"I'd have to disagree. There are many instances were DVCS are always superior. However, there are times - such as in corporate environments - where you simply do not want that kind of information floating around the organization."
That's just an empty string of words. What do you mean be 'not floating around'? Access controls? So use them on per-repository level, duh.
Information leaks? Developers can just use working copies for that.
With DVCS systems, like git, every node in the system hosts the whole repository. Aside from the initial access to the system, there's not much you can do per access controls - that's the nature of distributed systems where the information exists in the "cloud" and there is no centralized location.
You can only really only do access controls when you control the whole system - thus you require a centralized system instead of a decentralized system.
However, one person (e.g. Microsoft, gov't, gov't agency, etc.) controlling what all users - including all corporations - can do with the computers doesn't work. Thus 'trusted' computing is flawed...
Please go read the thread you're replying to. We're talking about a trust model where the greylists are created by weighted combinations of security threads from multiple sources, as weighted by the end user, with a very, very rarely user override option. There is no "one person" by the definition of what we're talking about.
While you do portend that there should be no single entity controlling it, there is nonetheless an entity other than the user controlling it - even if multiple entities, they will likely form together into a gov't agency or consortia of some sort in the end, thus a single entity any way.
Trusted computing (as you suggest with weights, etc for the user to adjust) still would not work. Why? If the user can modify what the system can trust, so can anything malicious. All it takes is for the malware to be one step ahead of those writing lists - e.g. just like all viruses today. So no, that is not a working model - it works no better than what we have today, and in fact makes it worse since it creates a false sense of security as well.
Thus, my point that Trusted Computing is flawed still stands.
Also, if you notice, my initial reply ignored your weights, etc - replying to what you had before that. The point, nonetheless, is still able to be inclusive of it though as shown above.
Their description of Subversion is almost blatantly wrong, and misses much of the improvements SVN brought about. It would have helped them to at least have read some of the Subversion Documentation - or even just the chapter on Subversion's Delta Editor in the book Beautiful Code.
Nope. Distributed systems are ALWAYS more useful than centralized ones for source code control.
Only sometimes their advantages are not that significant.
I'd have to disagree. There are many instances were DVCS are always superior. However, there are times - such as in corporate environments - where you simply do not want that kind of information floating around the organization. In those instances a Centralized VCS is superior as the main advantage (distributed version control) is in those situations the biggest disadvantage of a DVCS.
Anything stored in the BIOS as executable code isn't going to be removed via a jumper or removal of battery. Reflashing the BIOS maybe but not simply clearing volatile BIOS flags.
BIOS is stored on a motherboard in two fashions: (1) a flashable (read/write) memory chip, and (2) a hard-wired (read-only) memory chip. When you update (flash) the BIOS, you only overwrite the flashable memory chip. This chip requires power to keep the BIOS data alive; the power comes from either the battery on the motherboard or the power supply. If you disconnect the system from the wall, and remove the battery from the motherboard then the flashable BIOS _will_ be reset. (Yes, I've reset BIOS's this way. I have several systems that I use to update the BIOS regularly on.) It'll happen faster if you disconnect the power supply from the motherboard as well - since residual energy could be in the capacitors in the power supply. Supposing all power is disconnected from the motherboard it takes about 30 seconds to drain all power and reset the BIOS.
Some motherboards have jumpers or dip-switch settings to do the same thing automatically - via hardware.
All of this, of course, only reset the BIOS back to what was shipped from the factory - what is stored on the hard-wired memory chip.
Now, how many people bother with updating the motherboard BIOS now-a-days? No many, thus it may only be the settings in that case. But this will clear viruses out of the BIOS. It'll also reset the BIOS in case of a bad update.
Some motherboards even come with a dual-bios system where only one can be written to and it is non-executable, and the other can only be read and executed. The active BIOS (the executable one) has a number of safety checks in it to try to prevent a BIOS virus from taking hold. Though not impossible to put a virus on these motherboards, it is a lot more difficult to do.
But how that 99% of society wants to use the computer should not ( and cannot necessarily) be dictated by even the 1% as the 1% will not know every edge case for how the 99% wants to use the computer.
