Do you try to look for an explanation, whatever it may be, or do you sit on it?
You sit on it, because the evidence isn't enough to explain it. The point of an explanation is understanding the true cause of something, not filling unknown details with UFO nonsense. If you don't have enough evidence, you are simply out of luck, you just can't explain everything. We don't lock up random people in jails just because we can't find the real murderer either, if evidence isn't enough we simply "sit on it" too.
But they tend to focus on average Joe's UFO sightings rather than the well documented and really hard to explain ones.
You miss the point. If you want to prove the existence of alien spacecraft you don't have to find hard to explain cases, you have to find the opposite, well explained ones. You need clear pictures and well observed objects, not anomalies in the sky that you couldn't identify.
And my appeal to authority is justified in that you'd think that a highly trained elite pilot who's flown for years and even been to space would know what he's looking at when he looks in the sky around him.
The human visual system is build to measure distances and velocities at small scales. When you fly high up in the air you have the exact opposite, huge velocities, great distances and worst of all nothing to use as reference. No amount of experience will help you when you encounter an unexpected thing in the sky, as it becomes nearly impossible to judge size or velocity just by sight.
Sort them into the "unexplained" folder and move on. Just because you can't explain something doesn't mean it was an alien spacecraft. That case, as all the others, doesn't give you hard evidence for anything. Do we have now a clear picture of a space craft? Nope. Any idea how its propulsion system works? Nope. Who piloted it? Nope. Any idea about anything at all? Not really.
If you want to demonstrate that alien spacecrafts are real, you have to come up with some good evidence, not just an single unexplained anomaly. Find multiple anomalies that follow the same pattern and you might be getting somewhere, but a single one off doesn't really help you much with anything, especially not when you fill in the lack of facts with random UFO fiction.
When it comes to weird things happening in the air I like the story of British Airways Flight 9, full of mystery and suspense and it also happens to be fully explained in the end.
Seriously, guys who actually went and walked on the Moon
Does that give you magic power to detect alien spacecraft or immunity from mistaking something you see? Your appeal to authority just isn't a very good argument.
Now that's a bit off topic, but the real reason is that "sceptics" like you have settled on a frozen idea of what's possible and what's not,
No, they are skeptic because they haven't seen good evidence.
There's a plethora of UFO reports out there from civil and military pilots, as well as air traffic control staff, radar operators, military base personel, and yes, even astronauts who went to the Moon.
That is well and true, it however has a little problem: Those are UFOs in the literal sense, they are "unidentified flying object", not extra terrestrial spacecrafts. When you can't tell what something is, it simply means you don't know, it doesn't confirm that aliens are involved.
When it comes to hard evidence, there is simply nothing that points to E.T. Blobs of light in the sky can be lots of things, clouds, planets, satellites, floating lanterns, lense flare, insects and tons of more stuff. How many clear non-blurry pictures are there of alien space crafts? None. You'd guess in a time where every mobile has camera people would come up with some good pictures, but that hasn't happened.
That is only true when it comes to syntactic sugar, its not true when it comes to semantics. For example when I write:
for(int i = 0; i < 100; ++i)
lst[i] = do_math(lst[i]);
I don't just tell the compiler what I want to do, but I also say how I want it to be done. The variable i for example has nothing to do with my problem and neither has the forced sequential evaluation. And there isn't an easy way to figure out if do_math() has side effects when it is hidden in some external library either. A compiler will have a much harder time figuring out my intend then in a language that allows me to express what I want to do, instead of how:
map do_math lst
Here the compiler knows instantly that I don't care about the order of evaluation and it also knows that do_math doesn't have side effects because its a function, so it can split up the task across multiple CPUs without a problem.
Why isn't it used as often as say, Python, Perl, or Ruby?
You don't chose a language for their syntax, but for most part for their libraries and thus are lacking in Lua. Lua happens to be mostly intended as an language for embedding, not as a stand alone language.
