I assume that drivers can, but it's more about the efficiency of the applications running on it, and what they tell the OS to redraw and so on. Some of the best non-ultra portable laptops for battery life are Apple laptops.
My parents get upwards of 5:30 hours on their MacBook under normal use on OS X with "Better Energy Savings" selected. On a similar Dell with XP, it's only about 2 hours. If you put XP on my parent's laptop it gets about 2 hours or so. I always assumed that the reason this would be the case is that OS X is composited, which means that the reasonably CPU/Video intensive operation of recalculating and redrawing interface widgets isn't necessary when moving or minimizing or whatever, and that generally the UI is more efficient because it's based in video.
I know for a fact, that writing OpenGL interfaces need not be GPU/CPU intensive (since I am actually a programmer). Quite often, there is a significant increase in performance and reduction in power consumption when using a GL interface because the GPU consumes less power doing the same thing that a CPU could do, since it is more specialised and does it so damn quickly. I had, therefore, assumed that Vista would have better battery life than XP, particularly with all the more recent advances in power consumption reduction and research into how to make systems more efficient.
Maybe OpenGL are more concerned with efficiency in general, DX is a pretty huge API and there doesn't seem to be much of an efficiency/low power computing push in DirectX. At least not on the same level as OpenGL ES and so on: http://en.wikipedia.org/wiki/OpenGL_ES
Yeah, I thought about that, but it would seem that the "EU is good" is almost nonexistent but the "MS is bad" is very widespread, among comments and modders.
Great comeback, but in all seriousness, O(1) translates to constant time for a varying number of processes. O(n) increases depending on n in a linear fashion. So on, etc.
Why is it that the overall consensus on Slashdot is that the US government aren't doing enough to regulate Microsoft or help stop the monopoly, but as soon as the EU government try to get Microsoft to publish their specifications for interoperability, there's a huge outcry?
I think that, Slashdot being US based, seems to randomly hate the EU and their decisions. Quite honestly, it doesn't matter what Microsoft want here, and the EU probably don't care what economic damage their going to do to a US based company. Bear in mind that for Microsoft to release all their protocol specifications, it'd probably cost an extremely small amount of money compared to the amount they wasted* on Vista.
Considering Microsoft is a convicted monopolist and to interoperate with over 90% of the market, you need access to Microsoft's specifications, I hardly think it's unreasonable to demand MS provide access (even free access) to the specifications. I don't think Microsoft has a reason to demand that they maintain an illegal monopoly by not releasing their specifications that would merely allow others to interoperate. If anyone's dragging their feet here, it's Microsoft. I doubt a multi billion dollar company couldn't have just spent maybe a few weeks writing specifications and then released them on their devnet site. Hell, they could have bought a specifications firm to do it for them... It really isn't as difficult as they're making it out to be. The fact that their original specifications were deemed to be too low quality speaks volumes. I wonder how many of their file formats are internally undocumented other than the library itself.
And hey, if MS hates it, they can pull out of the EU. The EU could nullify the Windows copyright in the EU and we could all legally pirate Windows:).. Okay, unlikely.. I sort of hope not, really.
* I don't mean all of Vista was a waste, but there was probably a considerable amount of money wasted on things like WinFS that have been all but dropped now
O(1) just means that it doesn't lower in performance for the more processes that are running. In other words, scalable. This is good, and the current kernel does have modes for interactivity and server workloads, they're just not as good as this new scheduler. As I understand it, the scheduler in Linux has been better than the one in Windows for a good while now.
It's hard to port from the 360 to the PS3. It's easy to port from the PS3 to the 360. Actually, it's more the other way around. The PS3 can run run of the mill style, single threaded PC sort of games, but due to the architecture of the PS3, they won't run as quickly as they could if they were made specifically for the PS3. That's usually not too much of a problem but it's why ports from the 360 to the PS3 tend to not be as high framerates. The 360 is closer to the PC style of development, whereas the PS3 is pretty strange in comparison. Porting from the PS3 to the 360 will be far more difficult than the other way around, since there aren't SPEs and PPEs etc. in a 360.
