My responses (because his premise is flawed; that is, I do have a tablet):
1. Tablets are devices which fit niches I occupy, such as being a university student (fantastic for taking notes or annotating the professor's slides during lecture). Lots of people occupy such niches. 2. Remarkably enough, some of those who complained that the iPad doesn't have a full OS have tablets with full OSes, and love them. My tablet is also a highly portable laptop, and that is a *good* thing. 3. Depends on your definition of "high end" here; my tablet is 1.2 GHz Core 2 Duo, with 4GB of RAM. That's a lot higher than the iPad and a lot lower than the computer I type this on now. An ARM chip would give better battery life but would limit the computer's usefulness. A 2.8 GHz processor and nVidia GPU would give amazing performance but unless the battery got a lot bigger it wouldn't last me through a single lecture period. In other words, this is a good balance - it's fast enough for everything from taking notes to writing code to light gaming (nothing graphics intensive), yet the battery lasts 5 hours active (over a week in sleep) which is enough I typically only charge it every other day. 4. Interface is definitely important. On a tablet, the Win7 taskbar is larger so it's easier to hit with a stylus or finger. There are various gestures and pen flicks that the OS recognizes, that do things like "Go back" or "Cut"/"Copy"/"Paste" I'll grant you that a dedicated interface might be better for some things, but the Win7 one works surprisingly well after all the little tablet-related features are enabled (which they are by default on tablet computers, and aren't by default on non-tablets). 5. OneNote. Seriously, Microsoft Office OneNote is a damn good reason to get a tablet. It works on normal hardware, but it doesn't really shine until it's on a tablet. There are other programs that are great with tablets (including a lot of design or artistic stuff), but OneNote is the main one for me.
The IE9 preview, early though it is, blows Firefox et. al completely out of the water. Try running the number of images up to the max (256), zooming in, and holding the Shift key while moving your mouse. On my machine, IE9 preview will still get 28 FPS. Opera 10.5 will come next closest, at about 9.5 but with terrible background flicker. Firefox 3.6 gets about 5.5 FPS, not smooth at all.
This is true, but also irrelevant. We're talking about the possibility of a specific performance optimization here. Obviously, software rendering *can* do 3D - this has been known for decades, and is what other browsers use if you go to a 3D website like http://ie.microsoft.com/testdrive/Performance/01FlyingImages/Default.html . The fact that it's possible to do the same thing does not mean that it's possible to do it with the same performance though, which is the crux of the matter. WDDM drivers provide a lot of capabilities that non-WDDM drivers lack. WDDM drivers require kernel driver interfaces that XP lacks (if you added support for them to XP, it wouldn't really be XP anymore; this is a non-trivial modification of an old and long-stable portion of the kernel). The way IE9 gets the performance it does on 3D relies upon these capabilities. No Vista/Win7 = No WDDM = No support for those features = No IE9 hardware accelerated performance. There might be another way to get equivalent performance, but neither MS nor anybody else has demonstrated it yet.
I'm not sure what you think is "funny" about that, but maybe you just have an odd sense of humor. Actually, I suppose if you were the sort to bash anything that MS does which looks like a mistake, without any idea of what was involved, yeah it might be funny.
IE8: The default 36 images get nominally 4 FPS, but it doesn't even finish drawing the old frame before starting the new one so it tears abominably. At 4 images it gets about 30 PFS, smoothly. The follow-the-mouse function doesn't work.
Chrome 4.1: At 36 images, only 2 FPS, though no tearing. At 4 images, 30 FPS. Animation is still smooth, but framerate drops terribly if you zoom in oh hold the Shift key for faster rotation.
Firefox 3.6: 20 FPS at 36 images, and very clearly not smooth if you hold the shift key (fast spinning). 60 FPS and smooth at 16 images, though it drops to 12 FPS if you zoom in.
Opera 10.5: At 36 images, 25 FPS and smooth. Faster rotation is fine, but zooming in gives a flickering background and only 20 FPS. It'll do 100 images (at 12 FPS) but the background flickers fit to give you a seizure.
IE9 preview: At 36 images, 60 FPS and smooth, even zoomed in or with Shift. No change at 64 images. 100 images causes a drop to 57 FPS. I can go up to 256 images before the framerate drops to 30, and it's still smooth even with zoom and Shift.