Actually 99% of users will probably never do anything that would even be an issue. Malware primarily runs because users are not informed by the OS that it is malware or told that it is accessing their address book and starting a mail server or constantly spamming traffic at an address in Estonia. For the other 1% of cases the user needs the option to override the security system, but this should never be needed for normal use cases so when an app requests this it should be a red flag to users. Right now they're so conditioned by our poor OS UIs they just click through things. But if a users was never, ever (over the course of owning a machine and later over their lifetime) asked o override security and they were asked at some point with language worded to say doing so would allow someone else control of their computer forever, I think that would make a huge difference, don't you?
Who mentioned malware? Yes, malware is one thing that needs protected against; however, what a user wants to do with the computer may not necessarily be what the person controlling the system wants the user to do with the computer. This is almost fine in a corporate environment where the computer uses are dictated by the organization - though even then, that doesn't quite work as managers hire people to do things that IT didn't account for in their 'standard platform' (e.g. developers). However, one person (e.g. Microsoft, gov't, gov't agency, etc.) controlling what all users - including all corporations - can do with the computers doesn't work. Thus 'trusted' computing is flawed in the current model, and you don't even need to consider malware/viruses/etc to come to that analysis.
Have there actually been any MBR "bootkits" in the wild that have used flashable BIOS for storing copies? I always though that was a malware "urban legend". And shouldn't any flashable BIOS have some sort of jumper switch to prevent unauthorized flashing to being with?
Yes there are, and the symptoms are hard to relate. It's things like the PS/2 mouse won't be detected, or the floppy drive won't work right. Had one on my desktop back in college - only virus I ever had. And yes, the only way to get rid of them. Variants of the Monkey virus do store themselves into the BIOS.
Yeah, you could do all that, or you could just use MBRFIX and be done with it. "Extremely difficult" - I don't really think so. Typing 6 letters from a command prompt isn't something I would categorize as "extremely difficult".
Problem there - you could have the virus stored elsewhere on the computer. Many of those kinds of viruses will put themselves into start-up software as well; so running MBRFIX will remove it from the MBR yes, but then you'll re-infect the MBR on first boot.
And if you run MBRFIX from a machine that is infected - without booting from a CD first - then the virus will be in memory and just re-infect the MBR before you reboot.
So yes, it's not as simple as running MBRFIX - you actually have to disinfect the system too, which means eliminating any points where the infection may be - thus removing the battery on the motherboard to reset the BIOS, running the disk through an AV program on _another_ computer (or via a CD on the same computer), and fixing the MBR.
Old sk00l. When was the last MBR infector seen in the wild? 2002? Most of this class are from the DOS era, fercryingoutloud.
From the second paragraph of the fine article (emphasis added):
TDSS has been causing serious trouble for users for more than two years now, and is an example of a particularly pernicious type of rootkit that infects the master boot record of a PC. This type of malware often is referred to as a bootkit and can be extremely difficult to remove once it's detected. The older versions of TDSS--TDL1, TDL2 and TDL3--are detected by most antimalware suites now, but it's TDL4 that's the most problematic right now.
Easy way to remove it: (1) turn on the computer, (2) load the hard drives into another computer, scan, and dis-infect without loading the software on the computer - it should be read-write without execution permission; and finally (3) reset the firmware on the devices, especially the motherboard, on the computer. (Typically just the motherboard; but there could be other devices too that may need it.). Put it back-together, boot up, and your infection free.
OR
You could just run Linux/BSD/etc to start with and not have to worry about it.
Give me "trusted computing" where I control the keys and decide what software is "trusted" and I'd be fine w/ it.
The problem is, 99% of our society cannot properly decide whether software should be trusted or not, and even with more granular controls and proper feedback from the OS a lot of malware will slip through.
I don't think this is an unsolvable problem.
But how that 99% of society wants to use the computer should not ( and cannot necessarily) be dictated by even the 1% as the 1% will not know every edge case for how the 99% wants to use the computer. Thereby, "trusted" computing in that model is 100% flawed, and you then have to build in backdoors - like the register key that can disable requiring a signed driver so developers can test their drivers - so that the 99% can all do what they want/need to on the computer.