See, thats exactly the problem. If you invest everything you have into a single solution, you are kind of stuck with it, as you end up being to busy fixing the problems of your current solution to look for alternative solutions. Polywell presents an alternative and tries to tackle the problem from a completly different angle. Nobody knows if it would work out, but if you believe Bussards google talk it would cost a tiny fraction of ITERs cost to build a full scale Polywell reactor to find it out and it wouldn't take 15 years either. ITER has the problem of being way to expensive and way to unsuccessful so far, if you would pump similar amounts of money into alternative solution, you might already have found one.
but it's pretty trivial to put a system together in an evening...assuming you know what you're doing.
And the "knowing what you are doing" is exactly the problem. I don't buy many PCs, because there simply isn't a need for an upgrade every two years, instead I am more in the 5+ year range. So each time I look around for new stuff pretty much everything has changed. Almost all knowledge of previous generations is close to worthless, as power supplies, cables, cards, cooling, cases and so on all have changed. The screws holding things together are still the same, but thats about it.
He in no sense answers the question of why evolution couldn't have provided precisely the capacity he attributes to consciousness without any consciousness involved.
Just look at the last few decades of AI research, which has nicely shown that hard coded computation just doesn't get you very far, because its far to inflexible to react to unexpected events. Its good enough for winning chess, but just not very good to interact in the real world, which simply doesn't follow clear cut rules.
Planning != consciousness
It might not be the same, but its certainly something very similar, as consciousness is in large part about forming a model from the input that can then be used to plan and predict actions without actually acting them out in the real world. Try to build a robot that will act intelligently in the real world and you have to build up some internal model as well and sooner or later you might up with something that could be considered consciousness.
Or to put it another way: We really don't have any reason to assume that consciousness is something magical.
They seem to filter out or forget all off the crappy ones
Well, yeah, because there where enough good ones so that we didn't have to play the crappy ones. Look at the adventure genre, LucasArts had a new title ready every year and each of those was awesome. How often do high quality adventures released today? Not much at all, most are rather mediocre. And that is ignoring all the non-LucasArts adventure games from back in the day, some of them where pretty awesome too. And now don't let me start on the flightsim genre, as that is pretty much dead and burried today and the amount of good flightsim is close to zero.
that beats the crap out of joystick controls in terms of response.
Not really. Have you ever tried to hit anything on that setting without lock-on? Its near impossible because you constantly oversteer. Its not a problem for MP3 since you lock on all the time, but as a general FPS mechanics it doesn't really work. There is also the problem that you will get into a crazy spin each and every time you have a dialog with an NPC, as pointing to the center of the screen is not easy when you don't have any feedback from the interface.
So as a general dual-analogstick replacement I consider the Wiimote pretty much a failure. Aiming works very well, but rotation via the edge of the screen really doesn't. Thus one thing I would like to see is a game that makes only use of the aiming part and leaves all the rotation to the analogstick (i.e. classic Doom/MetroidPrime1 control, but you can aim at enemies that aren't in the center of the screen).
The question is, what market share is required to achieve this?
I doubt that it has much to do with market share, it seems more to be an issue with Linux being pretty much incompatible with how hardware manufacturers like to ship their drivers. Most drivers in the Windows world are not just drivers, they come bundled with a whole bunch of software and stuff that is tied to the specific piece of hardware (i.e. standard Windows Logitech mouse driver is 50MB instead of a few KB). A clean separation between the code that makes your hardware work on all that other additional software doesn't really exit, because the supplied software plays a big part in the marketing and feature lists you find on the box.
I think to get proper Linux support hardware vendors would first need to learn that their job is to produce hardware, not software. Once thats done they might have less problems with releasing specs, but I somewhat doubt that this is going to happen anytime soon because of Linux. The best thing for Linux hardware support in the end are really the open standards. Any USB HID or storage device works on Linux out of the box, not because the hardware vendor cared about Linux, but because he implemented the spec. The more specs we have for common hardware, the better the Linux support will be.
While I was growing up, playing video games pretty much meant you had an Atari, NES, or SNES.
Not where I grew up. C64, Amiga and all those other home computers have existed just as long and provides a ton of gaming goodness.
The gaming industry on the computer has grown by leaps and bounds.