I don't have Safari or Konqueror installed here, but I think you get the idea. IE8 can't even execute the page's code right. Chrome crawls at the 3D effect. Firefox is OK at wide angle and crawls when zoomed. Opera is the fastest of the released browsers, but has horrible flicker when images pass the edge of the screen. None of these browsers use hardware accelerated drawing.
The very, very early preview of IE9, which does use hardware acceleration, blows them all away. No performance degradation until it reaches the point where most other browsers drop to the single digits. No trouble with zooming or fast motion. No flicker or tearing. If you'd told me I was watching a demo of a 3D engine written in C++, I'd have believed you (not been terribly impressed, but believed you). For something using pure JavaScript I'm amazed.
My system has a mid-range GPU (GeForce 9600M) but pretty good CPU (Core 2 Duo T9600, 2.8GHz), running Win7 x64. I'm guessing IE9 uses vertical sync, since it maxes at my refresh rate (60 Hz). Clearly, simply compiling the JS to native binary isn't enough to get the really impressive performance, since the other browsers do that. Since Vista/Win7 use 3D to render the desktop anyhow, I can certainly believe it's easier to incorporate this kind of functionality into those operating systems. It may be possible with XP, but so far there's no indication that you can get comparable performance - none of the browsers that will run on XP can, at least.
As a side note, it's worth pointing out that openSuse is to Suse Linux Enterprise (commercial) as Fedora is to Red Hat - same codebase, but openSuse is typically running newer versions, is non-commercial (including for updates), and the base install does not include things like Flashplayer (although it includes an option to install them immediately after the system boots up for the first time, FlashPlayer and other non-F/OSS is not on the install media in most cases).
Back on topic, as with all IE versions there will be some features that other browsers probably won't support (such as Web Slices with IE8), but hopefully no page will actually require that browser version. IE8 already immensely cleaned up CSS compatibility,a nd it looks like IE9 is doing the same thing with JavaScript compatibility.
As for XP users, I can't really say I'm sorry to hear that. The OS is old, and missing many security features (such as the per-process security model that makes it possible to sandbox IE7/8 on Vista/Win7, or things like ASLR that make it so hard to write a working exploit even when a vulnerability is found in some program). MS doesn't want to extend support for it any longer than necessary. Heck, it would simplify things for ISVs if they could assume their customers were running an OS from this half of the decade.
They could address the issue by using encryption, checksums, and so forth. Of course, they could have done that for third-party cards too, but if they weren't planning to support externally read/write-capable devices they wouldn't have had a need for security on the data. Now that they are adding such support, it's *possible* that they will also add such encryption.
This comes pretty close to what I heard from a guy who worked on the Windows virtualization team about 9 months ago. I posted about it here http://slashdot.org/comments.pl?sid=1584998&cid=31505036 but basically the gist was that when they were developing the next Virtual PC version (started over 3 years ago), they asked Intel and AMD if they would have hardware assisted virtualization on all CPUs by then. Both companies said yes, so MS decided to save time and testing costs by removing the code that handles non-hardware-assisting CPUs.
At the time, Win7 hadn't quite even shipped yet, but he said they were considering adding the support back in after release, to make up for Intel screwing the pooch. Sounds like that's what happened.
as i pointed out, i have two 360s, and i need to move my profile from machine to machine (along with saves). The only easy way to do this is to have the profile and saves on a memory card, which are frickin expensive.
Why not just detach the hard drive and transfer it? They're external modules, designed to be easily removed and replaced. I'll admit I've not tried this - perhaps it doesn't work - but this seems like the logical approach.
It's entirely possible. The full set of step-by-step can be found easily via a web search (try "Enable Bitlocker without a TPM" or similar) but the general gist of it is as follows:
Open group policy editor (gpedit.msc) The setting you want is down this path:
Administrative Templates
Windows Components
BitLocker Drive Encryption
Operating System Drives Open the settings for "Require additional authentication at startup" Enable the configuration and select "Allow BitLocker without a compatible TPM" Make sure the other settings are sane, then save the policy change.
BitLocker will require the use of both a flashdrive (where the key is stored) and a PIN (used to unlock the key) at boot time. Lose the flashdrive of forget the PIN, and you'll have to use a recovery agent. That said, ti works fine for systems that want encryption but lack TPMs.
