Gee, that would be a stretch...
on
IE To Block Pop-Ups
·
· Score: 2, Insightful
They were turned down, so now watch what happens, on Longhorn there will already be a "MSN Search Deskbar" on bootup.
Keep in mind that Microsoft has had built-in internet searching via the Windows Shell through the Start Menu's "Search..." function for a few years now, so I don't think this is a huge surprise for anybody.
I also don't think it takes an incredible leap of imagination to think they might put an "Internet Search" control on the new Longhorn Sidebar. Claiming that by doing this they're trying to defeat Google's impending grasp on the desktop is a bit paranoid.
I mean, what else would they put on the new sidebar? A big clock?:-)
Bus-driving is a special skill that requires specific training. If you've only driven an automatic-transmission Camry and you hop behind the wheel on a Greyhound, you'd probably kill someone.
Many, MANY more people get killed on or by buses every year than on or by planes. And plane flight is much more automated that bus-driving, which is still completely manually controlled at every instance.
SO, by the above reasoning bus drivers are either vastly undercompensated in our society or major-airline pilots vastly overcompensated.
Flying any aircraft is so much more challenging than driving a bus that I don't know where to begin. To name a few challenges that do NOT have corresponding issues on the bus:
Landing, landing, landing (unless parking a bus is a challenge for you)
Engine failure (buses just coast to a stop)
Navigation (Ever had a bus wind up 50 miles off course because the prevailing winds changed direction? Didn't think so...)
Altitude effects like hypoxia (unless your bus happens to be driving up to Mount Everest Base camp)
Gas turbine engines (when's the last time you saw a bus with one of these?)
Weight and balance (ever had to shuffle passengers around on a bus to prevent a takeoff crash?)
And I haven't even scratched the surface! I could go into the thousands of emergency scenarios that bus drivers don't even DREAM about, but for which pilots must study and undergo training every few months using a full-motion simulator.
Don't get me wrong: some of the top-level United pilots just before United Airlines filed Chapter 11 were making ridiculous amounts of money, and it contributed to the company's restructuring. You can thank their stupid union for that ridiculousness (ALPA). But reducing their profession to "glorified bus driver" status is simply unenlightened balderdash.
Let me ask you this: does that mean any new feature added in Windows is now an anti-trust violation? Doesn't that seem a little harsh?
Or perhaps you are just objecting to the fact that Macromedia already has similar capabilities. The problem is, just about any feature you add to an OS today has been done by someone before. Does that mean that the OS must be stagnant?
It's not as though Macromedia has the patent on vector-based graphics...
It is also that Windows runs a ton of stupid, random crap in kernel space.
Like Windows Media Player. Like Internet Explorer. Like Outlook. Like a ton of office stuff.
This is one of the most blatantly false statements I have seen get modded up to +4 or +5 in a long, long time.
Windows Media Player, Internet Explorer, and Outlook do NOT run in kernel mode whatsoever. They may talk to kernel-mode drivers like 95% of all user-mode software does (read from a file, talk to the network), but they absolutely do not run in kernel-mode!
C'mon, people. If you want to bash MS, you can do better than make up ridiculous statements like that.
Please correct me if I'm wrong, but the first sentence in the article clearly states that "Portable Media Center 2004" is a piece of software, not a new hardware device:
MICROSOFT is jumping into the portable media market, launching software for a new line of mobile devices designed to free digital content from a computer and play music, videos and photos on the go.
It sounds like MS is developing platform software for such devices and will license the platform to the actual hardware vendors (Dell, iRiver, SonicBlue, etc). This seems similar to the kind of relationship Microsoft has with its Smartphone manufacturers: Microsoft supplies the software, Motorola et al. supply the gadgets.
iTunes Music Store sells AAC, but does not lock you in to buying ONLY AAC.
Yes it does. No online music retailer in its right mind would use non-DRM music formats today. And iTunes supports only one: AAC. So if you're buying licensed online music and using iTunes, you're using AAC. And since only one online store sells AAC music files compatible with iTunes, that's what you will use: iTunes Music Store. When MusicMatch actually releases something, then you can say you have a choice.
It doesn't even lock you into using iTunes (to play your music), as the DRM lives in QuickTime, so pretty much any software which supports QuickTime 6 can play iTMS AAC files.
Gee, that's a good defense of iTunes! Critic: "iTunes is too restrictive, users would better be served elsewhere to get more choices." Respondent: "What do you mean it's too restrictive? You don't have to use iTunes to listen to your music at all!" Critic: "Uh, that's exactly what we're saying... don't use iTunes and you won't be restricted..."
iTunes plays practically everything BUT WMA... so iTunes will happily work with any source of MP3, AAC, WAV, AIFF, etc. files.