The gaming industry has a whole has grown, and thus the PC market segment has grown too, in relative however the PC market segment has shrunk. For most developers doing PC-only games is no longer an option, most focus on multi-platform titles these days. And most of those titles are targeted for console first with the PC only getting a port much later and often not well adopted to the PC environment (i.e. GUI designed for game controller, not so good for use with a mouse).
Kind of, what PolicyKit does is give a user process a way to communicate to a root process, so that th user does not need root rights to accomplish certain things. That is fine for limiting access to root account, but it doesn't help with a user process having access to all the users files, as their is no way for a user to launch a process that has even less privileges then his user account has.
How old are you exactly? I'm not sure how anyone, who was old enough to know better, could claim that the state of PC video gaming was better in the 90's.
It was better. Back then PC games where very different from console games, you had your platformer and action adventures on the consoles and the PC was filled with adventures, flightsims, FPS and strategy games. Today on the other side most of those PC-exclusive genres are either dead or have been turned into console games. There no longer is a clear differentiation between PC and consoles, all get the same games, but the PC often gets them last, because it has the smallest market share of them (14%). Many once PC exclusive developers have either moved to consoles or at least gone multiplatform.
That doesn't mean PC gaming will die out, its the only open platform around, but it means PC gaming is shifting. The future of PC gaming lies in games like Peggle and indie titles not in AAA blockbuster titles, as those go where the money is and that is consoles.
The closest thing to an exclusive genre that the PC still has are the MMORPGs, but while they make tons of money, I am not so sure that they are actually good for PC gaming as a whole, as they suck up so much time, that other titles have a much harder time.
How exactly is this a problem in the operating system-
The problem is that mainstream Linux has just two ways to run an application: root or user. Both of those have the problem that they can with no problem delete or manipulate all the files a user has access to, for a single user system that means all applications can wreak as much havoc as they want. Its not an unfixable problem, there is SELinux and such that provides much more granular support of application rights, but that isn't something your average desktop distro uses and it would need quite a bit of work on the user interface side to integrate tightly locked applications smoothly (see OLPC for a nice example how it can be done).
You complain that your keyboard is broken in linux, and that's proof that linux is bad.
Its not a proof that Linux is bad, its just proof that it is imperfect and that the reason for that are not the evil hardware vendors, but just things that the open source crowd could fix by themselves.
Anywa, I've sucessfully build a single module (hardware driver) without a full kernel recompile,
That assumes that the feature you want to build is provided as a module in the first place and that the module doesn't require any changes outside the module itself which is often not the case. And even if you can build it as module, its not exactly a trivial single-click procedure.
Maybe you're saying that everything sucks....
Well, yeah, thats pretty much my point. Its simply that the Windows people have gotten used to their issues, so they happily ignore them, while the Linux issues are brand new and annoying for them. Linux simply has to be better then Windows if it wants to truly compete, not just almost as good.
That they are still problems even though they've been known for years - sometimes decades shows that they will never be addressed, or fixed.
I wouldn't be that pessimistic. The trouble with many of those problems is that they are in the infrastructure, not in a single application. So you can't just sit down, editing some code and be done. You have to write specs, write code for the specs and then convince all the other developers that your way of solving the problem is actually the right one and once that done, wait a year or two till everybody adopted their apps and then another few years till all of that is actually properly integrated into the distro. All that simply doesn't happen overnight, but takes a lot of time. And while things aren't improving quickly, they actually do improve. My graphic tablet went from "patch and compile a new kernel" over to "edit XF86Config" then it was "plug&play, but only some features support" to now being completly plug&play without touching any config file ever. Still not perfect, as their isn't a GUI tool to configure some aspects of the thing, but already quite close.
Agreed, but on the other side people have gotten used to all those quirks in Windows, on Linux those quirks are all new and fresh and completly annoying again. So its more a psychological problem then a technical one, but a problem non the less, as Linux has to be not just as good as Windows, but actually better to encourage people to do the switch.
That said, Windows is still far better in allowing things to be done via GUI then Linux. Most famous troublemaker in Linux is Xorg, which still lacks a proper standard and fully functioning GUI configuration tool.