Didn't another CEO get in trouble over having a Twitter account and other social media? Perhaps the SEC needs to investigate Microsoft over the matter since Ballmer could potentially be giving advise or hinting at matters and cause insider trading to occur, even if unintentional.
Yes, though typically only for specialty environments. Intel still sells the 80086 processor. You can still get the i386 processor too, and boards that support them. But again, you're dealing with specialty not consumer products.
The point is that it is this time it is going to be hardware that prompts the change not software as it has in the past. As I said Microsoft will release at least one and probably more than one new version of windows before this comes to pass. I'd guess 5 to 10 years before hardware presents a tangible incentive to move away from XP.
Well, Microsoft may not be around in 5 to 10 years; so that may prompt a software change in itself. Though I do agree, it's going to be hardware pushing some of the changes - namely the 32-bit to 64-bit or even to a 128-bit change. Software has increasingly become less relevant to speed and memory changes as the basic speeds and memory available are more than sufficient for 99% of tasks; with that 1% being the gaming and multimedia editing (yes big markets, but not very big compared to the rest of the entire computing industry).
Yes you can still find supported hardware on the retail level but how much of it is from large OEM outfits that can support an order of 100,000 units? And how long will they continue to offer those options?
Probably for quite a while still. If you can find it in retail, chances are you have a better chance of finding it OEM - especially for things that are less often modified (e.g. USB->Serial converters, etc.) so they have longer life cycles.
WinXP x64 while available was not used very much and was known for having all kinds of problems. However, you don't need the 64-bit WinXP version to do what I was saying - PAE (Physical Address Extension) is something from the processor level that allows addressing +4GB of memory. Linux used it a long time ago; but Windows didn't until WinXP - and even then you had to have the right version for a long time. MS may have enabled it by default for WinXP SP3, not sure. But any how, it's available.
Fortunately Vista 64 is a lot better than WinXP x64; though there are still some driver issues. Win7 improved on that I believe. And Win8 or Win9 probably won't have a 32-bit edition.
Windows XP, which can only see and use 2 GB. Time to roll in a 64-bit OS!
WinXP can see more than 2 GB. We have systems that use 3 GB; we'd put 4 GB in but there's something wrong with the BIOS so it only sees 3GB. I am aware of other XP systems that have 4 GB, and with PAE it can see more than 4GB too.
Of course, Windows still takes up 1/3rd of the RAM for itself. but that's a different issue.
his will become especially true if and when 32-bit systems disappear. In the mean time Microsoft is going to realse at least one more version of Windows and likely more.
You mean like the 16-bit or 8-bit support disappeared? You do realize that even the latest x86/AMD64 processors still support the original 80086 processor modes, right? That that is their _default_ state of running when power is applied?
The only reason to upgrade is going to be the limited hardware support of XP when it comes time to finally start replacing systems.
Also, vendors seem to be continuing support for XP even as Microsoft drops it. So where is this limited hardware support you speak of? Not to mention that it's only been in the last few years that support for Win9x was almost completely dropped - yes, you can still find hardware that officially supports Win9x/Me, and that's a lot older than XP.
Currently. But it's been falling. Apple+Linux adds up to nearly 10% now; when MS use to command nearly 95-98%. And guess what? It's growing as 1 in 3 PCs are non-Microsoft.
wasn't Windows 7 a huge hit
How are you counting? As with Vista, most Win7 licenses have the XP downgrading rights attached. How many that purchased it are still using Win7? Versus how many have downgraded to XP? Most probably.
Especially since there's a nasty little interop issue with Win7 and older systems for file sharing - due to changing the encryption method. Yes you can disable it, but it takes quite a while to figure out what exactly you need to do. The few Win7 PCs we had at work only stayed after one server system was upgraded to a newer more compatible version of Windows (due only to the death of the server it replaced). Before that, all Win7 systems that came in got downgraded to XP as it was just easier.