By listing the non-DRM formats that iTunes supports, you are dodging the issue at hand: online music distribution formats. Sites like Napster et. al. could not care less whether iTunes plays AIFF or u-law files (or even, dare I say, Mp3 files).
Online music retailers care only about DRM-ready file formats that prevent copying. For iTunes, this format is AAC. The days of online Mp3 sales are numbered, because record companies hate allowing kids to copy away and share at will.
To understand Microsoft's argument, you need answer only two questions:
Name some companies (other than Apple) that sell AAC music online for iTunes.
Name some devices other than iPod that sync with iTunes.
All of a sudden, the choices seem limited from my perspective...
If you want to buy music from the store then you're limited to iTunes (or Quicktime) and the iPod but how is this different from buying DRM-ed music in WMA format? You're limited to Windows Media Player and a portable machine that supports it (ok, so there are more of those available).
You're missing Microsoft's point, like just about everyone else who has posted today. The point is this:
If you choose to use iTunes, you can ONLY buy online music from Apple's iTunes Music Store. This is the way it will always be, since iTunes is integrated with iTMS and iTMS is the only online music store selling properly licensed (DRM) conent in the iTunes/iPod AAC format.
If you use Windows Media Player, you will be able to buy online music from serveral different services, like Napster, BuyTunes, etc. Several different companies are distributing online music using the WMA format. Not to mention, you will be able to play back this content on thousands more electronic devices that support WMA.
That is the choice MS is talking about. The choice of online music retailer, and the choice of playback device. It's about marketplace competition to sell you a service that doesn't exist when you run iTunes.
Having iTunes available for Windows offers yet another choice for Windows users, and that's fine and dandy. But the MS point is still valid.
Which is curious, considering the fact that the AAC format is an industry standard, while the WMA format is purely a Microsoft format.
AAC is a proprietary format designed by Dolby Laboratories (Sony and Fraunhofer signed on after the fact) which requires a license to use. The licensing terms can be found here.
WMA is a proprietary format developed by Microsoft. It is a proven industry standard, already in use by thousands of consumer-level electronic devices. The licensing terms can be found here.
Note that AAC is considerably more expensive to use, and therefore isn't any more "open" than WMA. And given that WMA is used by thousands more devices than AAC, wouldn't it be considered the "industry standard" and not the other way around?
I'm not sure what the output you listed is trying to demonstrate, but if you are suggesting that PowerPC doesn't use interrupts then you are quite mistaken...
And if you are trying to point out that PowerPC uses a different mechanism rather than using Interrupt Request Lines, then I'd like to point out that Intel has also used a different mechanism ever since the Pentium (since 1993, which pre-dates PowerPC IIRC). But motherboard manufacturers are only now getting around to including I/O APICs on all their motherboards to use this interrupt mechanism. I/O APICs send interrupt messages to the CPU rather than signalling a particular line. But to maintain backwards compatibility, Intel decided to leave support for IRQ lines intact.
Interestingly, all multi-processor systems with Pentium processors must use I/O APICs, so it's not as though mobo manufacturers didn't know how to support them...
IRQ sharing occurs with Windows 2k/XP on older computers (older than P4/Athlon XP) with ACPI enabled.
First of all, this "information" is misleading. IRQ sharing is NORMAL. It does NOT mean you have conflicts among your devices. IRQ sharing is a feature of ACPI (Advanced Configuration and Power Interface) computers, and it is a capability of all modern computers that have an ACPI-compatible BIOS and devices. Sharing IRQs is intentional, and all hardware that is ACPI-compatible is required to support a shared IRQ environment. This does not involve a significant performance penalty: shared interrupts are still "triaged" in a first-come, first-served basis and handled according to their priority. The "share" is resolved by having the driver query its device to see if it was the one requesting the IRQ. As long as the driver was written properly, this works out just fine. For reference, see this Anandtech FAQ.
Some computers have something called an I/O APIC (Advanced Programmable Interrupt Controller), traditionally only present on multi-CPU systems, that allows Win2k and WinXP to assign IRQ's above 15 to each device on the system, essentially giving every device its own "virtual" IRQ. If your computer has one of these, you probably won't see IRQ sharing, but there is no guarantee of that. For example, I type this on a brand new Compaq Evo that has an I/O Apic, and my ATI Radeon 7000 is sharing IRQ 18 with one of the USB Host Controllers. But like I said, they know how to work together, so it's all good, and I have no problems. The easiest way to tell if you have an I/O APIC is to go to Device Manager and expand the "computer" entry: if it says "ACPI Uniprocessor PC" rather than "Advanced Configuration and Power Interface PC", then you have an I/O APIC. All multi-processor computers have an I/O APIC because the I/O subsystem has to figure out a way to distribute interrupts between the multiple processors.