The "Dynamic Memory" thing sounds cool, but it sounds like specific to servers, i.e. for Hyper-V. This is also really too early to know exactly what the SP will or won't contain; everybody knew there would be one and it's easy to make an approximate timeline for it, but SP1 rarely contains any major new features anyhow. They can still add additional minor improvements like parallelizing more of the core code or something - you probably wouldn't notice specifically, but the system would be faster on a multi-core machine than it was before. It takes a lot of testing to be sure something like that doesn't cause a problem, though.
FlashPlayer, Acrobat Reader, RealPlayer, and Opera are all available for Linux. On my distro (openSuse) they're in the official "NON_OSS" repository, and will happily install and update automatically using the package manager.
Installing support for the Microsoft, Apple, and Real codecs, plus DVD decoding (without installing RealPlayer, which I hate) was also quite easy. These are in "community repositories" that the package manager suggests when you select "Add repository"; you don't even have to enter the URL or anything yourself.
I don't use console emulators personally, but I've noticed quite a variety of them in the repositories. NES, SNES, N64, several Atari, and others were all present.
As for games, there are a couple of good games available for Linux natively (Battle for Wesnoth and Heroes of Newerth come to mind, but there are many others). The majority of games will also run, often flawlessly and with similar performance, under Wine; I play WarCraft 3, StarCraft, Total Annihilation, EVE Online, Heroes of Might and Magic III, and more using this method. Don't get me wrong; there are games that won't work. However, they're actually rather rare these days.
(Yes, I know it's still technically possible to get a virus. But the chances are extremely slim, given the way I use my computer.)
Do you use FlashPlayer or Acrobat Reader, and if so do you run them as Admin? If so, you're not safe. There are numerous exploits for both, and Adobe is very slow about patching them. If your browser is configured to prompt you before downloading/displaying PDFs, then you at least have a chance to block those - I've had several completely legit sites attempt drive-by PDF downloads to my system, presumably containing malware (no other reason a webcomic or forum would try to send my a PDF upon loading the main page). For FlashPlayer, your only safe options are disable it or sandbox it. I went with sandboxing once I figured out how, and before that disabled it except for specific uses on specific sites.
Of course, if you run as Admin/root you're *still* asking to get hosed. That's just a terrible idea all around. Zero-day exploits happen, and the attacks can come from pretty much anywhere. Short of disconnecting entirely, the best you can hope for is that they do little to no damage (most malware expects full access and won't run if it doesn't have it).
Note that I'm a smart computer user who keeps everything patched and up to do, as well as knows how to configure a hardware router/firewall.
I see a lot of people claim things like this. The question I ask every one of them, especially if they run XP (an outdated OS missing a number of modern security features, like application sandboxing and ASLR), is whether they run as Administrator or not. 95% still say Yes (beats the approximately 99.9% otherwise, but... still too high). Running as Admin is a *terrible* idea - you might as well be running Windows ME, in terms of security - yet far too many people do so anyhow.
I'll grant you that running as a non-admin on XP or older is a pain - it was that pain which drove me to Linux in the first place. Now I dual-boot Win7 and Linux (Vista and Linux on my older machine) and things have worked out very well. I don't have any continuous monitoring AV running (I keep a copy of ClamAV for on-demand scans), I don't disable UAC or Protected Mode (in fact, I tweak the UAC settings and remove FlashPlayer's exemption regarding Protect Mode). A few UAC or sudo prompts a month is easily worth the extra protection that not running as Admin provides. Security is all about defense in depth, and relying solely on anti-intrusion methods is stupid.
Yes, there's still a lot of harm that can be done with standard user permissions. However, most malware authors, especially for Windows, assume that their code will run as Admin/root, and therefore it would fail on my system anyhow. Furthermore, without Admin, malware can't make itself un-removable. It might send spam or DDOS attempts, but it couldn't edit my firewall settings, hide itself from task manager, install kernel-mode code, or prevent me from deleting it.
There's an annoying but simple reason for their use: Slashdot tracks the URL of submitted articles, and if one submission is rejected you can't place another submission with the same URL. Supposedly it cuts down on the junk submissions, but it also means that if you have a good article with a bad submission, the bad submission gets rejected and somebody with a good submission now can't use that URL. Hence, the use of shorteners.
Other than multiple inheritance, which as far as I know is present in exactly one language, what do you consider critical for "full" inheritance then?