Also, Vista and Win7 licenses are returning lower profits for a company that depends on the high profit margins. Add the lower sales and profits of Office into the mix and you have a company with a major problem. It may still take them a while to bleed to death, but there's a cannonball sized whole through their corporate torso that some flesh eating bacteria is infecting.
isn't xbox 360 kicking ass right now,
Huh? Wii has the top sales. PS3 is doing lowsy, but Xbox isn't doing much better. We should see at least another generation from Nintendo, who basically changed the market they were targeting. Sony and Microsoft will try another generation of PlayStation/XBox respectively but there's not much life left there as the money is just drying up.
or are we just judging Windows Phone 7? Cause if we are then i gotta say it's a bit early for that
A bit early? Possible. Though there are very few OEMs supporting it, and Microsoft has to blackmail OEMs to use it by telling them it'll be cheaper to use WP7 than to use Android and pay MS for the patent rights that MS is claiming they need. Yeah failure coming around the bend there too.
Of course, you could have been beegin sarcastic - which I was hoping for, in which case people better mod you funny; but some how I doubt it.
XP mode is a virtual machine inside your windows 7 installation. Once installed, the user sees the application no differently than any other app, but when it is launched, it runs seamlessly in the VM.
That would depend on the application. I know of one application I use to work on that was a server, ran in the system tray, provided a GUI interface, and was installed a Windows Service. Guess what? Starting with Windows Vista all services are no longer allowed to have a GUI interface. Good luck integrating that with Vista/7; and I very much doubt it would work well under XP mode too. Why? We had problems with the app under normal WinXP when terminal services was used - guess where the service was told its GUI interface should go? Guess where the user got sent? Yep, not the same console; could we hard-code it? No - the user's console was indeterminate.
Oh, and it would have been a major rewrite to move the application into a server+admin-gui architecture like it should have been.)
Yes, the majority of user applications will probably run fine; but not all of them. Any IT administrator worth their salt would know they need to test every application that their users need to perform their job before doing a roll-out, and yes - that goes beyond what would be part of the official company standard. The larger the organization, the more software that has to get tracked and verified, so at the very least users can be notified that a problem will exist whereby they may need to purchase new software - either newer versions or move to another product, if that is even possible.
So using TFA's numbers, which is it?
Original run-time would have been: 1m^3 = 1,000,000,000,000,000,000 -> 1,000,000,000,000 million -> 1,000,000 million million -> 1 million million million
But which is the new run-time? it's vastly different depending on order of operation:
(1000000*log(1000000))^2 -> (1000000*6)^2 -> 36,000,000,000,000 ->36,000,000M->36 million million
1000000*(log(1000000)^2) 1000000*(6^2) -> 1000000*36 = 36 million
I've thought OpenOffice was much better at opening older versions of Word (especially 97 versions) than Office . It seemed something changed after Office 2003. But that's just my perception.
OpenOffice writes better Microsoft Office documents than Microsoft Office does. Whether they legacy binary formats, or the newer XML formats. I've taken to using OO Writer for most things, and then ensuring everything is still there when I make the Word version, though it's mostly just the cross-references that I have to redo for some reason.
Yeah. the testing I've done on it shows that Win7 runs nicely on decently old hardware - actually runs better than Vista on the same hardware - provided you turn off the Aero interface. Do you lose visual perks? Sure. But it runs just fine.
All due to a super huge effort to (i) reduce dependencies, and (ii) eliminate circular dependencies between kernel and user space. The effort started with Vista, and Win7 is a lot farther along. Win8 (whatever it will be) should do better yet. Though I believe that they already got to the point where dependencies only go one way with Win7 - that is, kernel space does not depend on anything in user space as of Win7. (If not Win7, certainly Win8 should be that way.)
We all trash Microsoft for making shitty products, but in the end we would all work for them given the chance.
I did a couple rounds of interviews with Microsoft just to be able to say I interviewed with them. I figured, if they offered the job, I could say I declined it as well.
While the interviews were fun, I have zero intention of working for Microsoft at any point in my career given their ethics, tactics, etc.
So no, I do not know of whom you speak.
No, it is no different because remotely controlling a model aircraft is legal, and driving a car is legal.
