's not. That's the WHOLE point.
From a technical perspective, it is.
**Yes it does, he retorted tersely.
You can retort as tersely as you want, it won't change the fact you're wrong, and every piece of documentation says you're wrong. An application can install in Windows - even to the point of associating itself with filetypes - without needing to "get its hooks into the OS".
**Oh were back to semantics about "no root" on windows. Whatever.
It's not semantics at all, it's an extremely important aspect of OS security. Superusers (ie: root) are security holes.
Added to that, your claim that you need to be "root" (or "Administrator, which is what you actually meant) to modify the Registry is not semantics, but fundamentally wrong, because each user has their own Registry hive - which they can write to - where applications store things like file extensions associations and the like.
(This actually existed even in DOS-based Windows since Windows 98, although obviously it didn't actually have any way of enforcing permissions. It's been around in NT since at least NT4, possibly earlier.)
**There are no application XML config files hiding in the your home directory on a mac.
Yes, there are. I suggest you start in ~/Library/Preferences. That is OS X's equivalent (very broadly speaking) to Windows's per-user Registry hives.
The only thing that ends up in the user prefs folder is the persistent user customization data. But that's not the same thing as what goes in the windows registry.
Yes, it is (well, as much as two platforms so different can be alike, anyway).
**Once again with feeling... since nearly everyone has to run as "root" (or whatever you want to call it on windows) on linux and windows to do installs those selective ACLs are not useful in practice.
Actually they are, because all you need to do to make them useful is (*gasp*) not run as Administrator.
I'll also point out that root on an OS X machine (ie: any time you type your password into that little dialog) has vastly more power over the system than an Administrator user does over a Windows system, because root is a superuser and Administrator is not.
Also, again, ("with feeling") this is an *application* problem, not a *Windows* problem.
That was the point. Once you decide to install as root, there is no selective exception to install here but
So you are admitting that building an anti-virus isn't actually making Windows itself better (or more secure).
Er, no. And I have no idea how you managed to reach that conclusion from a statement referring to the well-known fact that adding more programmers past a certain point rarely results in a product being finished any faster.
MS can still develop and sell an AV. But seeing as it's clearly not part of the OS, there's no reason they should be packaged together--especially when doing so would basically shutout all other AV manufactuers, not because of being a superior product, but because no one wants to go out and buy a second AV when Windows already comes with a pre-packaged one that you've basically already paid for.
Your "save the buggy whip manufacturers" argument is only marginally less worthless than your "it's not part of the OS" argument.
Five years ago people were saying it about a media player. Five years before that, it was a web browser. Five years before that, it was a network stack. Five years before that, it was a GUI shell. Five years before that, it was a CLI shell. Five years before that, it was a BASIC interpreter. Etc, etc.
All these things (well, except the BASIC interpreter) that were - at the time - "clearly not part of an OS" are now considered standard. It's just another step in the commoditisation of software. Get used it, because it's only going to continue. Indeed, enjoy it, because as the end user, you are the beneficiary of lower costs and more functionality.
If you wish to use the "it's not part of the OS argument", then you must also argue that OS vendors - *all* OS vendors - not be allowed to include anything that doesn't fit into the academic definition of "OS". Thus, only allowing them to sell a product that is - to the vast majority of users - pretty much useless (no shell, essentially no hardware support, no supporting libraries, no networking protocols, etc).
Residents get 5GB/week off-campus (unlimited on-campus). If they go over this limit, their off-campus connectivity speed is reduced until their traffic usage goes below a 4GB for the previous 7 day period. Campus traffic is never affected.
In principle this system sounds fair, but I can think of at least one improvement...
5G external is (more than) enough for education-related traffic (assuming you have decent internal mirrors), but you should have a system where students are allow to pay an additional (reasonable - comparable to commercial broadband in the area) fee to be able to increase their allowed bandwidth usage (again, on par with commercial broadband in the area).
I can agree than an on-campus network's primary (if not only) priority is to provide education-related connectivity, but if dorms in the US are anything like colleges in Australia, if you *do* want to have "extra curricular internet access", you're pretty muched fucked - because ADSL doesn't work through the internal phone systems and installing cable is not an option.
The solution is very simple, and I am amazed that TFA didn't at least mention it. The solution is not to base grades on such handed-in work. Instead, base grades on performance that you can ensure is the student's own. Higher (and lower) education have a name for this: exams. Conduct an exam under carefully-controlled conditions, and no cheating is possible.