The driver issue is not a Linux problem, it's problem of hw manufacturers who produce _closed_ source drivers only for MS.
Thats an old myth, the reality however is that a lot of the problems Linux has are completly on the Linux side. For example I can't use my Microsoft keyboard fully in Linux (a few multimedia buttons and the zoom-slider don't work), not because the specs aren't open, they have been known for years, but because the support for it wasn't properly integrated into the kernel. A few month ago some support actually made it into the kernel and more stuff works, but still not all, because X11 has some limitation in the protocol that says you must only have 256 buttons, to bad that keyboard needs more to function, can be worked around by patching the kernel some more and moving some ids around, but again, thats not exactly what an end user would do.
And that isn't an isolated issue, it happens all the time. Sometimes its due to old software cruft that hasn't kept up with times, sometimes its just the nastiness of the monolithic kernel that requires a full complete recompile to get a tiny little thing to work (like the button on my external HD) and sometimes its just stuff that gets broken due to not enough testing (Ubuntu 8.10 broke pretty much all input devices that weren't a mouse or keyboard, luckily fixed half a year later with Ubuntu9).
Now thats not to say that Windows is any better, yesterday I learned that my MX500 isn't worthy enough to receive support from Logitech, so I don't have no proper drivers for it in Vista (and whats up with mouse drivers being a 50MB download?!) and it looks like that Microsoft hasn't bothered to support my SideWinder Prec2 either. Both of them still work via standard driver, but missing quite a few features. Linux on the other side doesn't have those stupid issues, as software is properly separated from the hardware and so the software for rebinding a button doesn't care if you are using a Logitech, Microsoft or whatever device.
And lets not forget that Linux neither has a standard package format nor a standard API or ABI, so a hardware manufacture really doesn't have a good chance to build a proper binary-only driver even if they want to.
Slashdot Mirror
Do you try to look for an explanation, whatever it may be, or do you sit on it?
You sit on it, because the evidence isn't enough to explain it. The point of an explanation is understanding the true cause of something, not filling unknown details with UFO nonsense. If you don't have enough evidence, you are simply out of luck, you just can't explain everything. We don't lock up random people in jails just because we can't find the real murderer either, if evidence isn't enough we simply "sit on it" too.
But they tend to focus on average Joe's UFO sightings rather than the well documented and really hard to explain ones.
You miss the point. If you want to prove the existence of alien spacecraft you don't have to find hard to explain cases, you have to find the opposite, well explained ones. You need clear pictures and well observed objects, not anomalies in the sky that you couldn't identify.
And my appeal to authority is justified in that you'd think that a highly trained elite pilot who's flown for years and even been to space would know what he's looking at when he looks in the sky around him.
The human visual system is build to measure distances and velocities at small scales. When you fly high up in the air you have the exact opposite, huge velocities, great distances and worst of all nothing to use as reference. No amount of experience will help you when you encounter an unexpected thing in the sky, as it becomes nearly impossible to judge size or velocity just by sight.
What are you gonna do with those?
Sort them into the "unexplained" folder and move on. Just because you can't explain something doesn't mean it was an alien spacecraft. That case, as all the others, doesn't give you hard evidence for anything. Do we have now a clear picture of a space craft? Nope. Any idea how its propulsion system works? Nope. Who piloted it? Nope. Any idea about anything at all? Not really.
If you want to demonstrate that alien spacecrafts are real, you have to come up with some good evidence, not just an single unexplained anomaly. Find multiple anomalies that follow the same pattern and you might be getting somewhere, but a single one off doesn't really help you much with anything, especially not when you fill in the lack of facts with random UFO fiction.
When it comes to weird things happening in the air I like the story of British Airways Flight 9, full of mystery and suspense and it also happens to be fully explained in the end.
Seriously, guys who actually went and walked on the Moon
Does that give you magic power to detect alien spacecraft or immunity from mistaking something you see? Your appeal to authority just isn't a very good argument.
Now that's a bit off topic, but the real reason is that "sceptics" like you have settled on a frozen idea of what's possible and what's not,
No, they are skeptic because they haven't seen good evidence.
