Microsoft doesn't give you a compilable version of their code. That's the point.
Oh, really? So you have read and signed one of these "shared source" agreements with Microsoft in the past, then?
Microsoft already has a wing of MS Research in Beijing, if I recall correctly. Why would it be such a stretch of the imagination to have a "build group" over there that produces Chinese-only builds from the same source tree that is open for government inspection? How do you know this isn't what is being done?
Will this actually increase the interest in this seperatist group?
Absolutely!
Just look at what rock 'n roll music was in the USSR back in the 60s and 70s. It was a "politically subversive cultural influence" and therefore banned at the government level. But that only served to make groups like the Beatles ever more popular and mysterious.
Eh, NT is based of VMS, it's not from scratch. period.
The "similarities" between NT and VMS have been completely overblown. There are very few, and they are shared by just about any modern OS.
When you hire DEC engineers to architect a new product from scratch, you are going to see SOME similarities in the architecture with VMS. But that does NOT mean that the new product is "derived"!
I challenge you to put your OS architecture knowledge where your mouth is. Tell me all about the "derived" features that proves NT is a derivative of VMS and nothing more. I'll be waiting with bated breath.
you just do it better and viola', world domination.
This has never happened in the past... what makes you think it will start now?
of course that doesn't apply to microsoft. they didn't do it better, just marketed it better.
Microsoft didn't get to their position by marketing. Making that claim is simply being disingenuous. They rose to their current position through a combination of great timing, shrewd contract negotiation, and "good enough" products.
First they got a contract with IBM for DOS because they had an operating system to sell when others did not (or were unwilling). After that point, they rode the Intel x86 evolutional wave, improving their OS to basically keep up with Intel (with a sizable lag). No marketing was required: killer apps were written (Lotus 1-2-3 and Wordperfect come to mind), and history was made.
But even though their historical products achieved success mostly through timing genius, I think NT represents a new chapter for MS and shows that they DO know how to create a good product from scratch. Even if they DID have to hire some DEC engineers to do it!
Clearly, we can't assume that Microsoft strong-handed @stake. But I guess that's not the point here, is it?
It doesn't really bother me that this showed up on Slashdot.
But I am a bit annoyed that this was newsworthy for both the Washington Post and News.com. Are we so entertained by conspiracy theorists that we have to breast feed them with fodder like this?
You're forgetting the 8 new GPRs
on
Athlon 64 Debuts
·
· Score: 2, Informative
Also, try compiling the same app in 32-bit and 64-bit modes. The 64-bit app is a lot bigger and slower, since all the pointers doubled in size, so less code fits in cache, and I'm using more memory bandwidth.
Actually, this is only partly true. Because the new AMD64 instruction set includes 8 more general purpose registers, compilers now generate far less load/store code during periods of register contention. This alone ALMOST makes up for the extra byte per instruction that 64-bit instructions require.
In other words, full 64-bit AMD64 instructions require slightly more bits per instruction than their IA32 equivalents to encode. But at the same time the TOTAL number of instructions in a given program has decreased because of less register pressure (which produces load/store code). These two competing factors tend to offset each other, although the presentation I saw indicated that the total code size has increased slightly (5% or so, IIRC).
You are correct in pointing out that the instruction cache will now be less effective because bytes per instruction have increased slightly. I assume instruction cache sizes will need to increase to compensate. But instruction memory bandwidth isn't really a huge issue anymore because instruction caches do such a good job of prefetching (and code accesses tend to be highly local, unlike data accesses).
If you're referring to the latest RPC vulnerability, then here is the Knowledge Base Article that has everything you might want to know about the vulnerability, what systems it affects, including links to download the standalone patch.
Additionally, here is the Microsoft Technet Security Page where security bulletins, information, and all sorts of links get posted whenever a new Windows vulnerability is announced.
Well, if you remember, Microsoft is gearing up to offer MSN Licenses, which at least offers 3rd party clients like Trillian a solution to their predicament.
My guess is that such 3rd party clients would recoup the license costs by either selling their user database to advertising companies, or throwing some ads into the client itself. If you want to get RID of the ads, you can buy the professional version (which would also cover the cost of the client license).
It's just another example that web services aren't really free; if a company really can't make money off of them, they will be replaced with a model that generates revenue (directly or indirectly). Yahoo seems to just be applying the rule here.