It's got interfaces and abstract classes, overloadable and non-overloadable functions, visibility modifiers including things like protected (might use a different keyword, but same effect), you can get your superclass, every class inherits from a base class... seriously, that's a pretty good set of inheritance features.
I actually had a chance to talk with one of the guys behind Wirtual PC (including the one that powers Virtual XP Mode) a few months back. His answer was that when MS started work on the Windows Virtual PC (WVPC, the current version) 3 years ago (roughly when VPC2k7 shipped), they contacted Intel and AMD and asked if they were going to offer hardware virtualization on all models going forward. Both companies answered Yes, so to save development and testing costs, the (large block of tricky) code that enabled vitrualization on CPUs without hardware support was cut from WVPC. Skip forward a couple years to WVPC getting ready for release, and... still a lot of Intel CPUs without support for virtualization. At that point it was too late to add the feature without delaying shipping, so they didn't.
I asked whether they would add it in later, and he said they weren't planning to but might change their mainds if it became clear that this was a problem for enough of their customers. In the meantime, they were telling Intel to get its shit together the way it had said it would three years earlier, and he personally recommended going with AMD chips.
It's probably just a bug in the way VirtualPC handles the virtual TLBs or some such. It's not even present on Hyper-V, also from Microsoft, so I think the danger here is pretty low.
It's not like this actually makes the host OS vulnerable, either. I doubt it can even crash the VM software, although it could certainly lead to crashing the virtualized OS.
Honeypots are designed to get hit. This bug doesn't make the host system vulnerable, it just means that the client OS is easier to exploit.
If it worked on Hyper-V, this would be a big problem; that's a server-level technology where even the clients are expected to remain secure. On the other hand, Virtual PC isn't even a hypervisor; it requires a full OS onderneath it, running itself as just another Windows app. Up until 2007 didn't even require hardware support for virtualization.
This is definitley a bug, but all it does is allow bypassing of security features in the virtualized system. In other words, you can exploit the VM client, but you still can't get at the host.
It's worth of a patch, but not of a panic. If you're virtualizing for security, you don't really care what happens to the virtual system (that's the point). If you're virtualizing so you can run an old OS, it's going to be full of holes anyhow. If you're virtualizing for any other reason, why the hell are you using consumer-grade virtualization software?
You could just, you know, go *download* it. I mean the link is right there in TFS! Here, have link to the actual download site itself if you want:http://ie.microsoft.com/testdrive/info/ThankYou/Default.html
Not a full browser by any means, but the rending is a big step up from IE8 and I expect it will continue to improve.
Funny, but navigating via just links and the "Open URL" dialog (pops up when pressing Ctrl-O in the IE9 preview) works surprisingly well. I really don't ectually use the address bar that much.
That said, the lack of tabs, the lack of Back functionality (not just the button, the function itself is missing), and fact that it doesn't use my ad-blocker mean it's not something I'm going to use very heavily day-to-day. It is pretty damn awesome seeing Slashdot's discussion2 AJAX stuff go so fast, though.
Believe it or not, probably at least half of your list of people who know "nothing about browsers" still run Windows Update, which means they could easily have accepted the upgrade to IE7 or 8. It's not mandatory, but it does somewhat pop out at you. Outside of locked down corporate environments, I see very little of IE6 anymore. Then again, maybe I'm just lucky (it also helps that most of my friends are fairly technically inclined).
You also should have mentioned that IE8, the browser, is completely discoupled from Windows 7. In fact, in the EU if you buy Windows 7 it comes without IE. Anybody who wants to can easily remove IE8 from Win7 by using the "Turn Windows features on or off" control panel.
Mind you, this only removes the browser. The rendering engine (found is mshtml.dll) is required by an awful lot of things, including third-party software. Remove it and stuff stops working everywhere. Of course, one can replace Trident with another rendering engine - this is what Wine does, and it works in a lot of cases - but the existence of a rendering engine on an operating system is just something people take for granted these days, and on Windows it's expected that you have Trident.
They could, of course, just include *both* codecs. It's not like there's any technological or legal reason you can't do Theora decoding on Windows - they just don't ship with it.
Posted from IE9 preview, by the way. It's definitely not a fully functional drowser - just a thin shell around the rendering engine - but it works pretty damn well. Handles the JS on Slashdot rather nicely.
Question A: Where is it required that Nokia allow Apple (who provided no help toward developing 3G, unlike other phone manufacturers) to license the relevant patents at the same price as another company that directly helped develop the technology? Citation needed.