However, pretty much every state in the US requires that the driver have 2 hands on the wheel. Not really enforced, mind you; but having an AI drive the vehicle with a human on "stand-by" would indeed still be illegal in the same fashion as driving with your leg, or letting the passenger hold the steering wheel for you, etc. They might not enforce it, but that doesn't mean its legal - just that the law is unenforceable and itself is illegal (since laws are required to be enforceable to be legal and good law - mind you a lot of laws do not measure up to that standard).
Slashdot Mirror
"I'd have to disagree. There are many instances were DVCS are always superior. However, there are times - such as in corporate environments - where you simply do not want that kind of information floating around the organization."
That's just an empty string of words. What do you mean be 'not floating around'? Access controls? So use them on per-repository level, duh.
Information leaks? Developers can just use working copies for that.
With DVCS systems, like git, every node in the system hosts the whole repository. Aside from the initial access to the system, there's not much you can do per access controls - that's the nature of distributed systems where the information exists in the "cloud" and there is no centralized location.
You can only really only do access controls when you control the whole system - thus you require a centralized system instead of a decentralized system.
However, one person (e.g. Microsoft, gov't, gov't agency, etc.) controlling what all users - including all corporations - can do with the computers doesn't work. Thus 'trusted' computing is flawed...
Please go read the thread you're replying to. We're talking about a trust model where the greylists are created by weighted combinations of security threads from multiple sources, as weighted by the end user, with a very, very rarely user override option. There is no "one person" by the definition of what we're talking about.
While you do portend that there should be no single entity controlling it, there is nonetheless an entity other than the user controlling it - even if multiple entities, they will likely form together into a gov't agency or consortia of some sort in the end, thus a single entity any way.
Trusted computing (as you suggest with weights, etc for the user to adjust) still would not work. Why? If the user can modify what the system can trust, so can anything malicious. All it takes is for the malware to be one step ahead of those writing lists - e.g. just like all viruses today. So no, that is not a working model - it works no better than what we have today, and in fact makes it worse since it creates a false sense of security as well.
Thus, my point that Trusted Computing is flawed still stands.
Also, if you notice, my initial reply ignored your weights, etc - replying to what you had before that. The point, nonetheless, is still able to be inclusive of it though as shown above.
Their description of Subversion is almost blatantly wrong, and misses much of the improvements SVN brought about. It would have helped them to at least have read some of the Subversion Documentation - or even just the chapter on Subversion's Delta Editor in the book Beautiful Code.
Nope. Distributed systems are ALWAYS more useful than centralized ones for source code control.
Only sometimes their advantages are not that significant.
I'd have to disagree. There are many instances were DVCS are always superior. However, there are times - such as in corporate environments - where you simply do not want that kind of information floating around the organization. In those instances a Centralized VCS is superior as the main advantage (distributed version control) is in those situations the biggest disadvantage of a DVCS.
Anything stored in the BIOS as executable code isn't going to be removed via a jumper or removal of battery. Reflashing the BIOS maybe but not simply clearing volatile BIOS flags.
BIOS is stored on a motherboard in two fashions: (1) a flashable (read/write) memory chip, and (2) a hard-wired (read-only) memory chip. When you update (flash) the BIOS, you only overwrite the flashable memory chip. This chip requires power to keep the BIOS data alive; the power comes from either the battery on the motherboard or the power supply. If you disconnect the system from the wall, and remove the battery from the motherboard then the flashable BIOS _will_ be reset. (Yes, I've reset BIOS's this way. I have several systems that I use to update the BIOS regularly on.) It'll happen faster if you disconnect the power supply from the motherboard as well - since residual energy could be in the capacitors in the power supply. Supposing all power is disconnected from the motherboard it takes about 30 seconds to drain all power and reset the BIOS.
Some motherboards have jumpers or dip-switch settings to do the same thing automatically - via hardware.
All of this, of course, only reset the BIOS back to what was shipped from the factory - what is stored on the hard-wired memory chip.
Now, how many people bother with updating the motherboard BIOS now-a-days? No many, thus it may only be the settings in that case. But this will clear viruses out of the BIOS. It'll also reset the BIOS in case of a bad update.