There's a plethora of UFO reports out there from civil and military pilots, as well as air traffic control staff, radar operators, military base personel, and yes, even astronauts who went to the Moon.
That is well and true, it however has a little problem: Those are UFOs in the literal sense, they are "unidentified flying object", not extra terrestrial spacecrafts. When you can't tell what something is, it simply means you don't know, it doesn't confirm that aliens are involved.
When it comes to hard evidence, there is simply nothing that points to E.T. Blobs of light in the sky can be lots of things, clouds, planets, satellites, floating lanterns, lense flare, insects and tons of more stuff. How many clear non-blurry pictures are there of alien space crafts? None. You'd guess in a time where every mobile has camera people would come up with some good pictures, but that hasn't happened.
NOTHING to do with performance!
That is only true when it comes to syntactic sugar, its not true when it comes to semantics. For example when I write:
I don't just tell the compiler what I want to do, but I also say how I want it to be done. The variable i for example has nothing to do with my problem and neither has the forced sequential evaluation. And there isn't an easy way to figure out if do_math() has side effects when it is hidden in some external library either. A compiler will have a much harder time figuring out my intend then in a language that allows me to express what I want to do, instead of how:
Here the compiler knows instantly that I don't care about the order of evaluation and it also knows that do_math doesn't have side effects because its a function, so it can split up the task across multiple CPUs without a problem.
Why isn't it used as often as say, Python, Perl, or Ruby?
You don't chose a language for their syntax, but for most part for their libraries and thus are lacking in Lua. Lua happens to be mostly intended as an language for embedding, not as a stand alone language.
If you worry about file corruption, I wouldn't touch XFS, that thing shredded files for me on every single unclean shutdown.
See, thats exactly the problem. If you invest everything you have into a single solution, you are kind of stuck with it, as you end up being to busy fixing the problems of your current solution to look for alternative solutions. Polywell presents an alternative and tries to tackle the problem from a completly different angle. Nobody knows if it would work out, but if you believe Bussards google talk it would cost a tiny fraction of ITERs cost to build a full scale Polywell reactor to find it out and it wouldn't take 15 years either. ITER has the problem of being way to expensive and way to unsuccessful so far, if you would pump similar amounts of money into alternative solution, you might already have found one.
but it's pretty trivial to put a system together in an evening...assuming you know what you're doing.
And the "knowing what you are doing" is exactly the problem. I don't buy many PCs, because there simply isn't a need for an upgrade every two years, instead I am more in the 5+ year range. So each time I look around for new stuff pretty much everything has changed. Almost all knowledge of previous generations is close to worthless, as power supplies, cables, cards, cooling, cases and so on all have changed. The screws holding things together are still the same, but thats about it.
He in no sense answers the question of why evolution couldn't have provided precisely the capacity he attributes to consciousness without any consciousness involved.
Just look at the last few decades of AI research, which has nicely shown that hard coded computation just doesn't get you very far, because its far to inflexible to react to unexpected events. Its good enough for winning chess, but just not very good to interact in the real world, which simply doesn't follow clear cut rules.
Planning != consciousness
It might not be the same, but its certainly something very similar, as consciousness is in large part about forming a model from the input that can then be used to plan and predict actions without actually acting them out in the real world. Try to build a robot that will act intelligently in the real world and you have to build up some internal model as well and sooner or later you might up with something that could be considered consciousness.
Or to put it another way: We really don't have any reason to assume that consciousness is something magical.
They seem to filter out or forget all off the crappy ones
Well, yeah, because there where enough good ones so that we didn't have to play the crappy ones. Look at the adventure genre, LucasArts had a new title ready every year and each of those was awesome. How often do high quality adventures released today? Not much at all, most are rather mediocre. And that is ignoring all the non-LucasArts adventure games from back in the day, some of them where pretty awesome too. And now don't let me start on the flightsim genre, as that is pretty much dead and burried today and the amount of good flightsim is close to zero.
that beats the crap out of joystick controls in terms of response.