The problem with exams is that the scenario they present is typically not representative of how the student will be expected to perform in "real life". Their emphasis is generally more on memorisation and recall, than research and problem-solving.
But now, Apple has "installation", where install programs put stuff all over the place, and maybe change the state of the system. Just like Windows. Big step backwards.
It was an inescapable side effect of moving to an OS that actually had the concepts of "security" and "permissions".
I knew it was weird when I installed Parallels a few months ago and it added several kernel extensions without a password prompt. This is a serious design flaw, and yet another reason for developers and users to avoid installer packages unless absolutely necessary.
According to the documentation linked from TFA, this behaviour is neither "proper", nor expected. Assuming the documentation is correct, it's not a design flaw, it's just a bug.
Slashdot Mirror
I believe the Quadro chips also undergo more stringent testing - they are optimised for display quality, not performance.
Since Apple write their own drivers for OS X (based on the nVidia reference drivers), it is not entirely a fair comparison.
Nevertheless, it can't be validly claimed that a Mac Pro is massively cheaper than an equivalent dell when the video card difference alone accounts for ca. AU$1400 - $1500.
(This is not to detract from the excellent value the Mac Pro represents, although it's a shame the default video card is so weak and aftermarket options are still limited).
No, it's not. At *best* the two are equal and in most aspects Windows has the better design. OS X is, after all, just another unix under the hood.
There are a few important reasons for this (which may or may not all apply)
1. Return on Investment: Where is the benefit in gaining access to a machine only one out of every 100 people (roughly) uses and which is even more uncommon in business environments ?
2. Infection rates: Any "virus" infection is going to spread far, far more slowly on Macs than PCs. Heck, there's a pretty good chance it wouldn't even hit critical mass, because of Macs' relative scarcity, and never make it outside the initial infection zone.
3. Containment: Mac users are a much smaller and tighter-knit community and subsequently communication lines between them are relatively good. Put simply, its more likely someone will publically make note of any virus outbreaks and more likely other Mac users will find out about them.
To put it simply:
By targeting Windows, a cracker will - at a minimum, all else being equal - capture about two orders of magnitude more machines, will have access to machines much more likely to be "interesting" and will benefit from vastly higher propagation rate than they will targeting Macs.
Just consider it like a biological virus... Which one will spread faster, cause more damage and have a higher profile, the virus that only infects one in every 100 people or the virus that infects ~95 in every 100 people ?
Similarly, don't forget that the Precision 490 (at least here in Australia) comes by default with a professional 3D graphics card (Quadra FX 3450) that's worth ca. AU$1600 on its own (for comparison, the Mac Pro's pitiful 7300GT is worth ca. AU$120).
The Mac Pro is certainly a (truly amazingly, considering it's Apple) well priced machine (and it's a shame they don't have a less expensive version for us mere mortals), but the difference is not as large as the original poster would have you think. As close as I can get them, which requires adding Applecare to the Mac Pro and bumping both machines up to the Quadra FX 4500, puts the Dell at about AU$300 more.
The advantage Dell has is a model (and price) range that actually covers the massive gap between a 1.66Ghz Core Duo based machine and a quad-core Xeon based machine.
What "deep connections" ? IE is a user-mode component loaded on demand, just like its equivalents in KDE, GNOME and OS X are.
How do you figure that, given the design of Windows - from a security perspective - is *at least* as good as its contemporaries ?
Well, that's true in a kind of academic sense, but Microsoft has been telling people to write non-Admin-friendly software for a good eight years now and had the OS infrastructure to support it.
I'd have to say applications that only ran in Administrator accounts ceased being something that could be reasonably blamed on Microsoft back around 2000.
Also, I guess you haven't seen the various applications that I've seen that won't run with "Run As".
No, actually, I haven't. I sure do hear about them a lot on Slashdot, but I can't say I've ever actually seen one.
Also, giving an office user the ability to use "Run As" kind of defeats the purpose, [...]
Indeed it does, but "office users" are not "home desktop users". "Office users" should be in a managed environment and have an IT department capable of setting the appropriate file and Registry ACLs so that those broken applications work *without* having to hand out access to an Administrator account.
So no mail merge then ?
[...] and asking the user in plain English and with a good UI when an aberration occurs.
The problem is they'll still do it.
"Computers are too stupid" is a cop out for programmers don't want to do it right because it is hard and we have a monopoly so giving customers a good product does not matter.
No, "computers are too stupid" is the *truth*. The reason they're too stupid is because it's pretty much impossible to programmatically detect the difference between "good" and "bad" with any sort of reliability.