Some motherboards even come with a dual-bios system where only one can be written to and it is non-executable, and the other can only be read and executed. The active BIOS (the executable one) has a number of safety checks in it to try to prevent a BIOS virus from taking hold. Though not impossible to put a virus on these motherboards, it is a lot more difficult to do.
But how that 99% of society wants to use the computer should not ( and cannot necessarily) be dictated by even the 1% as the 1% will not know every edge case for how the 99% wants to use the computer.
Actually 99% of users will probably never do anything that would even be an issue. Malware primarily runs because users are not informed by the OS that it is malware or told that it is accessing their address book and starting a mail server or constantly spamming traffic at an address in Estonia. For the other 1% of cases the user needs the option to override the security system, but this should never be needed for normal use cases so when an app requests this it should be a red flag to users. Right now they're so conditioned by our poor OS UIs they just click through things. But if a users was never, ever (over the course of owning a machine and later over their lifetime) asked o override security and they were asked at some point with language worded to say doing so would allow someone else control of their computer forever, I think that would make a huge difference, don't you?
Who mentioned malware? Yes, malware is one thing that needs protected against; however, what a user wants to do with the computer may not necessarily be what the person controlling the system wants the user to do with the computer. This is almost fine in a corporate environment where the computer uses are dictated by the organization - though even then, that doesn't quite work as managers hire people to do things that IT didn't account for in their 'standard platform' (e.g. developers). However, one person (e.g. Microsoft, gov't, gov't agency, etc.) controlling what all users - including all corporations - can do with the computers doesn't work. Thus 'trusted' computing is flawed in the current model, and you don't even need to consider malware/viruses/etc to come to that analysis.
Have there actually been any MBR "bootkits" in the wild that have used flashable BIOS for storing copies? I always though that was a malware "urban legend". And shouldn't any flashable BIOS have some sort of jumper switch to prevent unauthorized flashing to being with?
Yes there are, and the symptoms are hard to relate. It's things like the PS/2 mouse won't be detected, or the floppy drive won't work right. Had one on my desktop back in college - only virus I ever had. And yes, the only way to get rid of them. Variants of the Monkey virus do store themselves into the BIOS.
Yeah, you could do all that, or you could just use MBRFIX and be done with it. "Extremely difficult" - I don't really think so. Typing 6 letters from a command prompt isn't something I would categorize as "extremely difficult".
Problem there - you could have the virus stored elsewhere on the computer. Many of those kinds of viruses will put themselves into start-up software as well; so running MBRFIX will remove it from the MBR yes, but then you'll re-infect the MBR on first boot.
And if you run MBRFIX from a machine that is infected - without booting from a CD first - then the virus will be in memory and just re-infect the MBR before you reboot.
So yes, it's not as simple as running MBRFIX - you actually have to disinfect the system too, which means eliminating any points where the infection may be - thus removing the battery on the motherboard to reset the BIOS, running the disk through an AV program on _another_ computer (or via a CD on the same computer), and fixing the MBR.
Of course, but the primary role of that lock down was to protect their DRM'd subsystems
In other words, the protection is there in order to prevent malicious code from stopping?
And turn over the keys to the RIAA and MPAA. Malicious AND Evil. Cthulhu would be pleased.
Old sk00l. When was the last MBR infector seen in the wild? 2002? Most of this class are from the DOS era, fercryingoutloud.
From the second paragraph of the fine article (emphasis added):
TDSS has been causing serious trouble for users for more than two years now, and is an example of a particularly pernicious type of rootkit that infects the master boot record of a PC. This type of malware often is referred to as a bootkit and can be extremely difficult to remove once it's detected. The older versions of TDSS--TDL1, TDL2 and TDL3--are detected by most antimalware suites now, but it's TDL4 that's the most problematic right now.
Easy way to remove it: (1) turn on the computer, (2) load the hard drives into another computer, scan, and dis-infect without loading the software on the computer - it should be read-write without execution permission; and finally (3) reset the firmware on the devices, especially the motherboard, on the computer. (Typically just the motherboard; but there could be other devices too that may need it.). Put it back-together, boot up, and your infection free.