Not really. Have you ever tried to hit anything on that setting without lock-on? Its near impossible because you constantly oversteer. Its not a problem for MP3 since you lock on all the time, but as a general FPS mechanics it doesn't really work. There is also the problem that you will get into a crazy spin each and every time you have a dialog with an NPC, as pointing to the center of the screen is not easy when you don't have any feedback from the interface.
So as a general dual-analogstick replacement I consider the Wiimote pretty much a failure. Aiming works very well, but rotation via the edge of the screen really doesn't. Thus one thing I would like to see is a game that makes only use of the aiming part and leaves all the rotation to the analogstick (i.e. classic Doom/MetroidPrime1 control, but you can aim at enemies that aren't in the center of the screen).
The question is, what market share is required to achieve this?
I doubt that it has much to do with market share, it seems more to be an issue with Linux being pretty much incompatible with how hardware manufacturers like to ship their drivers. Most drivers in the Windows world are not just drivers, they come bundled with a whole bunch of software and stuff that is tied to the specific piece of hardware (i.e. standard Windows Logitech mouse driver is 50MB instead of a few KB). A clean separation between the code that makes your hardware work on all that other additional software doesn't really exit, because the supplied software plays a big part in the marketing and feature lists you find on the box.
I think to get proper Linux support hardware vendors would first need to learn that their job is to produce hardware, not software. Once thats done they might have less problems with releasing specs, but I somewhat doubt that this is going to happen anytime soon because of Linux. The best thing for Linux hardware support in the end are really the open standards. Any USB HID or storage device works on Linux out of the box, not because the hardware vendor cared about Linux, but because he implemented the spec. The more specs we have for common hardware, the better the Linux support will be.
Thats hardly a problem in times of Bittorrent.
While I was growing up, playing video games pretty much meant you had an Atari, NES, or SNES.
Not where I grew up. C64, Amiga and all those other home computers have existed just as long and provides a ton of gaming goodness.
The gaming industry on the computer has grown by leaps and bounds.
The gaming industry has a whole has grown, and thus the PC market segment has grown too, in relative however the PC market segment has shrunk. For most developers doing PC-only games is no longer an option, most focus on multi-platform titles these days. And most of those titles are targeted for console first with the PC only getting a port much later and often not well adopted to the PC environment (i.e. GUI designed for game controller, not so good for use with a mouse).
Kind of, what PolicyKit does is give a user process a way to communicate to a root process, so that th user does not need root rights to accomplish certain things. That is fine for limiting access to root account, but it doesn't help with a user process having access to all the users files, as their is no way for a user to launch a process that has even less privileges then his user account has.
How old are you exactly? I'm not sure how anyone, who was old enough to know better, could claim that the state of PC video gaming was better in the 90's.
It was better. Back then PC games where very different from console games, you had your platformer and action adventures on the consoles and the PC was filled with adventures, flightsims, FPS and strategy games. Today on the other side most of those PC-exclusive genres are either dead or have been turned into console games. There no longer is a clear differentiation between PC and consoles, all get the same games, but the PC often gets them last, because it has the smallest market share of them (14%). Many once PC exclusive developers have either moved to consoles or at least gone multiplatform.
That doesn't mean PC gaming will die out, its the only open platform around, but it means PC gaming is shifting. The future of PC gaming lies in games like Peggle and indie titles not in AAA blockbuster titles, as those go where the money is and that is consoles.
The closest thing to an exclusive genre that the PC still has are the MMORPGs, but while they make tons of money, I am not so sure that they are actually good for PC gaming as a whole, as they suck up so much time, that other titles have a much harder time.
How exactly is this a problem in the operating system-
The problem is that mainstream Linux has just two ways to run an application: root or user. Both of those have the problem that they can with no problem delete or manipulate all the files a user has access to, for a single user system that means all applications can wreak as much havoc as they want. Its not an unfixable problem, there is SELinux and such that provides much more granular support of application rights, but that isn't something your average desktop distro uses and it would need quite a bit of work on the user interface side to integrate tightly locked applications smoothly (see OLPC for a nice example how it can be done).
You complain that your keyboard is broken in linux, and that's proof that linux is bad.