That means more 360 exclusives, less PS3 exclusives. I think I agree with the conclusion though, despite not agreeing with the premise. I think that it's because the PS3 is, at the moment, an alien style of development compared to the tried and tested development style on a 360. Development for a 360 will be easier and faster in the mean time, and there's more possibility of porting to other systems when the exclusive contract expires. The PS2 was similar, there weren't as many PS2 exclusives that ended up on the PC as there were 360 exclusives that did.
Then Verizon could create a small company to retain ownership of the patents and merely license the use of them to Verizon. The sub-Verizon would be smaller than you and could therefore sue you.
I don't think the iPhone is important because it runs OS X. I think OS X must be important to Apple since they've put it on the iPhone. I never said that Apple would be able to deliver a decent mobile phone, just that the market may be screaming for one. Despite your assertion that Nokia and Sony Ericsson make excellent mobile phones, I have found them to be pretty bad interface wise, even the ones that run my "beloved" Linux. Most phones now seem to have internet browsing, a camera and all that crap, so would seem to be a sort of PDA-phone.
Maybe Apple can pull this off, maybe they can't.... whatever, that's irrelevant to me anyway at the moment since I live in the UK and there is no UK release planned yet. I would much rather buy a 3G capable phone that can connect to my laptop with bluetooth and function as a modem, and obviously be good at making phone calls. I have an iPod so that my phone doesn't need MP3 functionality.
You may think I'm an Apple fanatic, but really I was trying to debate in a logical way. I actually prefer Linux, and wouldn't dream of buying the iPhone. If my argument contained logical flaws, by all means point them out, but all you've done here is attack a straw man by saying I said things I never said. Note how I never claimed the iPhone would be a success; I even said "potential success" a bit later to show that I wasn't being definite that the iPhone would be successful.
My point stands. I think the market is screaming for a decent cellphone. Whether Apple can deliver it or not and in what timeframe is another question; Anyone remember 1G iPods?
What's up with these binary comparisons? Just because OS X 10.5 was delayed a few months for the iPhone doesn't mean anything to do with OS X being unimportant. The iPhone runs OS X; it must be important.
Most users are happy with 10.4 and 10.5 is more of a luxury than a necessity. All this means is that 10.4 is sufficient that the general Apple buyer isn't screaming for OS improvements, but that the market may indeed be screaming for a decent mobile phone, like they were screaming for a decent MP3 player around when the iPod gained in popularity.
Anyway, a lot of the funds and improvements from the potential success from the iPhone will probably be funneled back into OS X and the Mac hardware. Haven't some of the improvements in 10.5, like Core Animation, been brought about due to the iPhone already?
The text in the address bar and Firefox widgets is indeed not subpixel rendered, however, the web page text is. The comparison I was making was between the web page text (rendered by OS X) and the image on the page (rendered by Microsoft's ClearType). The web page text is rendered by my system using subpixel rendering and is what I am referring to, since most text on my system is rendered that way. Firefox's GUI widgets are probably custom rendered or something, but Finder's widgets do indeed use subpixel rendering.
You have it right, I believe. Maybe it's something to do with the APIs for resizing images on the different systems. Different resampling algorithms can cause strange things like this.
I just looked at a screenshot of "Cleartype" on OS X (From http://www.microsoft.com/typography/ClearTypeFAQ.m spx). It looks revolting. I use Linux and OS X with subpixel rendering enabled. The first three lines look OK, but it starts to look blurry after that.
If you zoom in, you can see I'm definitely using subpixel rendering. I get the odd blurry looking font on my OS X system, but nowhere near the sort of stuff I see on other people's Windows laptops.
Actually, this is my point about diversity. I very nearly wrote almost what you wrote (as I run Gentoo also) as an example of diversity -- diverse configurations being similar to genetic diversity, but I was worried the anti-Linux/pro-Windows/pro-OSX/whatever trolls would mod me into the ground. ASLR works by randomising the program's memory space, which could be considered security through obscurity, though you could say that a password is security through obscurity, since it is secure based on the assumption that they don't know the password. Anyway, isn't Windows getting an ASLR technology in Vista or did I dream that?
Essentially though, I was saying the article was not describing security through obscurity, but rather just running a lesser used operating system as there is less script-kiddie interest.