If you follow some people's advice, they will tell you to ditch ACPI mode all together and go back to "Standard PC" mode where each device MUST have its own IRQ or things don't work properly. This is not a good idea. Some systems have too many devices to assign a separate IRQ to each device, and this is part of the reason why IRQ sharing was part of the ACPI requirements. Not only that, but ripping out the HAL that sits underneath Windows NT is a tricky operation to perform correctly without screwing up your system.
Instead, you should check to see if your computer has any BIOS Updates, and if your hardware devices (especially any PCI cards) have new drivers. These are generally the main reasons that IRQ sharing might be causing problems, if that indeed is the real culprit.
Re:It's like Googling your hard drive
on
CNet on WinFS
·
· Score: 1
LOL. Like Microsoft cares what a typical Linux user wants. That's funny.
In fact, I bet they have people working in shifts working on your request right now. I bet that's their next big marketing blitz: Longhorn, now with arcane Unix command-line tools! Can you imagine the sales boost that will give them? An easy 0.001% increase in market share, right there!
BTW: using dir with findstr will work on a Longhorn console just like it does on 2k and XP. And whaddya bet that WinFS itself is a service that can be disabled if you don't want it? Hmmm... seems like that would pretty much nullify your criticism.
It's like Googling your hard drive
on
CNet on WinFS
·
· Score: 1
This seems classic MS-predictive-FUD, where people hold their breath for the Next Release, which is a 1.0 version that sucks.
While I don't claim to know what version 1.0 of WinFS will be like, I don't think the idea of executing a system-level "metadata" search is a bad idea at all.
A filesystem is basically a clumsy database, with the fields being rather arbitrary "directories" and "filenames." The concept of a directory is not intuitive for computer illiterates at all. Allowing users to seach for their data, without worrying about the concept of where it physically (or even logically) resides, is a cool idea.
Lots of regular computer users don't understand the concept of directories and a directory hierarchy. But just about EVERYONE understands the idea of a Google search. I think that is the underlying principle behind this push.
I completely disagree. Joe User does NOT complain about patches. Joe User doesn't even know that patches exist, and this is a demonstrable (if unfortunate) fact.
You're vastly over-estimating the vast majority of computer users out there. Anyone earmarked for installation of patches on business machines has enough training to be considered an "IT Guy." Maybe not a Pro. But of course, anyone performing this operation on a number of machines is exactly the audience Microsoft is attempting to satisfy by rolling out patches on a monthly basis.
Why on earth do you think they sat on 4 critical updates until the middle of the month, and then issued a press release simultaneously mentioning that they were switching to a monthly schedule? You think they're ignoring their primary business customers? Hardly.
This is a classic case of being damned if they do, and damned if they don't. If they release patches as they come out, people say "Ooooh, oooh, my poor rollout schedule, I just got done deploying the last patch and here comes another just to spite me!" If they release patches monthly, people complain "Oooh, oooh Microsoft is witholding critical updates and making me insecure!" I'm tired of all this freaking complaining.
Joe User doesn't complain about the frequency of patches. IT pros are the ones who bitch about the frequency of patches. In this case, Microsoft is absolutely responding to pressure from its large customers.
When the CTO of a Fortune 50 company calls up Steve Ballmer and says, "How are you going to compensate us for all this time we're wasting deploying patches from you every other week?" you can bet that MS is going to come up with a way to ease that burden, or lose another customer to Linux.
They're trying to ease the IT burden by aggregating the patches into monthly releases (whenever exploits aren't already present) so that Admins have adequate time in between releases for testing, deployment, and preparation for the next batch of updates. It's a queueing mechanism, essentially.
Not adding Hebrew language support in a product available in Isreal is what isn't making strong economic sense I think.
Office Mac has always been available in Israel. Office Mac has never had support for Hebrew. People know this, and buy it anyway: it's their choice. At what point exactly did native language support for a country become a necessity to sell a product there?
So Yeda offered to take the job upon themsleves. MS does not have to pay anything to add Hebrew. Why does MS still refuse? Could it be they like their Windows platform better than Mac-OS?