Its not a proof that Linux is bad, its just proof that it is imperfect and that the reason for that are not the evil hardware vendors, but just things that the open source crowd could fix by themselves.
Anywa, I've sucessfully build a single module (hardware driver) without a full kernel recompile,
That assumes that the feature you want to build is provided as a module in the first place and that the module doesn't require any changes outside the module itself which is often not the case. And even if you can build it as module, its not exactly a trivial single-click procedure.
Maybe you're saying that everything sucks....
Well, yeah, thats pretty much my point. Its simply that the Windows people have gotten used to their issues, so they happily ignore them, while the Linux issues are brand new and annoying for them. Linux simply has to be better then Windows if it wants to truly compete, not just almost as good.
Why not make linux games that are live CDs?
Because it would be a total waste of time. Nobody wants to reboot just to play a game, we had that with DOS and Bootdisks and it sucked a lot.
That they are still problems even though they've been known for years - sometimes decades shows that they will never be addressed, or fixed.
I wouldn't be that pessimistic. The trouble with many of those problems is that they are in the infrastructure, not in a single application. So you can't just sit down, editing some code and be done. You have to write specs, write code for the specs and then convince all the other developers that your way of solving the problem is actually the right one and once that done, wait a year or two till everybody adopted their apps and then another few years till all of that is actually properly integrated into the distro. All that simply doesn't happen overnight, but takes a lot of time. And while things aren't improving quickly, they actually do improve. My graphic tablet went from "patch and compile a new kernel" over to "edit XF86Config" then it was "plug&play, but only some features support" to now being completly plug&play without touching any config file ever. Still not perfect, as their isn't a GUI tool to configure some aspects of the thing, but already quite close.
Agreed, but on the other side people have gotten used to all those quirks in Windows, on Linux those quirks are all new and fresh and completly annoying again. So its more a psychological problem then a technical one, but a problem non the less, as Linux has to be not just as good as Windows, but actually better to encourage people to do the switch.
That said, Windows is still far better in allowing things to be done via GUI then Linux. Most famous troublemaker in Linux is Xorg, which still lacks a proper standard and fully functioning GUI configuration tool.
Come on, you had normal copy&paste in Linux for like what, a decade? You can still use middle click paste, but you don't have to if you don't like it.
The driver issue is not a Linux problem, it's problem of hw manufacturers who produce _closed_ source drivers only for MS.
Thats an old myth, the reality however is that a lot of the problems Linux has are completly on the Linux side. For example I can't use my Microsoft keyboard fully in Linux (a few multimedia buttons and the zoom-slider don't work), not because the specs aren't open, they have been known for years, but because the support for it wasn't properly integrated into the kernel. A few month ago some support actually made it into the kernel and more stuff works, but still not all, because X11 has some limitation in the protocol that says you must only have 256 buttons, to bad that keyboard needs more to function, can be worked around by patching the kernel some more and moving some ids around, but again, thats not exactly what an end user would do.
And that isn't an isolated issue, it happens all the time. Sometimes its due to old software cruft that hasn't kept up with times, sometimes its just the nastiness of the monolithic kernel that requires a full complete recompile to get a tiny little thing to work (like the button on my external HD) and sometimes its just stuff that gets broken due to not enough testing (Ubuntu 8.10 broke pretty much all input devices that weren't a mouse or keyboard, luckily fixed half a year later with Ubuntu9).
Now thats not to say that Windows is any better, yesterday I learned that my MX500 isn't worthy enough to receive support from Logitech, so I don't have no proper drivers for it in Vista (and whats up with mouse drivers being a 50MB download?!) and it looks like that Microsoft hasn't bothered to support my SideWinder Prec2 either. Both of them still work via standard driver, but missing quite a few features. Linux on the other side doesn't have those stupid issues, as software is properly separated from the hardware and so the software for rebinding a button doesn't care if you are using a Logitech, Microsoft or whatever device.
And lets not forget that Linux neither has a standard package format nor a standard API or ABI, so a hardware manufacture really doesn't have a good chance to build a proper binary-only driver even if they want to.