This isn't security through obscurity. Security through obscurity would be saying "I'm safe because I run Windows and it's closed source". This is the claim that uncommon software is more secure because there are less exploits. While untrue mathematically, the reality is that you are still currently less likely to be exploited when running Mac OS X or Linux since script kiddies don't really care about you so much (for the same reason game developers don't, incidentally).
Same is true for biological systems - diversity is a good thing as it is less likely to be infected with a disease. Genetic diversity implies a more robust "operating system" species that's harder to destroy. Remember all the hell around the blaster worm. Imagine that MS, Apple, RedHat, Ubuntu... only had 10% marketshare each... it'd be bad, but not nearly as bad as it was.
If you're talking about a focussed professional attack on a specific system: to be honest, the OS you're running is probably pretty insignificant; the chances are there's a simple admin error somewhere along the line.
Why not get a tiny super-efficient PC, such as a Mini-ITX board or even try to get ahold of an ARM based system and put Linux or something... I'm almost positive that the latter if not the former will consume less power than your old P166.
The thing is, most of this work is, as far as I know, already done. The hard part is linking it all together in an easy to manage fashion. I love the opensource world's choice, but the need for collaboration is seen everywhere. Projects for unification exist here and there for things like desktops (Freedesktop, Tango), but as of yet I've not seen too many developer ones except for the plain old autoconf tools.
Looking around the OSS world there are some simply amazing projects that could be used for game development.... OpenGL/AL, SDL, OGRE 3D, Blender, Eclipse, ODE. ODE was used in BloodRayne2 and many other games I believe. And say a game was to choose to use Glib/Qt/GDK/GTK for operations like windowing, threads or whatever, it would be, in theory, compatible with Windows too. The beauty of most OSS libs is their cross platform. The Xbox consoles would seem to be the biggest barrier.. thanks Microsoft.
Oh, and also, why not the LGPL? I understand not using GPL libraries for game engines used in proprietary code, but an LGPL library would not require the game's code to be released in any shape or form.. only modified bits of the engine, if any.. no? I may have misunderstood, but I thought LGPL could be linked with proprietary code.
Component based programming for Linux turned out very different. Since Linux stuff tends to be developed from the bottom up, starting with a library, then a console app, then a GUI, most "components" ended up just being stuff lower on the stack, like libpng, gzip or whatever. That, and file formats tend to be standardised and read with a library, rather than file access being implemented in any specific application that needs to be loaded every time the file needs to be accessed.
On DOS and Windows, applications tended to be all-in-one blobs that use proprietary file formats and only depend on the operating system itself. Thus, OLE made sense. There was OLE-type technologies on Linux (such as Bonobo and KParts) but they never took off in any big way due to the general lack of need. In most cases, they were just used for automation, which better technologies such as DBus now exist.
As for "real code reuse" on Linux... try ldd-ing any application and seeing how many other libraries it uses to do it's bidding. Many application's I've written are just sticking together bits and pieces of other libraries, and that's in C++. I'd go as far as saying that code reuse on Linux is currently better than that in Windows. Generating quick applications with Linux tends to be pretty easy using bash, Python and even C, C++ or C#, where libraries have directly translated into "components" and "modules". The thing on Linux slowing down development isn't having to reinvent the wheel every time you write code, but rather, deciding what libraries you want to depend on and the difficulty of the languages favoured in Linux development, like C.
Since you bring it up, OLE seems to be taking a backseat in Windows now, where ActiveX is rarely used by anything other than IE plugins. Now clipboard/d'n'd/DDE and Microsoft Office are the only times I actively see it being used. COM is still heavily in use by libraries such as DirectX, however. It was never really used to it's full potential, such as to open Photoshop documents in Microsoft's preview application. Windows, with.NET, seems to be more of a developer friendly way of providing and using library components as a developer does in Linux..NET components remind me of how libraries are used in Linux. IMO this is a cleaner way of working and it's nice that Microsoft are starting to think this way.
When I installed the official Firefox 1.5 binary on Windows and OS X for the first time, it wouldn't perform any searches until I selected a search engine with the drop down...
The thing about expose that's good is that you see the whole of every window at the same time.
For quite a lot of websites such as the one you are reading, the header and left of the screen (in this case the menu bar) are identical. With my experience of Flip3d, it is impossible to differentiate between windows like this graphically, so you need to scroll through them.. It's quite a common use case to have those parts of the window being similar.