6) Any comment that defends anything that has even the slightest connection to Microsoft whatsoever, regardless of its interest, factual correctness, or insightfulness, is obviously just astroturfing from a member of the Evil Empire and, as such, should be instantly modded down as either "flamebait" or "troll".
...seeing as how you've been modded down to "Flamebait".
Personally, I thought most of your points were spot-on... I just got done reading a post that basically said, "MS code is clearly buggier than Open Source code." The sad thing is, many people will read that statement and NOT realize how fallacious or unprovable it really is...
On a semi-related note, I just previewed my comment, and the date on the preview says "31 December 1969." Is that normal?
...and the .NET Framework is language-neutral
on
Does C# Measure Up?
·
· Score: 5, Interesting
What language you use depends on your application. Comparing C, C++, and C# is like comparing a wrench and a screw driver.
And this is where the.NET Framework shines, because the CLR is a generic virtual machine to which any number of languages can be compiled. Currently there are C#, C++, VB, and even Java (under the moniker J#). There has been talk of writing a Python compiler and even possibly a Perl compiler. So you can choose your language of choice, and your resulting binaries or objects will fully interoperate with the other.NET languages and class libraries.
And as far as this article is concerned, I think the interesting point is not that they're comparing apples to oranges, but just that the performance numbers for CLR-compiled C# aren't so horrible that they should scare off the majority of developers.
A decent paper discussing the theory behind ISP to ISP peering is linked through Citeseer here. To download a copy of the paper, you click on the appropriate cached format in the top right corner of the page.
BGP peering contracts dictate some asymmetry
on
Where Is The Broadband?
·
· Score: 5, Informative
I remember reading about how all of the ISPs figure out how they bill each other whenever they "peer" (i.e., connect) to another network. Lots of these contracts are apparently very complicated, but the primary metric that stuck with me was this:most companies pay for the number of packets injected into (not received from) a peer network.
Now obviously, not ALL contracts are the same, but there are some important ramifications from this concept. There are two ends of the "spectrum" of ISP's, those that are net information sources (inject more packets than they receive), and those that are net information sinks (receive more than they inject) at any peering point. End-user ISP's are therefore usually better off when their users are primarily downloading information. When home users' computers start serving more packets, the end-user ISP is forced to pay more to its provider because it has injected more packets into the adjacent network(s) at the peering point. Hosting company ISPs (hosting web servers, for instance) pay significantly more because they are net information sources, and inject far more packets into the network than they receive.
Granted, this is a vast oversimplification of what is a very complex topic that not many people are familiar with, but in my opinion, it explains why it has traditionally always been cheaper to obtain download bandwidth than upload bandwidth: peering points generally "charge" based on packets sent. Anyone who knows differently can correct me... I'm still looking for the paper on BGP peering that I read that brought this all to my attention.
This kind of blatent abuse of the law is just another step towards neo-monarchism, and more loss of freedom for the common person.
OK. Let me get this straight. A private company introduces software that basically introduces built-in encryption for word documents, spreadsheets, and email. This technology is designed to allow companies to prevent emails and documents from accidentally "leaking" to the press or into the hands of corporate spies. This won't even affect the home user AT ALL because home users don't have the necessary software to make use of IRM anyway (it requires a separate Windows 2003 Server in addition to MS's Information Rights Management software).
And the availability of this product is somehow an example of "blatant abuse of the law"? I think some people here are suffering from some kind of paranoia.
Well, that is a strange view of PCB design. After many years in the industry, I don't share that view. That would imply that everyone who uses chipsets is simply "buying a design."
In the case of a USB wireless adapter, there are some significant translation layer issues with converting the 802.11x layer to USB that are not solved by the Atheros chipset in question. I'm not saying that this problem hasn't already been solved by several other add-on chipsets, but I'm pointing out that there is at least one more layer that must be designed into the board. And this is not performed by outsiders.
In the industry, "buying a design" implies contracting out the entire board design to meet your specs. Buying a chipset to use in your own design is most certainly not "buying a design." It's just common sense system integration.
Look, I'm not saying that MS had a hard time designing this box. And I understand that as chipsets continue to evolve and BiCMOS featuresets expand, PCB designs (generally) get much easier. But PCB layout isn't a pushover by a long shot, and a well designed board can spell the difference between a buggy piece of crap that's in your face all the time and a well-behaved utility that quietly goes about its business.
AOL has already changed their protocol on several occasions specifically to break the clients. This is nothing new.