Could it be that they don't want to allow 3rd party developers to modify their code and release something with the Microsoft logo on it?
Let me turn around the situation for you: Do you think Apple would allow 3rd party developers to add some features to OS X that a foreign government was requesting, and release that product with full Apple support? I think not.
Foolish of Microsoft to resist such an upgrade to their own software.
Feature management is a very, very simple concept. You look at what percent of your customers are demanding this or that feature, and what bugs are affecting X percentage of your customers, and make your decisions about how to allocate development resources from there.
With Apple having a sub 5% market share worldwide, MS Office being on perhaps half of those, and perhaps less than 1% of the worldwide Apples requesting Hebrew support, we're talking about a very small customer base here requesting this feature. Especially if there is a bug fix that is affecting 90% of your customers that hasn't been addressed yet.
If this article is true, the Israeli government seems to be strong-arming Microsoft to implement a feature that doesn't make economic sense. What's worse, according to the article, they are suspending MS contracts that have already been signed. I am no contract lawyer, but to me, that sounds very abusive and arbitrary, and would be illegal if anyone but the government were to attempt such an action...
Don't knock the NT kernel
on
Longhorn in 2006
·
· Score: 3, Insightful
Want to shut down Linux users? Write your next OS on a BSD kernel, make the old Windows apps work the way people want them to (it can be done... it's BEING done) and sell it to people. They will buy it because there are people out there who still trust you for some reason. Once you out out something with a *NIX kernel, you will see an amazing amount of curiosity and popularity.
This is rather amusing, because it points out an odd trend amongst "technophiles" in computing today. Somehow or other, *NIX kernels have become synonymous with "software excellence." When this trend started is not entirely clear to me, but I'd say post-1995 for sure. If it is indeed a FUD campaign, it seems to be succeeding, because 10 years ago if anyone had mentioned that *NIX kernels were superior to modern OS multi-threaded microkernels they would have been laughed into submission.
BSD-style *NIX kernels are NOT, contrary to what you may have heard, the end-all be-all of OS kernel design. In fact, most people who architect operating systems for a living will tell you that most of the concepts contained therein are good ideas, but they're somewhat stale and in need of some serious revision.
I don't have the time or the inclination to go through a feature-by-feature comparison between a modern *NIX kernel and NT, but I'll point out a few examples. The NT kernel's native support for threads and access control list kernel object security are superior to what the *BSD kernels offer. Other newer features like microarchitecture to support several different system call APIs are virtually on par feature for feature with *BSD.
So why would switching the kernel make the OS any better? If a kernel has the necessary features it requires, performs well, and provides remarkable stability, that's just about all that a kernel can do.
I think you're confusing the recent security problems discovered in the Windows system with problems in the kernel itself, which are few and far between. Holes in IIS, or SQL Server, or even the "RPC System Service" are NOT problems with the NT kernel, and they should not be confused with them.
Don't misunderstand me: I think the number of features that have been integrated into the Unix framework over the years (by the Linux and *BSD projects) is astounding and a telling tribute to what the research communities can accomplish when they work together. But that doesn't mean they're superior to what alternative OS kernels can do.
And did I mention that trivial bugs needn't be fatal flaws if the kernel enforces proper user level security? If I hadn't, then I will say it now. Trivial bugs needn't be fatal flaws if the kernel enforces proper user level security!!!!
I don't even know how to address this one. The NT kernel does MUCH more for security than any *NIX kernel. The trick is, people writing software that runs on the kernel have to make USE of these features properly. NT offers complete Access Control Lists and security descriptors for every possible kernel object. This is just about as granular as you can get, and better than the simple "rwx" permissions on file descriptors available in *NIX kernels.
Now, why everyone logs into the Windows Shell with a superuser security account is an entirely different matter, but it is NOT the kernel's fault!
You can run few, if any, games, many applications that weren't designed to work on NT will fail, and so on.
How is this the fault of Windows? Every single application that Microsoft releases works with restricted logins, and that is not by accident. If you have applications that do not, I would suggest contacting the manufacturer and letting them know they have a horrible security problem.
Plus, Windows XP pretty much encourages use of the Administrator account
Now that *is* a legitimate gripe about a default install of Windows. In my opinion, it should force you to add a restricted user account during installation. (But note that most linux distributions do not force you to do this either...) Luckily, it is easy to add a restricted user to the system by using the incredibly simple "User Accounts" control panel applet.
But anyone who knows about the control panel can easily add a restricted user account. It takes:
4 clicks (Start->Control Panel->User Accounts->Create an account)
typing in the user account name
2 more clicks (Limited Account->Create Account)
and you're done! What's more, setting an administrator password is just as easy...
