IF YOU ARE CORRECT, you must have a reasonable justification for the use of the registry that is credibly better than using a flat-file approach. I bet you don't have one.:)
* Better performance (parsing text files is expensive)
* Better security (best you can do with flat files is break them out into more flat files with less content in each, vs each Registry key having its own ACL)
* Greater reliability (Registry operations are atomic, hand-editing text files isn't)
* Better consistency (how many different ways of formatting a text file are there ? How many different ways of editing them ?)
* More human efficiency (how many times has the text-file-parser wheel been reinvented ? How much time has been wasted figuring out how to format the text file ? How much time has been wasted because sometimes there's a difference between spaces and tabs ? How much time is wasted trying to find badly-named files in inconsistent locations ?)
* Input validation (what stops you putting text where a number should go in a text file ?)
The Registry is sound software engineering. The train wreck of drunken-vomit-esque ASCII randomness in/etc, by comparison, has only a single, corner-case, redeeming feature - it's easier to recover by hand (as opposed to just restoring from backup) in case of disaster. Like most aspects of UNIX, it reeks of a quick, short-sighted fix thrown together in an afternoon that's subsequently been hacked, patched, bent, twisted, kludged and evolved for 30-odd years afterwards until it has become so embedded into the psyche of the typical UNIX nerd, that they're like child-abuse victims who simply cannot comprehend a life outside of the tiny locked room they get their nightly beatings in.
For example, is there a really good UI reason that when you are editing a document, the 'home' key takes your view to the top of the document but leaves the cursor where you were so that if you then use the cursor keys you zoom back to that location in the document? I'm sure that someone at Apple had a great reason for *that*...
I believe the theory is that the Home/End/Page Up/Page Down keys are to change the "viewport", but moving the insertion point (ie: cursor) should be done with either the arrow keys or the mouse. So moving the insertion point to the start and finish of a line, etc, is done by $MODIFIER+[arrow] (or the mouse), but moving the "view" is done with the dedicated keys. It's to enforce a separation between moving the insertion point and changing your view.
Again, this is behaviour that's been present on MacOS basically forever. Most (long-time) Mac users seem to prefer it.
How can someone write down (document) the undocumented protocol?
I don't want the "undocumented protocol" documented, I want examples of applications that broke because they'd been written to use a documented API which was subsequently changed without notice (or further documentation).
As we all know by now, MicroSoft cannot handle competition. So what did Microsoft do? What they always do, they changed the API and then didn't tell anyone. So now all kinds of MAPI complient applications started breaking, well except theirs of course, since they had all the documentation and the rest of the word didn't.
You can't currently install cygwin on vista so you are left with a pretty looking thing that is slower and less useful than the very old tools of find and grep.
What. The. Hell ? Comparing live indexed search facilities like Vista's (and OS X's, or Beagle) to tools like "find and grep" is like comparing a modern gas oven to a few bits of rock and some straw.
XP Built on top of 2000, which built on top of NT 4, etc. They were more or less business ready, stable OS's. Yes, Vista had to dump all that for security, but the constant delays for "security" purposes now just sound like a lot of bullshit.
I think you're confused.
Windows NT 4.0 = Windows NT 4.0
Windows 2000 = Windows NT 5.0
Windows XP = Windows NT 5.1
Windows 2003 = Windows NT 5.2 (although it should probably be 5.5 or so)
Windows Vista = Windows NT 6.0
There are significant changes in Vista, certainly, but it's still just the latest major revision of Windows NT. In particular, the vast majority of (visible) changes regarding security are in the UI and default configuration, not the "core" OS.
Windows 95 ran quite nicely on such a system (assuming you had 32 bit drivers for your hardware) - certainly better than Windows 3.x did.
Your hyperbole is self-serving and incorrect. Windows 95 ran poorly on very respectable (yet meagre) systems when it was released.
Windows 95 ran well enough on anything with 8MB or more of RAM (an average new PC in 1995). Certainly no worse - and arguably better - than OS/2 did on the same hardware.
They could have designed a dynamic prioritized engine for XP to make sure GUIs immediately recieve faster responses during high cpu load.
They did. The foreground application process(es) run at a higher priority than the rest of the GUI, which itself runs at a higher priority to non-interactive processes. This has been true, well, pretty much forever (even Windows 95 did it IIRC) on workstation/desktop versions of Windows.
(Unless you manually change the "Processor Scheduling" to "Background services", which basically makes it act like Windows "Server".)