It might just be how I work, but I tend to remember the shape of the text/buttons/whatever on the screen of a particular window, or the graphics. The 'details' such as title bars and text tend not to be so important as the general look of the window. Flip3d is good for seeing the details but it's harder to get the general impression of the window, so it (in my case) requires more concentration to use.
For windows I don't plan on using for a little while (e.g. to work on a sub-task), i'll command+h the application or minimise the window to hide them from expose. In my case, the time I last used the window doesn't bear too much relation on when I'm next going to use it if it's still visible, since it means it's involved with the current task or that I am checking it occasionally (such as a contact list).
The per-application expose and the desktop expose are also extremely useful, especially given the app/document management on OS X. I'd love it if Windows copied expose... it's difficult for me to work without it now -- it's habit for me to use the top-right corner hotspot for expose. I know there are some implementations for Windows, but when forced to use it elsewhere, such as at corporate desktops, you can't install stuff.
EVERYTHING else on the market has an interface specifically designed for it. Only computer systems use the keyboard/mouse interface for every application.
Except for those applications which work with game controllers, graphics tablets, MIDI keyboards, etc.
Have you used a modern GNOME or KDE desktop? They're very usable.
At least with GNOME, many new computer users I've tried find it easier than Windows. Naturally people who are used to Windows have some switching difficulties though (button order, menu names etc.). There are always improvements that could be made, but they are very nice to use (at least the core apps). Also, I am a programmer and use GNOME, and I've never had issues with not being able to do something. If there's no GUI for some advanced task, there's the command line, but usually those sort of tasks are ones which normal users won't be doing.
Personally, I find many of the cheaper proprietary apps (~$15 etc.) are clunky and badly designed from a usability aspect, and much opensource software is better. The reason for this is that very cheap proprietary apps (even ones that are popular) usually have only 1 or 2 developers, and their skillset is usually just coding and whatever the application is used for (graphics, sound, whatever). With popular opensource and large proprietary apps, there are often people contributing to the code who are specialised in usability.
Much of Windows' usability issues probably come from the marketing aspect (lets make hundreds of things appear at once so it seems powerful!) and backwards human compatibility. Microsoft get burned every time they change a string on their interface, unfortunately this results in strange UI conventions that exist only because developers and users are used to them, and every release the interface becomes further disjointed from the backend which can be a pain from a developer/power user point of view. Vista seems to be cleaning some of this up, however.
Slashdot Mirror
I assume that drivers can, but it's more about the efficiency of the applications running on it, and what they tell the OS to redraw and so on. Some of the best non-ultra portable laptops for battery life are Apple laptops.
My parents get upwards of 5:30 hours on their MacBook under normal use on OS X with "Better Energy Savings" selected. On a similar Dell with XP, it's only about 2 hours. If you put XP on my parent's laptop it gets about 2 hours or so. I always assumed that the reason this would be the case is that OS X is composited, which means that the reasonably CPU/Video intensive operation of recalculating and redrawing interface widgets isn't necessary when moving or minimizing or whatever, and that generally the UI is more efficient because it's based in video.
I know for a fact, that writing OpenGL interfaces need not be GPU/CPU intensive (since I am actually a programmer). Quite often, there is a significant increase in performance and reduction in power consumption when using a GL interface because the GPU consumes less power doing the same thing that a CPU could do, since it is more specialised and does it so damn quickly. I had, therefore, assumed that Vista would have better battery life than XP, particularly with all the more recent advances in power consumption reduction and research into how to make systems more efficient.
Maybe OpenGL are more concerned with efficiency in general, DX is a pretty huge API and there doesn't seem to be much of an efficiency/low power computing push in DirectX. At least not on the same level as OpenGL ES and so on: http://en.wikipedia.org/wiki/OpenGL_ES
Yeah, I thought about that, but it would seem that the "EU is good" is almost nonexistent but the "MS is bad" is very widespread, among comments and modders.
Great comeback, but in all seriousness, O(1) translates to constant time for a varying number of processes. O(n) increases depending on n in a linear fashion. So on, etc.
Why is it that the overall consensus on Slashdot is that the US government aren't doing enough to regulate Microsoft or help stop the monopoly, but as soon as the EU government try to get Microsoft to publish their specifications for interoperability, there's a huge outcry?