A rootkit is injected manually by a human being AFTER they burrow through an exploit.
His point is that those exploits that allow rootkits were obviously not counted, and the "perhaps 40 for Linux" statistic is therefore entirely misleading when taken out of context, like it was.
The other misleading issue about the 60,000 Windows viruses statistic is that it includes every variation of any virus written for Windows since Windows 1.0. After eliminating variations on a theme and duplicates, the total number of unique exploitable holes in Windows is some smaller number. And nearly ALL of those are eliminated when using a permissions-based filesystem like NTFS and restricted user logins, both of which have mainstream since NT 4.0 in 1996.
This "build group" would still work for Microsoft, not for the government or local companies?
Exactly. It would be a Microsoft group that is overseen by the security audit team from the Chinese government. The Chinese officials would be there to verify that the builds are indeed coming from the same codebase their "security experts" had inspected.
Ensuring that the code being built has already been inspected can be accomplished in any number of ways, but signing the code using some sort of cryptographic hash during transportation from one machine to another is an obvious solution. With suitable oversight, such a build process can be made very secure.
My point is just that there are many ways this agreement could be accomplished, and no one here can say what the details are for sure. Assuming that the Chinese government doesn't know about these potential security loopholes is underestimating their intelligence a bit, no?
Slashdot Mirror
Keep in mind that Microsoft has had built-in internet searching via the Windows Shell through the Start Menu's "Search..." function for a few years now, so I don't think this is a huge surprise for anybody.
I also don't think it takes an incredible leap of imagination to think they might put an "Internet Search" control on the new Longhorn Sidebar. Claiming that by doing this they're trying to defeat Google's impending grasp on the desktop is a bit paranoid.
I mean, what else would they put on the new sidebar? A big clock? :-)
Flying any aircraft is so much more challenging than driving a bus that I don't know where to begin. To name a few challenges that do NOT have corresponding issues on the bus:
- Landing, landing, landing (unless parking a bus is a challenge for you)
- Engine failure (buses just coast to a stop)
- Navigation (Ever had a bus wind up 50 miles off course because the prevailing winds changed direction? Didn't think so...)
- Altitude effects like hypoxia (unless your bus happens to be driving up to Mount Everest Base camp)
- Gas turbine engines (when's the last time you saw a bus with one of these?)
- Weight and balance (ever had to shuffle passengers around on a bus to prevent a takeoff crash?)
And I haven't even scratched the surface! I could go into the thousands of emergency scenarios that bus drivers don't even DREAM about, but for which pilots must study and undergo training every few months using a full-motion simulator.Don't get me wrong: some of the top-level United pilots just before United Airlines filed Chapter 11 were making ridiculous amounts of money, and it contributed to the company's restructuring. You can thank their stupid union for that ridiculousness (ALPA). But reducing their profession to "glorified bus driver" status is simply unenlightened balderdash.
Let me ask you this: does that mean any new feature added in Windows is now an anti-trust violation? Doesn't that seem a little harsh?
Or perhaps you are just objecting to the fact that Macromedia already has similar capabilities. The problem is, just about any feature you add to an OS today has been done by someone before. Does that mean that the OS must be stagnant?
It's not as though Macromedia has the patent on vector-based graphics...
This is one of the most blatantly false statements I have seen get modded up to +4 or +5 in a long, long time.
Windows Media Player, Internet Explorer, and Outlook do NOT run in kernel mode whatsoever. They may talk to kernel-mode drivers like 95% of all user-mode software does (read from a file, talk to the network), but they absolutely do not run in kernel-mode!
C'mon, people. If you want to bash MS, you can do better than make up ridiculous statements like that.
Please correct me if I'm wrong, but the first sentence in the article clearly states that "Portable Media Center 2004" is a piece of software, not a new hardware device:
It sounds like MS is developing platform software for such devices and will license the platform to the actual hardware vendors (Dell, iRiver, SonicBlue, etc). This seems similar to the kind of relationship Microsoft has with its Smartphone manufacturers: Microsoft supplies the software, Motorola et al. supply the gadgets.
Wow, that was a good one. I love how people like you make constructive and rational arguments, rather than resorting to puerile name-calling.
Keep it up, and your viewpoint will surely be endorsed!
Yes it does. No online music retailer in its right mind would use non-DRM music formats today. And iTunes supports only one: AAC. So if you're buying licensed online music and using iTunes, you're using AAC. And since only one online store sells AAC music files compatible with iTunes, that's what you will use: iTunes Music Store. When MusicMatch actually releases something, then you can say you have a choice.