I don't understand the big deal here. The MSN Messenger servers are Microsoft property. If they want to charge 3rd party clients to use them, that's their prerogative. And it seems to be a perfectly legitimate business move, unless you're of the persuasion that believes the public is "entitled" to use these servers in any way they choose. I disagree, however, and so do private property laws in the US.
It'd be news if they DIDN'T buy a design and slap the X-Box logo on it.
Nowhere in the article does it say anything about "buying a design." MS has excellent hardware engineers.
Furthermore, if you're referring to their using the Atheros chipset, I don't know of any companies that roll their own wireless chipset. So basically, I don't understand what you're referring to at all.
This in turn means that code compiled for the 64-bit mode will run faster than the code compiled for the 32-bit mode.
There are at least two significant factors that would tend to offset the speedup obtained by adding AMD64's 8 extra general purpose registers: higher average instruction length (contributing to higher instruction read I/O), and 64-bit cache pollution.
Higher average instruction length is a factor because each 64-bit instruction (requiring use of 64-bit registers, or simply using any of the new "extended" general purpose registers) must be prefaced by a "64-bit mode" byte. That adds 8 bits to every 64-bit instruction. Increasing average instruction length has several consequences. First, the increased absolute size of code generates more instruction fetch traffic, using more memory bandwidth. Second, larger average instruction length decreases the effectiveness of the instruction cache because more space is needed to store the same number of total instructions (decreasing locality).
64-bit cache pollution is similar to the problem with the instruction cache, but generalized to include data. Using 64-bit data types and pointers increases (often unnecessarily) the size of each piece of data brought into a cache line. Once again, this increased size of each unit of data decreases data locality and increases the likelihood of capacity cache misses. The primary way to account for this is to simply increase the size of the data cache.
So in other words, it is not altogether obvious that the extra GPRs alone will compensate for these two issues that plague 64-bit computing. (For those who don't know why extra GPR's would speed things up: they offer the compiler some extra temporary storage space when performing computations, thereby reducing the number of data load/stores to main memory which are orders of magnitude slower than keeping things in registers.)
Now, don't get me wrong. I'm excited about these new AMD64 systems too. But I'm waiting for the legitimate 3rd party performance tests to come out so that I can verify the truth of these claims.
I also wonder if MS is going to make non-partnered products appear with some kind of friendly warning or desparagement, thus making Joe SixPack think that they're unsafe to use or won't work completely.
MS already warns you when you try to install a driver that has not been signed by the WHQL team that it has not been certified to work with Windows. To me, this is a *good thing* and is what has been sorely lacking on the Windows side of the house for years now: 3rd party driver quality control.
With more than half of Windows crash reports attributed to faulty 3rd party drivers, it seems that forcing hardware vendors to work more closely with Microsoft can only benefit the end result. Microsoft is not going to use its vendor relationships to push one piece of hardware over another, because Windows thrives on offerring support for a HUGE number of cool devices.
Microsoft's focus in the vendor arena is helping vendors producing a wide variety of devices with the best possible quality drivers. That is the path to making more $$. And that is the path that MS will take.
Personally, wouldn't mind seeing punitive charges for continuing to spread this kind of thing some number of weeks past the initial outbreak.
That idea isn't politically correct, but it may be the ONLY way to get people to pay attention to the fact that their computer can cause huge amounts of problems if it's neglected.
In some states, if you leave a firearm unlocked and someone steals it and uses it to commit a crime, you can be charged with the crime of "Failure to Secure a Firearm."
For example, Massachusetts General laws Chapter 140-131L requires that firearms MUST be secured in a locked container or equipped with a tamper resistant mechanical lock. Failure to secure a large capacity weapon is a felony with a minimum of one year imprisonment and/or $1000 fine. Failure to secure a non-large capacity firearm is misdemeanor.
Granted, we're not talking about life/death stakes here, but such unprotected systems can cause HUGE damages through denial of service attacks, as well as points of launch for worms and e-mail viruses. The lost time and productivity around the globe usually adds up to a significant amount of $$. That alone should be worth at LEAST a misdemeanor or citation with a fine!
OSX works very well with Windows and Unix network shares meaning no third party software to provide such support. OSX Server is very well priced per client than Windows Server 2003 or 2000 Server. The OSX Server can also provide services for Linux, Unix, and Windows systems as well as Macs.
I spent about two weeks trying to help a small private high school (about 200 students) near Memphis, TN get their systems online because the school year had already begun and they couldn't get their new infrastructure to work.