I think Vista is a crappy OS that was pushed off the shelf too early because MS was unwilling ot fix it's kernel. The very core behind their OS is worthless.
Apples theory is that their memory management is so good and so transparent that it shouldn't matter. (I disagree -- when I close an application window I *intend* for the application to exit.)
It's got nothing to do with the memory manager and everything to do with the UI model. That's just the way the MacOS GUI works, and it's been like that since MacOS first became capable of running multiple programs.
MacOS's GUI is application-centric, and hence makes a distinction between an application and the windows ("documents") it has open. You hit the button to close a document, you quit the application to stop it running. There's nothing "deceptive" about this behaviour unless you're coming in with preconceived notions about how it should work, rather than learning how it actually works.
What that means is that trying to work on any design/documentation etc while rendering in the background feels like wading through molasses. There hasn't been an OS that'll multitask applications effectively on commodity hardware since the Amiga in the late 80's, and Vista won't change that dismal record.
I've been happily (and heavily) multitasking applications on Windows NT variants (and OS/2 before that) for a decade and a half now. WTF are you on about ?
An "OS for the masses" should make it obvious what memory is in active use and what memory is just being used for aggressive caching. The interface should be designed such that my computer illiterate mother-in-law would be able to figure out that for herself. It should not require the saavy of Geeky Linux Zealot to make the distinction.
Your computer illiterate mother-in-law doesn't even know what RAM is. A display that differentiates between the various different types of RAM usage is, pretty much by definition, utterly irrelevant to her.
Outside of obsessive-compulsive nerds (/wannabe nerds), there is a vanishingly small number of people who give two hoots about how much of their RAM is being used for what (and even fewer of them who have any real reason to).
The reason this is a nonsensical argument is that windows vista does not provide any features substantially in advance of windows xp.
Yes, it does.
In fact, Microsoft claimed that Windows Vista would be the fastest windows yet. But in spite of its limited improvements in functionality - which are almost all supposedly speed-related - it is dramatically slower.
Depends on the system. Vista will be faster on a higher end system, XP will probably be faster on a lower end system. This is a quite common (pretty much standard, really) state of affairs for major OS updates.
Everything is slower on Vista.
Untrue. Its vastly improved IO scheduling, for one example, will improve performance across the board.
Intel would *love* to see an end to the Microsoft monopoly. MS has had Intel by the short and curlies for some time; MS is the reason that Intel cannot work with non-x86 CPUs, and what killed the (somewhat) competitive Itanium 2.
Given that Microsoft have had a portable, multiplatform OS for ~15 years now (the development of which was in part to get away from being chained to a single hardware platform) and that said OS has been mainstream for ~6 years, exactly what is your rationale here ?
Why do you think that Intel has such excellent linux drivers cross the board? You can bet that Intel, although a MS ally, is tired of living under the Wintel shadow.
Why would Intel care what OS is running on their hardware ?
I mean, I like cubes, but it was a bad idea. Apple used to offer half-crippled computers like that, with one PCI or NuBus slot, and 90% of them went unused. That's why Apple just ships machines with no expansion; they realized that the only people who would use them were people who would have no business owning such a computer.
Hardware-wise the Cube was quite a reasonable machine. People even (eventually) managed to shoehorn dual G4s into them.
The reason the Cube failed in the market was its outrageous pricetag, not any significant failing in the hardware.
I'm flattered by all your attention to my original post, but clearly there are at least SOME consumers who want a smaller, more efficient operating system. Why can't Microsoft provide such a product?
Because when a multi-Ghz, multiprocessor, gigabytes of RAM computer costs less a thousand US$, the number of people interested in running Vista on a 10+ year old computer is vanishingly small (ie: unprofitable).
There are what, like 23 version of Vista? Couldn't they have made ONE version that left out all the DRM and the slow file moving and copying, and all the supposed "new features"?
Sure, they probably could. But with the miniscule (and largely disinterested) potential customer base, why bother ? People who are too cheap to spend a few hundred bucks on a PC with something faster than a P2, aren't likely to be spending a few hundred bucks on Vista.
Surely there was room in all those version for a more efficient operating system that was targeted to professional A/V producers or coders or other people that have to make a living with their computer.