Gee, that's a good defense of iTunes! Critic: "iTunes is too restrictive, users would better be served elsewhere to get more choices." Respondent: "What do you mean it's too restrictive? You don't have to use iTunes to listen to your music at all!" Critic: "Uh, that's exactly what we're saying... don't use iTunes and you won't be restricted..."
By listing the non-DRM formats that iTunes supports, you are dodging the issue at hand: online music distribution formats. Sites like Napster et. al. could not care less whether iTunes plays AIFF or u-law files (or even, dare I say, Mp3 files).
Online music retailers care only about DRM-ready file formats that prevent copying. For iTunes, this format is AAC. The days of online Mp3 sales are numbered, because record companies hate allowing kids to copy away and share at will.
To understand Microsoft's argument, you need answer only two questions:
All of a sudden, the choices seem limited from my perspective...
You're missing Microsoft's point, like just about everyone else who has posted today. The point is this:
If you choose to use iTunes, you can ONLY buy online music from Apple's iTunes Music Store. This is the way it will always be, since iTunes is integrated with iTMS and iTMS is the only online music store selling properly licensed (DRM) conent in the iTunes/iPod AAC format.
If you use Windows Media Player, you will be able to buy online music from serveral different services, like Napster, BuyTunes, etc. Several different companies are distributing online music using the WMA format. Not to mention, you will be able to play back this content on thousands more electronic devices that support WMA.
That is the choice MS is talking about. The choice of online music retailer, and the choice of playback device. It's about marketplace competition to sell you a service that doesn't exist when you run iTunes.
Having iTunes available for Windows offers yet another choice for Windows users, and that's fine and dandy. But the MS point is still valid.
AAC is a proprietary format designed by Dolby Laboratories (Sony and Fraunhofer signed on after the fact) which requires a license to use. The licensing terms can be found here.
WMA is a proprietary format developed by Microsoft. It is a proven industry standard, already in use by thousands of consumer-level electronic devices. The licensing terms can be found here.
Note that AAC is considerably more expensive to use, and therefore isn't any more "open" than WMA. And given that WMA is used by thousands more devices than AAC, wouldn't it be considered the "industry standard" and not the other way around?
Odd, considering that software has no effect on whether or not a CD drive can read a scratched CD...
I'm not sure what the output you listed is trying to demonstrate, but if you are suggesting that PowerPC doesn't use interrupts then you are quite mistaken...
And if you are trying to point out that PowerPC uses a different mechanism rather than using Interrupt Request Lines, then I'd like to point out that Intel has also used a different mechanism ever since the Pentium (since 1993, which pre-dates PowerPC IIRC). But motherboard manufacturers are only now getting around to including I/O APICs on all their motherboards to use this interrupt mechanism. I/O APICs send interrupt messages to the CPU rather than signalling a particular line. But to maintain backwards compatibility, Intel decided to leave support for IRQ lines intact.
Interestingly, all multi-processor systems with Pentium processors must use I/O APICs, so it's not as though mobo manufacturers didn't know how to support them...
First of all, this "information" is misleading. IRQ sharing is NORMAL. It does NOT mean you have conflicts among your devices. IRQ sharing is a feature of ACPI (Advanced Configuration and Power Interface) computers, and it is a capability of all modern computers that have an ACPI-compatible BIOS and devices. Sharing IRQs is intentional, and all hardware that is ACPI-compatible is required to support a shared IRQ environment. This does not involve a significant performance penalty: shared interrupts are still "triaged" in a first-come, first-served basis and handled according to their priority. The "share" is resolved by having the driver query its device to see if it was the one requesting the IRQ. As long as the driver was written properly, this works out just fine. For reference, see this Anandtech FAQ.
Some computers have something called an I/O APIC (Advanced Programmable Interrupt Controller), traditionally only present on multi-CPU systems, that allows Win2k and WinXP to assign IRQ's above 15 to each device on the system, essentially giving every device its own "virtual" IRQ. If your computer has one of these, you probably won't see IRQ sharing, but there is no guarantee of that. For example, I type this on a brand new Compaq Evo that has an I/O Apic, and my ATI Radeon 7000 is sharing IRQ 18 with one of the USB Host Controllers. But like I said, they know how to work together, so it's all good, and I have no problems. The easiest way to tell if you have an I/O APIC is to go to Device Manager and expand the "computer" entry: if it says "ACPI Uniprocessor PC" rather than "Advanced Configuration and Power Interface PC", then you have an I/O APIC. All multi-processor computers have an I/O APIC because the I/O subsystem has to figure out a way to distribute interrupts between the multiple processors.