OR
You could just run Linux/BSD/etc to start with and not have to worry about it.
Give me "trusted computing" where I control the keys and decide what software is "trusted" and I'd be fine w/ it.
The problem is, 99% of our society cannot properly decide whether software should be trusted or not, and even with more granular controls and proper feedback from the OS a lot of malware will slip through.
I don't think this is an unsolvable problem.
But how that 99% of society wants to use the computer should not ( and cannot necessarily) be dictated by even the 1% as the 1% will not know every edge case for how the 99% wants to use the computer. Thereby, "trusted" computing in that model is 100% flawed, and you then have to build in backdoors - like the register key that can disable requiring a signed driver so developers can test their drivers - so that the 99% can all do what they want/need to on the computer.
Yes, he's the CEO of Microsoft. No it is NOT an important technology company.
Didn't another CEO get in trouble over having a Twitter account and other social media? Perhaps the SEC needs to investigate Microsoft over the matter since Ballmer could potentially be giving advise or hinting at matters and cause insider trading to occur, even if unintentional.
Yes, though typically only for specialty environments. Intel still sells the 80086 processor. You can still get the i386 processor too, and boards that support them. But again, you're dealing with specialty not consumer products.
Well, Microsoft may not be around in 5 to 10 years; so that may prompt a software change in itself. Though I do agree, it's going to be hardware pushing some of the changes - namely the 32-bit to 64-bit or even to a 128-bit change. Software has increasingly become less relevant to speed and memory changes as the basic speeds and memory available are more than sufficient for 99% of tasks; with that 1% being the gaming and multimedia editing (yes big markets, but not very big compared to the rest of the entire computing industry).
Probably for quite a while still. If you can find it in retail, chances are you have a better chance of finding it OEM - especially for things that are less often modified (e.g. USB->Serial converters, etc.) so they have longer life cycles.
WinXP x64 while available was not used very much and was known for having all kinds of problems. However, you don't need the 64-bit WinXP version to do what I was saying - PAE (Physical Address Extension) is something from the processor level that allows addressing +4GB of memory. Linux used it a long time ago; but Windows didn't until WinXP - and even then you had to have the right version for a long time. MS may have enabled it by default for WinXP SP3, not sure. But any how, it's available. Fortunately Vista 64 is a lot better than WinXP x64; though there are still some driver issues. Win7 improved on that I believe. And Win8 or Win9 probably won't have a 32-bit edition.
WinXP can see more than 2 GB. We have systems that use 3 GB; we'd put 4 GB in but there's something wrong with the BIOS so it only sees 3GB. I am aware of other XP systems that have 4 GB, and with PAE it can see more than 4GB too.
Of course, Windows still takes up 1/3rd of the RAM for itself. but that's a different issue.
You mean like the 16-bit or 8-bit support disappeared? You do realize that even the latest x86/AMD64 processors still support the original 80086 processor modes, right? That that is their _default_ state of running when power is applied?
Also, vendors seem to be continuing support for XP even as Microsoft drops it. So where is this limited hardware support you speak of? Not to mention that it's only been in the last few years that support for Win9x was almost completely dropped - yes, you can still find hardware that officially supports Win9x/Me, and that's a lot older than XP.
Currently. But it's been falling. Apple+Linux adds up to nearly 10% now; when MS use to command nearly 95-98%. And guess what? It's growing as 1 in 3 PCs are non-Microsoft.
How are you counting? As with Vista, most Win7 licenses have the XP downgrading rights attached. How many that purchased it are still using Win7? Versus how many have downgraded to XP? Most probably. Especially since there's a nasty little interop issue with Win7 and older systems for file sharing - due to changing the encryption method. Yes you can disable it, but it takes quite a while to figure out what exactly you need to do. The few Win7 PCs we had at work only stayed after one server system was upgraded to a newer more compatible version of Windows (due only to the death of the server it replaced). Before that, all Win7 systems that came in got downgraded to XP as it was just easier.
Also, Vista and Win7 licenses are returning lower profits for a company that depends on the high profit margins. Add the lower sales and profits of Office into the mix and you have a company with a major problem. It may still take them a while to bleed to death, but there's a cannonball sized whole through their corporate torso that some flesh eating bacteria is infecting.