If you're a "professional A/V producer or coder", you have a fairly modern machine and Vista will be faster or, at worst, no slower. The performance gap will improve as hardware resources increase. Over the next ~7 years, Vista will make vastly better use of the 4 - 16 core machines with multiple PCIe 3D GPUs and 4 - 16G of RAM that will become common, than XP ever could.
Perhaps you can send your boss a memo to this effect asking them if they could think about this?
Oh. I see what you did there. How very clever ! I imagine you nearly need to change your underwear after that exercise in wittiness.
It is, assuming you have a reasonable computer to start with. This gap will widen in the future as more powerful hardware becomes available.
Or a secure, latest technology file system?
Keep waving your hands. You might be able to get airborne.
Or how about the feature that lets me turn off all the DRM that doesn't have anything to do with protecting MY content?
That would defeat the purpose of DRM, which is ostensibly to protect *other people's* "rights".
Avoiding DRM is trivial. Don't buy DRM-encumbered media. This is true regardless of OS, or anything else. DRM is an attribute of the media, not the player. Vista will not impact your usage of DRM-free media, and non-DRM capable software/hardware will not play DRM-encumbered media. Criticisms of Vista regarding DRM are *inherently, completely meritless*.
Or how about just being able to run all of my programs that ran under XP?
Blame the developer of your broken program, it's their fault. While Microsoft expend herculean efforts to keep old, broken software working, they can't always make it happen.
I'm a consumer and Vista has nothing I want that XP doesn't already have.
The plural of anecdote, is not data.
Oh wait, a weird docking bar over on the right that takes up 30 percent of the screen! NOT!
Then why didn't you? Was it not also trivial to post a few?
Precisely because they _are_ so ridiculously trivial that I should not need to. However:
A new version is supposed to have at least the same functionalities as the previous versions.
* The previous version may have had incorrect behaviour. For example, a security bug might allow programs to access parts of the system they should not be able to, and badly written programs may be written assuming this access is present. Future versions patch the hole and break these programs. This kind of problem seems to be particularly prevalent on Windows and scenarios like this are common to the vast majority of "the latest version of Windows breaks my apps" stories.
* A new version may implement a system to only allow loading drivers that have passed certain correctness and validation checks, to block drivers with obvious and preventable bugs in them from crashing the OS. Therefore old, buggy drivers would no longer run, resulting in "less functionality".
In both cases, clearly net improvements, yet less functionality - broken programs - and both perfectly justifiable.
When using exactly the same functionalities as the previous version, one could expect the new version to take less resource or at least, to not take more.
* The new version may introduce checks of security and/or input correctness. These checks would have greater overheads than not making them. A clear improvement, but at a performance cost.
* Scheduling, locking, etc improvements may be introduced to allow greater performance scalability or consistency, at the cost of individual application performance (ie: individual apps run slower, but the system as a whole can run more applications faster). Again, an overall improvement, but by some measures, worse performance.
These are some obvious examples. They are relevant to most platforms, as they have been improved over time to take advantage of hardware resources that were nearly unimaginable only 10-15 years ago. For another example, a VM system tuned to work well on systems with relatively limited amounts of real RAM (say, tens of megabytes), will not work as well in systems with plentiful real RAM (say, hundreds of megabytes), as a VM system tuned with the assumption of such an environment. For a more practical example, a Linux 2.0 kernel will perform substantially better running a single program on an 8MB 386 than a 2.6 kernel - but a 2.6 kernel will stomp all over a 2.0 kernel on a 16 CPU machine with 64GB of RAM handling a few hundred interactive logins.
Here in Dundee we have a lot of people or chinese and asian origin, lots of them are born and bred here in Scotland with parents or grandparents who emigrated to Scotland. In the shop I work in we get non-whites in every day, and Glasgow for instance has a huge asian population.
Indeed. One of the most surreal experiences in my recent trip through the UK was hearing an old Asian lady - not a day under 80, I'm sure - start talking in one of the thickest Scottish accents I've ever heard (the type where all you can answer with is "what ?").
This wasn't in Glasgow though. Edinburgh, I think it was.
Slashdot Mirror
IF YOU ARE CORRECT, you must have a reasonable justification for the use of the registry that is credibly better than using a flat-file approach. I bet you don't have one. :)
* Better performance (parsing text files is expensive)
* Better security (best you can do with flat files is break them out into more flat files with less content in each, vs each Registry key having its own ACL)
* Greater reliability (Registry operations are atomic, hand-editing text files isn't)
* Better consistency (how many different ways of formatting a text file are there ? How many different ways of editing them ?)