:) .. Okay, unlikely.. I sort of hope not, really.
I think that, Slashdot being US based, seems to randomly hate the EU and their decisions. Quite honestly, it doesn't matter what Microsoft want here, and the EU probably don't care what economic damage their going to do to a US based company. Bear in mind that for Microsoft to release all their protocol specifications, it'd probably cost an extremely small amount of money compared to the amount they wasted* on Vista.
Considering Microsoft is a convicted monopolist and to interoperate with over 90% of the market, you need access to Microsoft's specifications, I hardly think it's unreasonable to demand MS provide access (even free access) to the specifications. I don't think Microsoft has a reason to demand that they maintain an illegal monopoly by not releasing their specifications that would merely allow others to interoperate. If anyone's dragging their feet here, it's Microsoft. I doubt a multi billion dollar company couldn't have just spent maybe a few weeks writing specifications and then released them on their devnet site. Hell, they could have bought a specifications firm to do it for them... It really isn't as difficult as they're making it out to be. The fact that their original specifications were deemed to be too low quality speaks volumes. I wonder how many of their file formats are internally undocumented other than the library itself.
And hey, if MS hates it, they can pull out of the EU. The EU could nullify the Windows copyright in the EU and we could all legally pirate Windows
* I don't mean all of Vista was a waste, but there was probably a considerable amount of money wasted on things like WinFS that have been all but dropped now
O(1) just means that it doesn't lower in performance for the more processes that are running. In other words, scalable. This is good, and the current kernel does have modes for interactivity and server workloads, they're just not as good as this new scheduler. As I understand it, the scheduler in Linux has been better than the one in Windows for a good while now.
Then Verizon could create a small company to retain ownership of the patents and merely license the use of them to Verizon. The sub-Verizon would be smaller than you and could therefore sue you.
Great misquoting.
I don't think the iPhone is important because it runs OS X. I think OS X must be important to Apple since they've put it on the iPhone. I never said that Apple would be able to deliver a decent mobile phone, just that the market may be screaming for one. Despite your assertion that Nokia and Sony Ericsson make excellent mobile phones, I have found them to be pretty bad interface wise, even the ones that run my "beloved" Linux. Most phones now seem to have internet browsing, a camera and all that crap, so would seem to be a sort of PDA-phone.
Maybe Apple can pull this off, maybe they can't.... whatever, that's irrelevant to me anyway at the moment since I live in the UK and there is no UK release planned yet. I would much rather buy a 3G capable phone that can connect to my laptop with bluetooth and function as a modem, and obviously be good at making phone calls. I have an iPod so that my phone doesn't need MP3 functionality.
You may think I'm an Apple fanatic, but really I was trying to debate in a logical way. I actually prefer Linux, and wouldn't dream of buying the iPhone. If my argument contained logical flaws, by all means point them out, but all you've done here is attack a straw man by saying I said things I never said. Note how I never claimed the iPhone would be a success; I even said "potential success" a bit later to show that I wasn't being definite that the iPhone would be successful.
My point stands. I think the market is screaming for a decent cellphone. Whether Apple can deliver it or not and in what timeframe is another question; Anyone remember 1G iPods?
What's up with these binary comparisons? Just because OS X 10.5 was delayed a few months for the iPhone doesn't mean anything to do with OS X being unimportant. The iPhone runs OS X; it must be important.
Most users are happy with 10.4 and 10.5 is more of a luxury than a necessity. All this means is that 10.4 is sufficient that the general Apple buyer isn't screaming for OS improvements, but that the market may indeed be screaming for a decent mobile phone, like they were screaming for a decent MP3 player around when the iPod gained in popularity.
Anyway, a lot of the funds and improvements from the potential success from the iPhone will probably be funneled back into OS X and the Mac hardware. Haven't some of the improvements in 10.5, like Core Animation, been brought about due to the iPhone already?
The text in the address bar and Firefox widgets is indeed not subpixel rendered, however, the web page text is. The comparison I was making was between the web page text (rendered by OS X) and the image on the page (rendered by Microsoft's ClearType). The web page text is rendered by my system using subpixel rendering and is what I am referring to, since most text on my system is rendered that way. Firefox's GUI widgets are probably custom rendered or something, but Finder's widgets do indeed use subpixel rendering.