Question B: Hypothetically, suppose you're right and Nokia was obligated to offer Apple the same pricing. In that case, which of the following options is legal for Apple to take: (A) File a lawsuit against Nokia (B) Sell their product using unlicensed technology anyhow?
Slashdot Mirror
My responses (because his premise is flawed; that is, I do have a tablet):
1. Tablets are devices which fit niches I occupy, such as being a university student (fantastic for taking notes or annotating the professor's slides during lecture). Lots of people occupy such niches.
2. Remarkably enough, some of those who complained that the iPad doesn't have a full OS have tablets with full OSes, and love them. My tablet is also a highly portable laptop, and that is a *good* thing.
3. Depends on your definition of "high end" here; my tablet is 1.2 GHz Core 2 Duo, with 4GB of RAM. That's a lot higher than the iPad and a lot lower than the computer I type this on now. An ARM chip would give better battery life but would limit the computer's usefulness. A 2.8 GHz processor and nVidia GPU would give amazing performance but unless the battery got a lot bigger it wouldn't last me through a single lecture period. In other words, this is a good balance - it's fast enough for everything from taking notes to writing code to light gaming (nothing graphics intensive), yet the battery lasts 5 hours active (over a week in sleep) which is enough I typically only charge it every other day.
4. Interface is definitely important. On a tablet, the Win7 taskbar is larger so it's easier to hit with a stylus or finger. There are various gestures and pen flicks that the OS recognizes, that do things like "Go back" or "Cut"/"Copy"/"Paste" I'll grant you that a dedicated interface might be better for some things, but the Win7 one works surprisingly well after all the little tablet-related features are enabled (which they are by default on tablet computers, and aren't by default on non-tablets).
5. OneNote. Seriously, Microsoft Office OneNote is a damn good reason to get a tablet. It works on normal hardware, but it doesn't really shine until it's on a tablet. There are other programs that are great with tablets (including a lot of design or artistic stuff), but OneNote is the main one for me.
And at that, it's actually already got some very impressive capabilities. In particular, and relevant to the discussion at hand (hardware-accelerated 3D) try this page in you favorite browser(s): http://ie.microsoft.com/testdrive/Performance/01FlyingImages/Default.html .
The IE9 preview, early though it is, blows Firefox et. al completely out of the water. Try running the number of images up to the max (256), zooming in, and holding the Shift key while moving your mouse. On my machine, IE9 preview will still get 28 FPS. Opera 10.5 will come next closest, at about 9.5 but with terrible background flicker. Firefox 3.6 gets about 5.5 FPS, not smooth at all.
This is true, but also irrelevant. We're talking about the possibility of a specific performance optimization here. Obviously, software rendering *can* do 3D - this has been known for decades, and is what other browsers use if you go to a 3D website like http://ie.microsoft.com/testdrive/Performance/01FlyingImages/Default.html . The fact that it's possible to do the same thing does not mean that it's possible to do it with the same performance though, which is the crux of the matter. WDDM drivers provide a lot of capabilities that non-WDDM drivers lack. WDDM drivers require kernel driver interfaces that XP lacks (if you added support for them to XP, it wouldn't really be XP anymore; this is a non-trivial modification of an old and long-stable portion of the kernel). The way IE9 gets the performance it does on 3D relies upon these capabilities. No Vista/Win7 = No WDDM = No support for those features = No IE9 hardware accelerated performance. There might be another way to get equivalent performance, but neither MS nor anybody else has demonstrated it yet.
I'm not sure what you think is "funny" about that, but maybe you just have an odd sense of humor. Actually, I suppose if you were the sort to bash anything that MS does which looks like a mistake, without any idea of what was involved, yeah it might be funny.
Go visit this page, in whatever browser(s) you please: http://ie.microsoft.com/testdrive/Performance/01FlyingImages/Default.html
I don't have Safari or Konqueror installed here, but I think you get the idea. IE8 can't even execute the page's code right. Chrome crawls at the 3D effect. Firefox is OK at wide angle and crawls when zoomed. Opera is the fastest of the released browsers, but has horrible flicker when images pass the edge of the screen. None of these browsers use hardware accelerated drawing.