* More human efficiency (how many times has the text-file-parser wheel been reinvented ? How much time has been wasted figuring out how to format the text file ? How much time has been wasted because sometimes there's a difference between spaces and tabs ? How much time is wasted trying to find badly-named files in inconsistent locations ?)
* Input validation (what stops you putting text where a number should go in a text file ?)
The Registry is sound software engineering. The train wreck of drunken-vomit-esque ASCII randomness in /etc, by comparison, has only a single, corner-case, redeeming feature - it's easier to recover by hand (as opposed to just restoring from backup) in case of disaster. Like most aspects of UNIX, it reeks of a quick, short-sighted fix thrown together in an afternoon that's subsequently been hacked, patched, bent, twisted, kludged and evolved for 30-odd years afterwards until it has become so embedded into the psyche of the typical UNIX nerd, that they're like child-abuse victims who simply cannot comprehend a life outside of the tiny locked room they get their nightly beatings in.
For example, is there a really good UI reason that when you are editing a document, the 'home' key takes your view to the top of the document but leaves the cursor where you were so that if you then use the cursor keys you zoom back to that location in the document? I'm sure that someone at Apple had a great reason for *that*...
I believe the theory is that the Home/End/Page Up/Page Down keys are to change the "viewport", but moving the insertion point (ie: cursor) should be done with either the arrow keys or the mouse. So moving the insertion point to the start and finish of a line, etc, is done by $MODIFIER+[arrow] (or the mouse), but moving the "view" is done with the dedicated keys. It's to enforce a separation between moving the insertion point and changing your view.
Again, this is behaviour that's been present on MacOS basically forever. Most (long-time) Mac users seem to prefer it.
How can someone write down (document) the undocumented protocol?
I don't want the "undocumented protocol" documented, I want examples of applications that broke because they'd been written to use a documented API which was subsequently changed without notice (or further documentation).
As we all know by now, MicroSoft cannot handle competition. So what did Microsoft do? What they always do, they changed the API and then didn't tell anyone. So now all kinds of MAPI complient applications started breaking, well except theirs of course, since they had all the documentation and the rest of the word didn't.
For example ?
They all "work". Most are Win XP. One is Win 2000. Why would I change? They work!
Do your Linux boxes "work" ? Will you ever be upgrading them ? Did they "work" before the last time you upgraded them ?
Yeah, I feel your pain. Just last week I had the same problem trying to buy a new BMW without an engine.
You can't currently install cygwin on vista so you are left with a pretty looking thing that is slower and less useful than the very old tools of find and grep.
What. The. Hell ? Comparing live indexed search facilities like Vista's (and OS X's, or Beagle) to tools like "find and grep" is like comparing a modern gas oven to a few bits of rock and some straw.
XP Built on top of 2000, which built on top of NT 4, etc. They were more or less business ready, stable OS's. Yes, Vista had to dump all that for security, but the constant delays for "security" purposes now just sound like a lot of bullshit.
I think you're confused.
Windows NT 4.0 = Windows NT 4.0
Windows 2000 = Windows NT 5.0
Windows XP = Windows NT 5.1
Windows 2003 = Windows NT 5.2 (although it should probably be 5.5 or so)
Windows Vista = Windows NT 6.0
There are significant changes in Vista, certainly, but it's still just the latest major revision of Windows NT. In particular, the vast majority of (visible) changes regarding security are in the UI and default configuration, not the "core" OS.
Who cares? Anyone who had 8M of RAM and a 486.
Windows 95 ran quite nicely on such a system (assuming you had 32 bit drivers for your hardware) - certainly better than Windows 3.x did.
Your hyperbole is self-serving and incorrect. Windows 95 ran poorly on very respectable (yet meagre) systems when it was released.
Windows 95 ran well enough on anything with 8MB or more of RAM (an average new PC in 1995). Certainly no worse - and arguably better - than OS/2 did on the same hardware.
That is to say, Vista imposes a large and pointless cost on the vast majority of users for the purchase of "well supported hardware".
An Aero-capable video card costs ca. US$30, or is included "for free" in modern video-capable chipsets.
They could have designed a dynamic prioritized engine for XP to make sure GUIs immediately recieve faster responses during high cpu load.
They did. The foreground application process(es) run at a higher priority than the rest of the GUI, which itself runs at a higher priority to non-interactive processes. This has been true, well, pretty much forever (even Windows 95 did it IIRC) on workstation/desktop versions of Windows.