Because if they actually used that info for anything like that, the Slashdot headline would be in YRO, have a little Borg icon beside it and say "Microsoft uses XBox 360 to read your email, defraud your credit card and kick your dog !".
Then why is there so much of it ? Why was there so much of it before copyright was even an idea, let alone a law ?
But why should you be paid for the one bit of work more than once ? No-one else (except others riding the copyright gravy train) is.
But that's exactly why we need the protections offered by copyright. If the Foo Fighters have a great song that everyone wants, they should have the right to make the money from the song.
Abolishing copyright does not remove their "right" to make money from the song. All it does is remove the ease with which they can make money from the song by not allowing them to sell it over and over and over and over again without actually having to produce it more than once.
However, if anyone can copy and give away or sell the song on any media, then the band loses money, and will lose the incentive.
Not making money != losing money.
Just a nit-pick here, but copyright is not just a "legislated" thing, it's a constitutional thing.
Note that the "constitutional thing" doesn't say anything about the right to make money, or even control distribution. Simply ensuring that no one person could ever claim credit for another's work, would meet the constitutional "requirements".
Copyright - by every meaningful definition of the word - is most certainly a "legislative thing".
Note that (legally) being considered the "sole owner" of something and also allowing other people to copy it are *not* mutually exclusive.
Ideas and information are not physical items. Any system which tries to treat them as such, and denies their fundamental properties, is broken by design.
I've just spent a week in our Zurich office, since I'm considering a move there for a year or two and I found the price comparisons interesting...
(Bear in mind this is comparing to Australian prices. It was actually very easy, since a Swiss Franc and an Australian Dollar are worth almost the same).
I found the most noticably higher expenses were eating (and drinking) out. You're luck to get away with a relatively simple meal for two and a couple of beers, for less than $80 - $100. Considering $80-$100 in Australia will easily buy you a nice meal *and* a bottle of wine, even in a Sydney restaurant, that's damn expensive.
Food & drink from supermarkets, OTOH, was only marginally more expensive. Except for some types of meats, which were very expensive.
Luxury items - PSPs, iPods, etc - seemed to cost the same as they would in Australia.
Public transport (well, I only used the trams) was quite reasonably priced and worked well.
I couldn't really compare rents, but from what I understand they're relatively high and bargains are pretty much nonexistant (ie: no matter what you look at in a certain price bracket, they're all going to basically the same). Buying property - at least in Zurich - is practically unheard of for anyone who isn't extremely wealthy.
All in all, it was a very nice place. The most amazing thing, to me, was that the river and lake in Zurich is so clean you can see the bottom in shallower areas, and people happily swim in it. All I need to do now is learn to speak German :).
Yes. For about ten years now.
Some programs don't even run unless you are administrator.
Which is why you have "Run As".
Now if we are talking about a work enviornment then sure, give everyone in the building (except engineering) non-admin accounts, but I would never recommend doing it to someone who didn't have a high level of computer knowledge and patience or an equivalant IT staff on hand to help out with any issues.
True to an extent, but this is the application's fault, not Windows'.
No, this is because computers are fundamentally devices that do what you tell them to do, not what you want them to do.
A computer does not know what you mean by "don't do anything bad" unless you actually tell it what "bad" equates to.
Neither does Windows.
Because to most people, file extensions are meaningless gibberish at best and confusing computer stuff at worst.
I guess you must mean OS/2, because Windows sure as hell doesn't do this.
Active Desktop let you embed ActiveX controls in your Desktop (hence the name...). Since ActiveX controls - as everyone who loves to hate them knows - are just win32 programs, this effectively let you embed whatever the hell you wanted into your desktop.
However, given the specs of the average PC back in 1997, for obvious reasons it never really took off.
No, the author stated the behaviour was "proper" based on the documentation he linked to, when the documentation quite clearly - in the second row of "Table 1" - states that it is not.
**Reality intrudes here: on Linux and Windows ubiquitously the applications need root, or whatever you want to call it, so often that everyone does run and install as root level user. Your reply is really pretty strange and weasly lawyer speak
My point was (pretty clearly) that this is not a *Windows* problem, it is an *application* problem.
**Oh come on...this is stupid. Have you ever tried to do large numbers of package installs this way and not basically break the usability of installed libraries for other users, or maintain any consistency between package mangers or had to hand edit other make files unaware of your non-standard installs??? This is completely disingenuous or you are not considering multi-user systems
Er, if you're installing for other users then you *should* require higher than normal privileges. That's kind of the point of a multiuser system.