The school was tech-heavy. Every student was issued an IBM Thinkpad with WinXP, much to the chagrin of the school's IT director, who was an Apple "distinguished educator" (she had requested iBooks but the Board of Directors had vetoed her, citing numerous parental complaints).
But here's the curious part. When it came to servers, she decided to go with 3 OSX Server machines, to the tune of almost $5000 each. They were to run file shares, email servers, a web server, and some library database software the school already had.
To make a long story short, it was a mess. She and her assistant had no clue how to administer the Windows laptops. They had no idea what a domain was. Their experience setting up Appleshare networks with AtEase for authentication just didn't prepare them for administering a real heterogenous network.
The file sharing was the real problem. The IT director eventually broke down and bought a Win2k domain controller for the Windows machines for authentication and print services. But the OSX file server could not seem to use the Windows domain controller for authentication. The students not only needed a domain account, but also needed separate OSX accounts just to get to their file shares. The problem lay in Apple's LDAP authentication implementation, which did not integrate with Samba very well. Authenticating the OSX users over LDAP worked, but it refused to authenticate the WinXP users. And this was a problem because the students needed to access the file shares from their laptops. Apple did not have a nice, GUI way to configure such a situation.
The whole time this was happening, I just kept thinking how her $15,000 could have been much better spent buying a few PCs and a few copies of Win2k Server. They probably would have had a few $thousand leftover to send her assistant to some Windows administration classes. Her decision to buy Apple OSX Servers was more like a stubborn attempt to get back at the Board of Directors. At each problem that came up, she always blamed the Windows machines because "she'd never run into such problems in her old Apple lab." Nevermind that her old Apple lab was trivial...
There are restrictions on changing the registry- random users aren't allowed to change other users' preferences, for instance- but if a worm can get system privileges then it can alter anything it wants.
Exactly! But what more can you do? I mean, if I get root access to a Red Hat box, I could corrupt the RPM database just as easily as a Windows virus could corrupt the registry. It's just a programmer's API, any way you look at it. If you have the intention and the permissions, you can screw up any OS.
The registry is protected with ACLs just as well as your average access-controlled filesystem (NTFS), so complaining about it being "easy to modify" is irrelevant. Files are easy to modify too, if you have the right permissions. But you EXPECT the permissions to block stupid programs from messing with your files.
Now I'm assuming someone is going to say, "But Windows users run as Administrator!" Well, if that's the case, then running a trojan horse or spyware app is their own damn fault. Running as Administrator all the time basically makes your NT system as secure as Windows 9x was.
If every company that Microsoft directly competed with (Oracle, AOL, Sony, Nintendo, IBM, Palm, RealNetworks, Novell, just to name a few) were to boycott Microsoft products for their internal use...
If these companies were to execute such a strategy, would they not be guilty of criminal conspiracy to engage in anticompetitive practices in the first degree? (collusion to put a specific company out of business?) Isn't this just as illegal as what Microsoft has been charged with?
Call me idealistic, but I don't see how two wrongs would make a right here...
Let them fight Microsoft by making superior products that make desktop users or corporate IT departments want to switch. THAT is what is good for consumers! Don't force them to switch just because a bitter CEO issued a directive ordering the switch for political reasons...
And anyway, just because a company might sue somebody for building an emulator doesn't mean the law agrees with them.
Agreed. But then again, the law doesn't really have to be on the side of the party that "wins". After a mountain of legal fees and paperwork, some projects simply fold, even if they eventually win in court. I still have an Aureal sound card around here somewhere as a reminder to what can happen...
Slashdot Mirror
Oh, really? So you have read and signed one of these "shared source" agreements with Microsoft in the past, then?
Microsoft already has a wing of MS Research in Beijing, if I recall correctly. Why would it be such a stretch of the imagination to have a "build group" over there that produces Chinese-only builds from the same source tree that is open for government inspection? How do you know this isn't what is being done?
Absolutely!
Just look at what rock 'n roll music was in the USSR back in the 60s and 70s. It was a "politically subversive cultural influence" and therefore banned at the government level. But that only served to make groups like the Beatles ever more popular and mysterious.
The "similarities" between NT and VMS have been completely overblown. There are very few, and they are shared by just about any modern OS.
When you hire DEC engineers to architect a new product from scratch, you are going to see SOME similarities in the architecture with VMS. But that does NOT mean that the new product is "derived"!