(Unless you manually change the "Processor Scheduling" to "Background services", which basically makes it act like Windows "Server".)
I think Vista is a crappy OS that was pushed off the shelf too early because MS was unwilling ot fix it's kernel. The very core behind their OS is worthless.
What's wrong with it ?
Apples theory is that their memory management is so good and so transparent that it shouldn't matter. (I disagree -- when I close an application window I *intend* for the application to exit.)
It's got nothing to do with the memory manager and everything to do with the UI model. That's just the way the MacOS GUI works, and it's been like that since MacOS first became capable of running multiple programs.
MacOS's GUI is application-centric, and hence makes a distinction between an application and the windows ("documents") it has open. You hit the button to close a document, you quit the application to stop it running. There's nothing "deceptive" about this behaviour unless you're coming in with preconceived notions about how it should work, rather than learning how it actually works.
What that means is that trying to work on any design/documentation etc while rendering in the background feels like wading through molasses. There hasn't been an OS that'll multitask applications effectively on commodity hardware since the Amiga in the late 80's, and Vista won't change that dismal record.
I've been happily (and heavily) multitasking applications on Windows NT variants (and OS/2 before that) for a decade and a half now. WTF are you on about ?
An "OS for the masses" should make it obvious what memory is in active use and what memory is just being used for aggressive caching. The interface should be designed such that my computer illiterate mother-in-law would be able to figure out that for herself. It should not require the saavy of Geeky Linux Zealot to make the distinction.
Your computer illiterate mother-in-law doesn't even know what RAM is. A display that differentiates between the various different types of RAM usage is, pretty much by definition, utterly irrelevant to her.
Outside of obsessive-compulsive nerds (/wannabe nerds), there is a vanishingly small number of people who give two hoots about how much of their RAM is being used for what (and even fewer of them who have any real reason to).
The reason this is a nonsensical argument is that windows vista does not provide any features substantially in advance of windows xp.
Yes, it does.
In fact, Microsoft claimed that Windows Vista would be the fastest windows yet. But in spite of its limited improvements in functionality - which are almost all supposedly speed-related - it is dramatically slower.
Depends on the system. Vista will be faster on a higher end system, XP will probably be faster on a lower end system. This is a quite common (pretty much standard, really) state of affairs for major OS updates.
Everything is slower on Vista.
Untrue. Its vastly improved IO scheduling, for one example, will improve performance across the board.
Intel would *love* to see an end to the Microsoft monopoly. MS has had Intel by the short and curlies for some time; MS is the reason that Intel cannot work with non-x86 CPUs, and what killed the (somewhat) competitive Itanium 2.
Given that Microsoft have had a portable, multiplatform OS for ~15 years now (the development of which was in part to get away from being chained to a single hardware platform) and that said OS has been mainstream for ~6 years, exactly what is your rationale here ?
Why do you think that Intel has such excellent linux drivers cross the board? You can bet that Intel, although a MS ally, is tired of living under the Wintel shadow.
Why would Intel care what OS is running on their hardware ?
I mean, I like cubes, but it was a bad idea. Apple used to offer half-crippled computers like that, with one PCI or NuBus slot, and 90% of them went unused. That's why Apple just ships machines with no expansion; they realized that the only people who would use them were people who would have no business owning such a computer.
Hardware-wise the Cube was quite a reasonable machine. People even (eventually) managed to shoehorn dual G4s into them.
The reason the Cube failed in the market was its outrageous pricetag, not any significant failing in the hardware.
Because everyone knew they were coming eventually.
Now, a single-socket, dual-core Mac Pro (or similar), *that* would be news.
I'm flattered by all your attention to my original post, but clearly there are at least SOME consumers who want a smaller, more efficient operating system. Why can't Microsoft provide such a product?
Because when a multi-Ghz, multiprocessor, gigabytes of RAM computer costs less a thousand US$, the number of people interested in running Vista on a 10+ year old computer is vanishingly small (ie: unprofitable).
There are what, like 23 version of Vista? Couldn't they have made ONE version that left out all the DRM and the slow file moving and copying, and all the supposed "new features"?
Sure, they probably could. But with the miniscule (and largely disinterested) potential customer base, why bother ? People who are too cheap to spend a few hundred bucks on a PC with something faster than a P2, aren't likely to be spending a few hundred bucks on Vista.