The very, very early preview of IE9, which does use hardware acceleration, blows them all away. No performance degradation until it reaches the point where most other browsers drop to the single digits. No trouble with zooming or fast motion. No flicker or tearing. If you'd told me I was watching a demo of a 3D engine written in C++, I'd have believed you (not been terribly impressed, but believed you). For something using pure JavaScript I'm amazed.
My system has a mid-range GPU (GeForce 9600M) but pretty good CPU (Core 2 Duo T9600, 2.8GHz), running Win7 x64. I'm guessing IE9 uses vertical sync, since it maxes at my refresh rate (60 Hz). Clearly, simply compiling the JS to native binary isn't enough to get the really impressive performance, since the other browsers do that. Since Vista/Win7 use 3D to render the desktop anyhow, I can certainly believe it's easier to incorporate this kind of functionality into those operating systems. It may be possible with XP, but so far there's no indication that you can get comparable performance - none of the browsers that will run on XP can, at least.
As a side note, it's worth pointing out that openSuse is to Suse Linux Enterprise (commercial) as Fedora is to Red Hat - same codebase, but openSuse is typically running newer versions, is non-commercial (including for updates), and the base install does not include things like Flashplayer (although it includes an option to install them immediately after the system boots up for the first time, FlashPlayer and other non-F/OSS is not on the install media in most cases).
Back on topic, as with all IE versions there will be some features that other browsers probably won't support (such as Web Slices with IE8), but hopefully no page will actually require that browser version. IE8 already immensely cleaned up CSS compatibility,a nd it looks like IE9 is doing the same thing with JavaScript compatibility.
As for XP users, I can't really say I'm sorry to hear that. The OS is old, and missing many security features (such as the per-process security model that makes it possible to sandbox IE7/8 on Vista/Win7, or things like ASLR that make it so hard to write a working exploit even when a vulnerability is found in some program). MS doesn't want to extend support for it any longer than necessary. Heck, it would simplify things for ISVs if they could assume their customers were running an OS from this half of the decade.
They could address the issue by using encryption, checksums, and so forth. Of course, they could have done that for third-party cards too, but if they weren't planning to support externally read/write-capable devices they wouldn't have had a need for security on the data. Now that they are adding such support, it's *possible* that they will also add such encryption.
This comes pretty close to what I heard from a guy who worked on the Windows virtualization team about 9 months ago. I posted about it here http://slashdot.org/comments.pl?sid=1584998&cid=31505036 but basically the gist was that when they were developing the next Virtual PC version (started over 3 years ago), they asked Intel and AMD if they would have hardware assisted virtualization on all CPUs by then. Both companies said yes, so MS decided to save time and testing costs by removing the code that handles non-hardware-assisting CPUs.
At the time, Win7 hadn't quite even shipped yet, but he said they were considering adding the support back in after release, to make up for Intel screwing the pooch. Sounds like that's what happened.
as i pointed out, i have two 360s, and i need to move my profile from machine to machine (along with saves). The only easy way to do this is to have the profile and saves on a memory card, which are frickin expensive.
Why not just detach the hard drive and transfer it? They're external modules, designed to be easily removed and replaced. I'll admit I've not tried this - perhaps it doesn't work - but this seems like the logical approach.
It's entirely possible. The full set of step-by-step can be found easily via a web search (try "Enable Bitlocker without a TPM" or similar) but the general gist of it is as follows:
Open group policy editor (gpedit.msc)
The setting you want is down this path:
Administrative Templates
Windows Components
BitLocker Drive Encryption
Operating System Drives
Open the settings for "Require additional authentication at startup"
Enable the configuration and select "Allow BitLocker without a compatible TPM"
Make sure the other settings are sane, then save the policy change.
BitLocker will require the use of both a flashdrive (where the key is stored) and a PIN (used to unlock the key) at boot time. Lose the flashdrive of forget the PIN, and you'll have to use a recovery agent. That said, ti works fine for systems that want encryption but lack TPMs.
The "Dynamic Memory" thing sounds cool, but it sounds like specific to servers, i.e. for Hyper-V. This is also really too early to know exactly what the SP will or won't contain; everybody knew there would be one and it's easy to make an approximate timeline for it, but SP1 rarely contains any major new features anyhow. They can still add additional minor improvements like parallelizing more of the core code or something - you probably wouldn't notice specifically, but the system would be faster on a multi-core machine than it was before. It takes a lot of testing to be sure something like that doesn't cause a problem, though.