Try again. It's less coloured than the Microsoft font, but it is definitely coloured, as eyedropper in Photoshop reveals.
You have it right, I believe. Maybe it's something to do with the APIs for resizing images on the different systems. Different resampling algorithms can cause strange things like this.
I just looked at a screenshot of "Cleartype" on OS X (From http://www.microsoft.com/typography/ClearTypeFAQ.m spx). It looks revolting. I use Linux and OS X with subpixel rendering enabled. The first three lines look OK, but it starts to look blurry after that.
r p7.png
Screenshot of my OS X system: http://img248.imageshack.us/my.php?image=picture3
If you zoom in, you can see I'm definitely using subpixel rendering. I get the odd blurry looking font on my OS X system, but nowhere near the sort of stuff I see on other people's Windows laptops.
Actually, this is my point about diversity. I very nearly wrote almost what you wrote (as I run Gentoo also) as an example of diversity -- diverse configurations being similar to genetic diversity, but I was worried the anti-Linux/pro-Windows/pro-OSX/whatever trolls would mod me into the ground. ASLR works by randomising the program's memory space, which could be considered security through obscurity, though you could say that a password is security through obscurity, since it is secure based on the assumption that they don't know the password. Anyway, isn't Windows getting an ASLR technology in Vista or did I dream that?
Essentially though, I was saying the article was not describing security through obscurity, but rather just running a lesser used operating system as there is less script-kiddie interest.
This isn't security through obscurity. Security through obscurity would be saying "I'm safe because I run Windows and it's closed source". This is the claim that uncommon software is more secure because there are less exploits. While untrue mathematically, the reality is that you are still currently less likely to be exploited when running Mac OS X or Linux since script kiddies don't really care about you so much (for the same reason game developers don't, incidentally).
Same is true for biological systems - diversity is a good thing as it is less likely to be infected with a disease. Genetic diversity implies a more robust "operating system" species that's harder to destroy. Remember all the hell around the blaster worm. Imagine that MS, Apple, RedHat, Ubuntu... only had 10% marketshare each... it'd be bad, but not nearly as bad as it was.
If you're talking about a focussed professional attack on a specific system: to be honest, the OS you're running is probably pretty insignificant; the chances are there's a simple admin error somewhere along the line.
Why not get a tiny super-efficient PC, such as a Mini-ITX board or even try to get ahold of an ARM based system and put Linux or something... I'm almost positive that the latter if not the former will consume less power than your old P166.
The thing is, most of this work is, as far as I know, already done. The hard part is linking it all together in an easy to manage fashion. I love the opensource world's choice, but the need for collaboration is seen everywhere. Projects for unification exist here and there for things like desktops (Freedesktop, Tango), but as of yet I've not seen too many developer ones except for the plain old autoconf tools.
Looking around the OSS world there are some simply amazing projects that could be used for game development.... OpenGL/AL, SDL, OGRE 3D, Blender, Eclipse, ODE. ODE was used in BloodRayne2 and many other games I believe. And say a game was to choose to use Glib/Qt/GDK/GTK for operations like windowing, threads or whatever, it would be, in theory, compatible with Windows too. The beauty of most OSS libs is their cross platform. The Xbox consoles would seem to be the biggest barrier.. thanks Microsoft.
Oh, and also, why not the LGPL? I understand not using GPL libraries for game engines used in proprietary code, but an LGPL library would not require the game's code to be released in any shape or form.. only modified bits of the engine, if any.. no? I may have misunderstood, but I thought LGPL could be linked with proprietary code.
Component based programming for Linux turned out very different. Since Linux stuff tends to be developed from the bottom up, starting with a library, then a console app, then a GUI, most "components" ended up just being stuff lower on the stack, like libpng, gzip or whatever. That, and file formats tend to be standardised and read with a library, rather than file access being implemented in any specific application that needs to be loaded every time the file needs to be accessed.
.NET, seems to be more of a developer friendly way of providing and using library components as a developer does in Linux. .NET components remind me of how libraries are used in Linux. IMO this is a cleaner way of working and it's nice that Microsoft are starting to think this way.