I'll add that if the average Linux system came with the same (flawed, IMHO) default permissions for its equivalents of /Applications and friends, "admin" users on Linux could also do what they do on OS X, and "install" applications without needing to use sudo.
**errr no it's not. That's the WHOLE point.
From a technical perspective, it is.
**Yes it does, he retorted tersely.
You can retort as tersely as you want, it won't change the fact you're wrong, and every piece of documentation says you're wrong. An application can install in Windows - even to the point of associating itself with filetypes - without needing to "get its hooks into the OS".
**Oh were back to semantics about "no root" on windows. Whatever.
It's not semantics at all, it's an extremely important aspect of OS security. Superusers (ie: root) are security holes.
Added to that, your claim that you need to be "root" (or "Administrator, which is what you actually meant) to modify the Registry is not semantics, but fundamentally wrong, because each user has their own Registry hive - which they can write to - where applications store things like file extensions associations and the like.
(This actually existed even in DOS-based Windows since Windows 98, although obviously it didn't actually have any way of enforcing permissions. It's been around in NT since at least NT4, possibly earlier.)
**There are no application XML config files hiding in the your home directory on a mac.
Yes, there are. I suggest you start in ~/Library/Preferences. That is OS X's equivalent (very broadly speaking) to Windows's per-user Registry hives.
The only thing that ends up in the user prefs folder is the persistent user customization data. But that's not the same thing as what goes in the windows registry.
Yes, it is (well, as much as two platforms so different can be alike, anyway).
**Once again with feeling... since nearly everyone has to run as "root" (or whatever you want to call it on windows) on linux and windows to do installs those selective ACLs are not useful in practice.
Actually they are, because all you need to do to make them useful is (*gasp*) not run as Administrator.
I'll also point out that root on an OS X machine (ie: any time you type your password into that little dialog) has vastly more power over the system than an Administrator user does over a Windows system, because root is a superuser and Administrator is not.
Also, again, ("with feeling") this is an *application* problem, not a *Windows* problem.
That was the point. Once you decide to install as root, there is no selective exception to install here but
Er, no. And I have no idea how you managed to reach that conclusion from a statement referring to the well-known fact that adding more programmers past a certain point rarely results in a product being finished any faster.
MS can still develop and sell an AV. But seeing as it's clearly not part of the OS, there's no reason they should be packaged together--especially when doing so would basically shutout all other AV manufactuers, not because of being a superior product, but because no one wants to go out and buy a second AV when Windows already comes with a pre-packaged one that you've basically already paid for.
Your "save the buggy whip manufacturers" argument is only marginally less worthless than your "it's not part of the OS" argument.
Five years ago people were saying it about a media player. Five years before that, it was a web browser. Five years before that, it was a network stack. Five years before that, it was a GUI shell. Five years before that, it was a CLI shell. Five years before that, it was a BASIC interpreter. Etc, etc.
All these things (well, except the BASIC interpreter) that were - at the time - "clearly not part of an OS" are now considered standard. It's just another step in the commoditisation of software. Get used it, because it's only going to continue. Indeed, enjoy it, because as the end user, you are the beneficiary of lower costs and more functionality.
If you wish to use the "it's not part of the OS argument", then you must also argue that OS vendors - *all* OS vendors - not be allowed to include anything that doesn't fit into the academic definition of "OS". Thus, only allowing them to sell a product that is - to the vast majority of users - pretty much useless (no shell, essentially no hardware support, no supporting libraries, no networking protocols, etc).
In principle this system sounds fair, but I can think of at least one improvement...
5G external is (more than) enough for education-related traffic (assuming you have decent internal mirrors), but you should have a system where students are allow to pay an additional (reasonable - comparable to commercial broadband in the area) fee to be able to increase their allowed bandwidth usage (again, on par with commercial broadband in the area).
I can agree than an on-campus network's primary (if not only) priority is to provide education-related connectivity, but if dorms in the US are anything like colleges in Australia, if you *do* want to have "extra curricular internet access", you're pretty muched fucked - because ADSL doesn't work through the internal phone systems and installing cable is not an option.
The problem with exams is that the scenario they present is typically not representative of how the student will be expected to perform in "real life". Their emphasis is generally more on memorisation and recall, than research and problem-solving.
It was an inescapable side effect of moving to an OS that actually had the concepts of "security" and "permissions".
According to the documentation linked from TFA, this behaviour is neither "proper", nor expected. Assuming the documentation is correct, it's not a design flaw, it's just a bug.