Surely there was room in all those version for a more efficient operating system that was targeted to professional A/V producers or coders or other people that have to make a living with their computer.
If you're a "professional A/V producer or coder", you have a fairly modern machine and Vista will be faster or, at worst, no slower. The performance gap will improve as hardware resources increase. Over the next ~7 years, Vista will make vastly better use of the 4 - 16 core machines with multiple PCIe 3D GPUs and 4 - 16G of RAM that will become common, than XP ever could.
Perhaps you can send your boss a memo to this effect asking them if they could think about this?
Oh. I see what you did there. How very clever ! I imagine you nearly need to change your underwear after that exercise in wittiness.
How about being faster than XP, for starters?
It is, assuming you have a reasonable computer to start with. This gap will widen in the future as more powerful hardware becomes available.
Or a secure, latest technology file system?
Keep waving your hands. You might be able to get airborne.
Or how about the feature that lets me turn off all the DRM that doesn't have anything to do with protecting MY content?
That would defeat the purpose of DRM, which is ostensibly to protect *other people's* "rights".
Avoiding DRM is trivial. Don't buy DRM-encumbered media. This is true regardless of OS, or anything else. DRM is an attribute of the media, not the player. Vista will not impact your usage of DRM-free media, and non-DRM capable software/hardware will not play DRM-encumbered media. Criticisms of Vista regarding DRM are *inherently, completely meritless*.
Or how about just being able to run all of my programs that ran under XP?
Blame the developer of your broken program, it's their fault. While Microsoft expend herculean efforts to keep old, broken software working, they can't always make it happen.
I'm a consumer and Vista has nothing I want that XP doesn't already have.
The plural of anecdote, is not data.
Oh wait, a weird docking bar over on the right that takes up 30 percent of the screen! NOT!
Might be time to upgrade that 14" monitor.
Then why didn't you? Was it not also trivial to post a few?
Precisely because they _are_ so ridiculously trivial that I should not need to. However:
* The previous version may have had incorrect behaviour. For example, a security bug might allow programs to access parts of the system they should not be able to, and badly written programs may be written assuming this access is present. Future versions patch the hole and break these programs. This kind of problem seems to be particularly prevalent on Windows and scenarios like this are common to the vast majority of "the latest version of Windows breaks my apps" stories.
* A new version may implement a system to only allow loading drivers that have passed certain correctness and validation checks, to block drivers with obvious and preventable bugs in them from crashing the OS. Therefore old, buggy drivers would no longer run, resulting in "less functionality".
In both cases, clearly net improvements, yet less functionality - broken programs - and both perfectly justifiable.
* The new version may introduce checks of security and/or input correctness. These checks would have greater overheads than not making them. A clear improvement, but at a performance cost.
* Scheduling, locking, etc improvements may be introduced to allow greater performance scalability or consistency, at the cost of individual application performance (ie: individual apps run slower, but the system as a whole can run more applications faster). Again, an overall improvement, but by some measures, worse performance.
These are some obvious examples. They are relevant to most platforms, as they have been improved over time to take advantage of hardware resources that were nearly unimaginable only 10-15 years ago. For another example, a VM system tuned to work well on systems with relatively limited amounts of real RAM (say, tens of megabytes), will not work as well in systems with plentiful real RAM (say, hundreds of megabytes), as a VM system tuned with the assumption of such an environment. For a more practical example, a Linux 2.0 kernel will perform substantially better running a single program on an 8MB 386 than a 2.6 kernel - but a 2.6 kernel will stomp all over a 2.0 kernel on a 16 CPU machine with 64GB of RAM handling a few hundred interactive logins.
We just need to stop suing them so much and torturing them with inhuman work hours.
If the hours are "inhuman", then pretty much by definition there aren't enough of them.
Here in Dundee we have a lot of people or chinese and asian origin, lots of them are born and bred here in Scotland with parents or grandparents who emigrated to Scotland. In the shop I work in we get non-whites in every day, and Glasgow for instance has a huge asian population.
Indeed. One of the most surreal experiences in my recent trip through the UK was hearing an old Asian lady - not a day under 80, I'm sure - start talking in one of the thickest Scottish accents I've ever heard (the type where all you can answer with is "what ?").
This wasn't in Glasgow though. Edinburgh, I think it was.
Hey... which version comes without the DRM feature?
You're looking in the wrong place. DRM is applied to the media, and has nothing to do with the OS.