If you follow some people's advice, they will tell you to ditch ACPI mode all together and go back to "Standard PC" mode where each device MUST have its own IRQ or things don't work properly. This is not a good idea. Some systems have too many devices to assign a separate IRQ to each device, and this is part of the reason why IRQ sharing was part of the ACPI requirements. Not only that, but ripping out the HAL that sits underneath Windows NT is a tricky operation to perform correctly without screwing up your system.
Instead, you should check to see if your computer has any BIOS Updates, and if your hardware devices (especially any PCI cards) have new drivers. These are generally the main reasons that IRQ sharing might be causing problems, if that indeed is the real culprit.
LOL. Like Microsoft cares what a typical Linux user wants. That's funny.
In fact, I bet they have people working in shifts working on your request right now. I bet that's their next big marketing blitz: Longhorn, now with arcane Unix command-line tools! Can you imagine the sales boost that will give them? An easy 0.001% increase in market share, right there!
BTW: using dir with findstr will work on a Longhorn console just like it does on 2k and XP. And whaddya bet that WinFS itself is a service that can be disabled if you don't want it? Hmmm... seems like that would pretty much nullify your criticism.
While I don't claim to know what version 1.0 of WinFS will be like, I don't think the idea of executing a system-level "metadata" search is a bad idea at all.
A filesystem is basically a clumsy database, with the fields being rather arbitrary "directories" and "filenames." The concept of a directory is not intuitive for computer illiterates at all. Allowing users to seach for their data, without worrying about the concept of where it physically (or even logically) resides, is a cool idea.
Lots of regular computer users don't understand the concept of directories and a directory hierarchy. But just about EVERYONE understands the idea of a Google search. I think that is the underlying principle behind this push.
I completely disagree. Joe User does NOT complain about patches. Joe User doesn't even know that patches exist, and this is a demonstrable (if unfortunate) fact.
You're vastly over-estimating the vast majority of computer users out there. Anyone earmarked for installation of patches on business machines has enough training to be considered an "IT Guy." Maybe not a Pro. But of course, anyone performing this operation on a number of machines is exactly the audience Microsoft is attempting to satisfy by rolling out patches on a monthly basis.
Why on earth do you think they sat on 4 critical updates until the middle of the month, and then issued a press release simultaneously mentioning that they were switching to a monthly schedule? You think they're ignoring their primary business customers? Hardly.
This is a classic case of being damned if they do, and damned if they don't. If they release patches as they come out, people say "Ooooh, oooh, my poor rollout schedule, I just got done deploying the last patch and here comes another just to spite me!" If they release patches monthly, people complain "Oooh, oooh Microsoft is witholding critical updates and making me insecure!" I'm tired of all this freaking complaining.
Joe User doesn't complain about the frequency of patches. IT pros are the ones who bitch about the frequency of patches. In this case, Microsoft is absolutely responding to pressure from its large customers.
When the CTO of a Fortune 50 company calls up Steve Ballmer and says, "How are you going to compensate us for all this time we're wasting deploying patches from you every other week?" you can bet that MS is going to come up with a way to ease that burden, or lose another customer to Linux.
They're trying to ease the IT burden by aggregating the patches into monthly releases (whenever exploits aren't already present) so that Admins have adequate time in between releases for testing, deployment, and preparation for the next batch of updates. It's a queueing mechanism, essentially.
Office Mac has always been available in Israel. Office Mac has never had support for Hebrew. People know this, and buy it anyway: it's their choice. At what point exactly did native language support for a country become a necessity to sell a product there?
Could it be that they don't want to allow 3rd party developers to modify their code and release something with the Microsoft logo on it?
Let me turn around the situation for you: Do you think Apple would allow 3rd party developers to add some features to OS X that a foreign government was requesting, and release that product with full Apple support? I think not.
Feature management is a very, very simple concept. You look at what percent of your customers are demanding this or that feature, and what bugs are affecting X percentage of your customers, and make your decisions about how to allocate development resources from there.
With Apple having a sub 5% market share worldwide, MS Office being on perhaps half of those, and perhaps less than 1% of the worldwide Apples requesting Hebrew support, we're talking about a very small customer base here requesting this feature. Especially if there is a bug fix that is affecting 90% of your customers that hasn't been addressed yet.