On DOS and Windows, applications tended to be all-in-one blobs that use proprietary file formats and only depend on the operating system itself. Thus, OLE made sense. There was OLE-type technologies on Linux (such as Bonobo and KParts) but they never took off in any big way due to the general lack of need. In most cases, they were just used for automation, which better technologies such as DBus now exist.
As for "real code reuse" on Linux... try ldd-ing any application and seeing how many other libraries it uses to do it's bidding. Many application's I've written are just sticking together bits and pieces of other libraries, and that's in C++. I'd go as far as saying that code reuse on Linux is currently better than that in Windows. Generating quick applications with Linux tends to be pretty easy using bash, Python and even C, C++ or C#, where libraries have directly translated into "components" and "modules". The thing on Linux slowing down development isn't having to reinvent the wheel every time you write code, but rather, deciding what libraries you want to depend on and the difficulty of the languages favoured in Linux development, like C.
Since you bring it up, OLE seems to be taking a backseat in Windows now, where ActiveX is rarely used by anything other than IE plugins. Now clipboard/d'n'd/DDE and Microsoft Office are the only times I actively see it being used. COM is still heavily in use by libraries such as DirectX, however. It was never really used to it's full potential, such as to open Photoshop documents in Microsoft's preview application. Windows, with
Yeah. It caught me out at first... I thought search was just broken since it'd defaulted to Google on my previous installations.
When I installed the official Firefox 1.5 binary on Windows and OS X for the first time, it wouldn't perform any searches until I selected a search engine with the drop down...
The thing about expose that's good is that you see the whole of every window at the same time.
For quite a lot of websites such as the one you are reading, the header and left of the screen (in this case the menu bar) are identical. With my experience of Flip3d, it is impossible to differentiate between windows like this graphically, so you need to scroll through them.. It's quite a common use case to have those parts of the window being similar.
It might just be how I work, but I tend to remember the shape of the text/buttons/whatever on the screen of a particular window, or the graphics. The 'details' such as title bars and text tend not to be so important as the general look of the window. Flip3d is good for seeing the details but it's harder to get the general impression of the window, so it (in my case) requires more concentration to use.
For windows I don't plan on using for a little while (e.g. to work on a sub-task), i'll command+h the application or minimise the window to hide them from expose. In my case, the time I last used the window doesn't bear too much relation on when I'm next going to use it if it's still visible, since it means it's involved with the current task or that I am checking it occasionally (such as a contact list).
The per-application expose and the desktop expose are also extremely useful, especially given the app/document management on OS X. I'd love it if Windows copied expose... it's difficult for me to work without it now -- it's habit for me to use the top-right corner hotspot for expose. I know there are some implementations for Windows, but when forced to use it elsewhere, such as at corporate desktops, you can't install stuff.
EVERYTHING else on the market has an interface specifically designed for it. Only computer systems use the keyboard/mouse interface for every application.
Except for those applications which work with game controllers, graphics tablets, MIDI keyboards, etc.
A browser crash is not the website's fault, therefore, IE7 or Windows (or something else linked into IE7) was definitely at fault here.
Have you used a modern GNOME or KDE desktop? They're very usable.
At least with GNOME, many new computer users I've tried find it easier than Windows. Naturally people who are used to Windows have some switching difficulties though (button order, menu names etc.). There are always improvements that could be made, but they are very nice to use (at least the core apps). Also, I am a programmer and use GNOME, and I've never had issues with not being able to do something. If there's no GUI for some advanced task, there's the command line, but usually those sort of tasks are ones which normal users won't be doing.
Personally, I find many of the cheaper proprietary apps (~$15 etc.) are clunky and badly designed from a usability aspect, and much opensource software is better. The reason for this is that very cheap proprietary apps (even ones that are popular) usually have only 1 or 2 developers, and their skillset is usually just coding and whatever the application is used for (graphics, sound, whatever). With popular opensource and large proprietary apps, there are often people contributing to the code who are specialised in usability.
Much of Windows' usability issues probably come from the marketing aspect (lets make hundreds of things appear at once so it seems powerful!) and backwards human compatibility. Microsoft get burned every time they change a string on their interface, unfortunately this results in strange UI conventions that exist only because developers and users are used to them, and every release the interface becomes further disjointed from the backend which can be a pain from a developer/power user point of view. Vista seems to be cleaning some of this up, however.