FlashPlayer, Acrobat Reader, RealPlayer, and Opera are all available for Linux. On my distro (openSuse) they're in the official "NON_OSS" repository, and will happily install and update automatically using the package manager.
Installing support for the Microsoft, Apple, and Real codecs, plus DVD decoding (without installing RealPlayer, which I hate) was also quite easy. These are in "community repositories" that the package manager suggests when you select "Add repository"; you don't even have to enter the URL or anything yourself.
I don't use console emulators personally, but I've noticed quite a variety of them in the repositories. NES, SNES, N64, several Atari, and others were all present.
As for games, there are a couple of good games available for Linux natively (Battle for Wesnoth and Heroes of Newerth come to mind, but there are many others). The majority of games will also run, often flawlessly and with similar performance, under Wine; I play WarCraft 3, StarCraft, Total Annihilation, EVE Online, Heroes of Might and Magic III, and more using this method. Don't get me wrong; there are games that won't work. However, they're actually rather rare these days.
(Yes, I know it's still technically possible to get a virus. But the chances are extremely slim, given the way I use my computer.)
Do you use FlashPlayer or Acrobat Reader, and if so do you run them as Admin? If so, you're not safe. There are numerous exploits for both, and Adobe is very slow about patching them. If your browser is configured to prompt you before downloading/displaying PDFs, then you at least have a chance to block those - I've had several completely legit sites attempt drive-by PDF downloads to my system, presumably containing malware (no other reason a webcomic or forum would try to send my a PDF upon loading the main page). For FlashPlayer, your only safe options are disable it or sandbox it. I went with sandboxing once I figured out how, and before that disabled it except for specific uses on specific sites.
Of course, if you run as Admin/root you're *still* asking to get hosed. That's just a terrible idea all around. Zero-day exploits happen, and the attacks can come from pretty much anywhere. Short of disconnecting entirely, the best you can hope for is that they do little to no damage (most malware expects full access and won't run if it doesn't have it).
Note that I'm a smart computer user who keeps everything patched and up to do, as well as knows how to configure a hardware router/firewall.
I see a lot of people claim things like this. The question I ask every one of them, especially if they run XP (an outdated OS missing a number of modern security features, like application sandboxing and ASLR), is whether they run as Administrator or not. 95% still say Yes (beats the approximately 99.9% otherwise, but... still too high). Running as Admin is a *terrible* idea - you might as well be running Windows ME, in terms of security - yet far too many people do so anyhow.
I'll grant you that running as a non-admin on XP or older is a pain - it was that pain which drove me to Linux in the first place. Now I dual-boot Win7 and Linux (Vista and Linux on my older machine) and things have worked out very well. I don't have any continuous monitoring AV running (I keep a copy of ClamAV for on-demand scans), I don't disable UAC or Protected Mode (in fact, I tweak the UAC settings and remove FlashPlayer's exemption regarding Protect Mode). A few UAC or sudo prompts a month is easily worth the extra protection that not running as Admin provides. Security is all about defense in depth, and relying solely on anti-intrusion methods is stupid.
Yes, there's still a lot of harm that can be done with standard user permissions. However, most malware authors, especially for Windows, assume that their code will run as Admin/root, and therefore it would fail on my system anyhow. Furthermore, without Admin, malware can't make itself un-removable. It might send spam or DDOS attempts, but it couldn't edit my firewall settings, hide itself from task manager, install kernel-mode code, or prevent me from deleting it.
There's an annoying but simple reason for their use: Slashdot tracks the URL of submitted articles, and if one submission is rejected you can't place another submission with the same URL. Supposedly it cuts down on the junk submissions, but it also means that if you have a good article with a bad submission, the bad submission gets rejected and somebody with a good submission now can't use that URL. Hence, the use of shorteners.
Other than multiple inheritance, which as far as I know is present in exactly one language, what do you consider critical for "full" inheritance then?
It's got interfaces and abstract classes, overloadable and non-overloadable functions, visibility modifiers including things like protected (might use a different keyword, but same effect), you can get your superclass, every class inherits from a base class... seriously, that's a pretty good set of inheritance features.