If this article is true, the Israeli government seems to be strong-arming Microsoft to implement a feature that doesn't make economic sense. What's worse, according to the article, they are suspending MS contracts that have already been signed. I am no contract lawyer, but to me, that sounds very abusive and arbitrary, and would be illegal if anyone but the government were to attempt such an action...
This is rather amusing, because it points out an odd trend amongst "technophiles" in computing today. Somehow or other, *NIX kernels have become synonymous with "software excellence." When this trend started is not entirely clear to me, but I'd say post-1995 for sure. If it is indeed a FUD campaign, it seems to be succeeding, because 10 years ago if anyone had mentioned that *NIX kernels were superior to modern OS multi-threaded microkernels they would have been laughed into submission.
BSD-style *NIX kernels are NOT, contrary to what you may have heard, the end-all be-all of OS kernel design. In fact, most people who architect operating systems for a living will tell you that most of the concepts contained therein are good ideas, but they're somewhat stale and in need of some serious revision.
I don't have the time or the inclination to go through a feature-by-feature comparison between a modern *NIX kernel and NT, but I'll point out a few examples. The NT kernel's native support for threads and access control list kernel object security are superior to what the *BSD kernels offer. Other newer features like microarchitecture to support several different system call APIs are virtually on par feature for feature with *BSD.
So why would switching the kernel make the OS any better? If a kernel has the necessary features it requires, performs well, and provides remarkable stability, that's just about all that a kernel can do.
I think you're confusing the recent security problems discovered in the Windows system with problems in the kernel itself, which are few and far between. Holes in IIS, or SQL Server, or even the "RPC System Service" are NOT problems with the NT kernel, and they should not be confused with them.
Don't misunderstand me: I think the number of features that have been integrated into the Unix framework over the years (by the Linux and *BSD projects) is astounding and a telling tribute to what the research communities can accomplish when they work together. But that doesn't mean they're superior to what alternative OS kernels can do.
And did I mention that trivial bugs needn't be fatal flaws if the kernel enforces proper user level security? If I hadn't, then I will say it now. Trivial bugs needn't be fatal flaws if the kernel enforces proper user level security!!!!
I don't even know how to address this one. The NT kernel does MUCH more for security than any *NIX kernel. The trick is, people writing software that runs on the kernel have to make USE of these features properly. NT offers complete Access Control Lists and security descriptors for every possible kernel object. This is just about as granular as you can get, and better than the simple "rwx" permissions on file descriptors available in *NIX kernels.
Now, why everyone logs into the Windows Shell with a superuser security account is an entirely different matter, but it is NOT the kernel's fault!
Umm... a jury already awarded Eolas $521 million in damages due to the patent infringement. What else do you want them to lose?
How is this the fault of Windows? Every single application that Microsoft releases works with restricted logins, and that is not by accident. If you have applications that do not, I would suggest contacting the manufacturer and letting them know they have a horrible security problem.
Now that *is* a legitimate gripe about a default install of Windows. In my opinion, it should force you to add a restricted user account during installation. (But note that most linux distributions do not force you to do this either...) Luckily, it is easy to add a restricted user to the system by using the incredibly simple "User Accounts" control panel applet.
But anyone who knows about the control panel can easily add a restricted user account. It takes:
- 4 clicks (Start->Control Panel->User Accounts->Create an account)
- typing in the user account name
- 2 more clicks (Limited Account->Create Account)
and you're done! What's more, setting an administrator password is just as easy...His point is that those exploits that allow rootkits were obviously not counted, and the "perhaps 40 for Linux" statistic is therefore entirely misleading when taken out of context, like it was.
The other misleading issue about the 60,000 Windows viruses statistic is that it includes every variation of any virus written for Windows since Windows 1.0. After eliminating variations on a theme and duplicates, the total number of unique exploitable holes in Windows is some smaller number. And nearly ALL of those are eliminated when using a permissions-based filesystem like NTFS and restricted user logins, both of which have mainstream since NT 4.0 in 1996.
Very nice personal attack. Did you learn that one in economics class, Troll?
Exactly. It would be a Microsoft group that is overseen by the security audit team from the Chinese government. The Chinese officials would be there to verify that the builds are indeed coming from the same codebase their "security experts" had inspected.
Ensuring that the code being built has already been inspected can be accomplished in any number of ways, but signing the code using some sort of cryptographic hash during transportation from one machine to another is an obvious solution. With suitable oversight, such a build process can be made very secure.
My point is just that there are many ways this agreement could be accomplished, and no one here can say what the details are for sure. Assuming that the Chinese government doesn't know about these potential security loopholes is underestimating their intelligence a bit, no?