Huh? Wii has the top sales. PS3 is doing lowsy, but Xbox isn't doing much better. We should see at least another generation from Nintendo, who basically changed the market they were targeting. Sony and Microsoft will try another generation of PlayStation/XBox respectively but there's not much life left there as the money is just drying up.
A bit early? Possible. Though there are very few OEMs supporting it, and Microsoft has to blackmail OEMs to use it by telling them it'll be cheaper to use WP7 than to use Android and pay MS for the patent rights that MS is claiming they need. Yeah failure coming around the bend there too.
Of course, you could have been beegin sarcastic - which I was hoping for, in which case people better mod you funny; but some how I doubt it.
++True
XP mode is a virtual machine inside your windows 7 installation. Once installed, the user sees the application no differently than any other app, but when it is launched, it runs seamlessly in the VM.
That would depend on the application. I know of one application I use to work on that was a server, ran in the system tray, provided a GUI interface, and was installed a Windows Service. Guess what? Starting with Windows Vista all services are no longer allowed to have a GUI interface. Good luck integrating that with Vista/7; and I very much doubt it would work well under XP mode too. Why? We had problems with the app under normal WinXP when terminal services was used - guess where the service was told its GUI interface should go? Guess where the user got sent? Yep, not the same console; could we hard-code it? No - the user's console was indeterminate.
Oh, and it would have been a major rewrite to move the application into a server+admin-gui architecture like it should have been.)
Yes, the majority of user applications will probably run fine; but not all of them. Any IT administrator worth their salt would know they need to test every application that their users need to perform their job before doing a roll-out, and yes - that goes beyond what would be part of the official company standard. The larger the organization, the more software that has to get tracked and verified, so at the very least users can be notified that a problem will exist whereby they may need to purchase new software - either newer versions or move to another product, if that is even possible.
So using TFA's numbers, which is it? Original run-time would have been: 1m^3 = 1,000,000,000,000,000,000 -> 1,000,000,000,000 million -> 1,000,000 million million -> 1 million million million
But which is the new run-time? it's vastly different depending on order of operation:
(1000000*log(1000000))^2 -> (1000000*6)^2 -> 36,000,000,000,000 ->36,000,000M->36 million million
1000000*(log(1000000)^2) 1000000*(6^2) -> 1000000*36 = 36 million
Unless it was the politicians that were going...
I've thought OpenOffice was much better at opening older versions of Word (especially 97 versions) than Office . It seemed something changed after Office 2003. But that's just my perception.
OpenOffice writes better Microsoft Office documents than Microsoft Office does. Whether they legacy binary formats, or the newer XML formats. I've taken to using OO Writer for most things, and then ensuring everything is still there when I make the Word version, though it's mostly just the cross-references that I have to redo for some reason.
All due to a super huge effort to (i) reduce dependencies, and (ii) eliminate circular dependencies between kernel and user space. The effort started with Vista, and Win7 is a lot farther along. Win8 (whatever it will be) should do better yet. Though I believe that they already got to the point where dependencies only go one way with Win7 - that is, kernel space does not depend on anything in user space as of Win7. (If not Win7, certainly Win8 should be that way.)
We all trash Microsoft for making shitty products, but in the end we would all work for them given the chance.
I did a couple rounds of interviews with Microsoft just to be able to say I interviewed with them. I figured, if they offered the job, I could say I declined it as well.
While the interviews were fun, I have zero intention of working for Microsoft at any point in my career given their ethics, tactics, etc.
So no, I do not know of whom you speak.
However, pretty much every state in the US requires that the driver have 2 hands on the wheel. Not really enforced, mind you; but having an AI drive the vehicle with a human on "stand-by" would indeed still be illegal in the same fashion as driving with your leg, or letting the passenger hold the steering wheel for you, etc. They might not enforce it, but that doesn't mean its legal - just that the law is unenforceable and itself is illegal (since laws are required to be enforceable to be legal and good law - mind you a lot of laws do not measure up to that standard).