I actually had a chance to talk with one of the guys behind Wirtual PC (including the one that powers Virtual XP Mode) a few months back. His answer was that when MS started work on the Windows Virtual PC (WVPC, the current version) 3 years ago (roughly when VPC2k7 shipped), they contacted Intel and AMD and asked if they were going to offer hardware virtualization on all models going forward. Both companies answered Yes, so to save development and testing costs, the (large block of tricky) code that enabled vitrualization on CPUs without hardware support was cut from WVPC. Skip forward a couple years to WVPC getting ready for release, and... still a lot of Intel CPUs without support for virtualization. At that point it was too late to add the feature without delaying shipping, so they didn't.
I asked whether they would add it in later, and he said they weren't planning to but might change their mainds if it became clear that this was a problem for enough of their customers. In the meantime, they were telling Intel to get its shit together the way it had said it would three years earlier, and he personally recommended going with AMD chips.
On a side note, I didn't hear anything at all about it requiring x64. According to http://www.microsoft.com/windows/virtual-pc/download.aspx, it works just fine on x86.
It's probably just a bug in the way VirtualPC handles the virtual TLBs or some such. It's not even present on Hyper-V, also from Microsoft, so I think the danger here is pretty low.
It's not like this actually makes the host OS vulnerable, either. I doubt it can even crash the VM software, although it could certainly lead to crashing the virtualized OS.
Honeypots are designed to get hit. This bug doesn't make the host system vulnerable, it just means that the client OS is easier to exploit.
If it worked on Hyper-V, this would be a big problem; that's a server-level technology where even the clients are expected to remain secure. On the other hand, Virtual PC isn't even a hypervisor; it requires a full OS onderneath it, running itself as just another Windows app. Up until 2007 didn't even require hardware support for virtualization.
This is definitley a bug, but all it does is allow bypassing of security features in the virtualized system. In other words, you can exploit the VM client, but you still can't get at the host.
It's worth of a patch, but not of a panic. If you're virtualizing for security, you don't really care what happens to the virtual system (that's the point). If you're virtualizing so you can run an old OS, it's going to be full of holes anyhow. If you're virtualizing for any other reason, why the hell are you using consumer-grade virtualization software?
You could just, you know, go *download* it. I mean the link is right there in TFS! Here, have link to the actual download site itself if you want:http://ie.microsoft.com/testdrive/info/ThankYou/Default.html
Not a full browser by any means, but the rending is a big step up from IE8 and I expect it will continue to improve.
Funny, but navigating via just links and the "Open URL" dialog (pops up when pressing Ctrl-O in the IE9 preview) works surprisingly well. I really don't ectually use the address bar that much.
That said, the lack of tabs, the lack of Back functionality (not just the button, the function itself is missing), and fact that it doesn't use my ad-blocker mean it's not something I'm going to use very heavily day-to-day. It is pretty damn awesome seeing Slashdot's discussion2 AJAX stuff go so fast, though.
Believe it or not, probably at least half of your list of people who know "nothing about browsers" still run Windows Update, which means they could easily have accepted the upgrade to IE7 or 8. It's not mandatory, but it does somewhat pop out at you. Outside of locked down corporate environments, I see very little of IE6 anymore. Then again, maybe I'm just lucky (it also helps that most of my friends are fairly technically inclined).
You also should have mentioned that IE8, the browser, is completely discoupled from Windows 7. In fact, in the EU if you buy Windows 7 it comes without IE. Anybody who wants to can easily remove IE8 from Win7 by using the "Turn Windows features on or off" control panel.
Mind you, this only removes the browser. The rendering engine (found is mshtml.dll) is required by an awful lot of things, including third-party software. Remove it and stuff stops working everywhere. Of course, one can replace Trident with another rendering engine - this is what Wine does, and it works in a lot of cases - but the existence of a rendering engine on an operating system is just something people take for granted these days, and on Windows it's expected that you have Trident.
They could, of course, just include *both* codecs. It's not like there's any technological or legal reason you can't do Theora decoding on Windows - they just don't ship with it.
Posted from IE9 preview, by the way. It's definitely not a fully functional drowser - just a thin shell around the rendering engine - but it works pretty damn well. Handles the JS on Slashdot rather nicely.
Question A: Where is it required that Nokia allow Apple (who provided no help toward developing 3G, unlike other phone manufacturers) to license the relevant patents at the same price as another company that directly helped develop the technology? Citation needed.
Question B: Hypothetically, suppose you're right and Nokia was obligated to offer Apple the same pricing. In that case, which of the following options is legal for Apple to take:
(A) File a lawsuit against Nokia
(B) Sell their product using unlicensed technology anyhow?