I challenge you to put your OS architecture knowledge where your mouth is. Tell me all about the "derived" features that proves NT is a derivative of VMS and nothing more. I'll be waiting with bated breath.
This has never happened in the past... what makes you think it will start now?
Microsoft didn't get to their position by marketing. Making that claim is simply being disingenuous. They rose to their current position through a combination of great timing, shrewd contract negotiation, and "good enough" products.
First they got a contract with IBM for DOS because they had an operating system to sell when others did not (or were unwilling). After that point, they rode the Intel x86 evolutional wave, improving their OS to basically keep up with Intel (with a sizable lag). No marketing was required: killer apps were written (Lotus 1-2-3 and Wordperfect come to mind), and history was made.
But even though their historical products achieved success mostly through timing genius, I think NT represents a new chapter for MS and shows that they DO know how to create a good product from scratch. Even if they DID have to hire some DEC engineers to do it!
Clearly, we can't assume that Microsoft strong-handed @stake. But I guess that's not the point here, is it?
It doesn't really bother me that this showed up on Slashdot.
But I am a bit annoyed that this was newsworthy for both the Washington Post and News.com. Are we so entertained by conspiracy theorists that we have to breast feed them with fodder like this?
Actually, this is only partly true. Because the new AMD64 instruction set includes 8 more general purpose registers, compilers now generate far less load/store code during periods of register contention. This alone ALMOST makes up for the extra byte per instruction that 64-bit instructions require.
In other words, full 64-bit AMD64 instructions require slightly more bits per instruction than their IA32 equivalents to encode. But at the same time the TOTAL number of instructions in a given program has decreased because of less register pressure (which produces load/store code). These two competing factors tend to offset each other, although the presentation I saw indicated that the total code size has increased slightly (5% or so, IIRC).
You are correct in pointing out that the instruction cache will now be less effective because bytes per instruction have increased slightly. I assume instruction cache sizes will need to increase to compensate. But instruction memory bandwidth isn't really a huge issue anymore because instruction caches do such a good job of prefetching (and code accesses tend to be highly local, unlike data accesses).
If you're referring to the latest RPC vulnerability, then here is the Knowledge Base Article that has everything you might want to know about the vulnerability, what systems it affects, including links to download the standalone patch.
Additionally, here is the Microsoft Technet Security Page where security bulletins, information, and all sorts of links get posted whenever a new Windows vulnerability is announced.
Well, if you remember, Microsoft is gearing up to offer MSN Licenses, which at least offers 3rd party clients like Trillian a solution to their predicament.
My guess is that such 3rd party clients would recoup the license costs by either selling their user database to advertising companies, or throwing some ads into the client itself. If you want to get RID of the ads, you can buy the professional version (which would also cover the cost of the client license).
It's just another example that web services aren't really free; if a company really can't make money off of them, they will be replaced with a model that generates revenue (directly or indirectly). Yahoo seems to just be applying the rule here.
...seeing as how you've been modded down to "Flamebait".
Personally, I thought most of your points were spot-on... I just got done reading a post that basically said, "MS code is clearly buggier than Open Source code." The sad thing is, many people will read that statement and NOT realize how fallacious or unprovable it really is...
On a semi-related note, I just previewed my comment, and the date on the preview says "31 December 1969." Is that normal?
And this is where the .NET Framework shines, because the CLR is a generic virtual machine to which any number of languages can be compiled. Currently there are C#, C++, VB, and even Java (under the moniker J#). There has been talk of writing a Python compiler and even possibly a Perl compiler. So you can choose your language of choice, and your resulting binaries or objects will fully interoperate with the other .NET languages and class libraries.
And as far as this article is concerned, I think the interesting point is not that they're comparing apples to oranges, but just that the performance numbers for CLR-compiled C# aren't so horrible that they should scare off the majority of developers.
A decent paper discussing the theory behind ISP to ISP peering is linked through Citeseer here. To download a copy of the paper, you click on the appropriate cached format in the top right corner of the page.
I remember reading about how all of the ISPs figure out how they bill each other whenever they "peer" (i.e., connect) to another network. Lots of these contracts are apparently very complicated, but the primary metric that stuck with me was this: most companies pay for the number of packets injected into (not received from) a peer network .
Now obviously, not ALL contracts are the same, but there are some important ramifications from this concept. There are two ends of the "spectrum" of ISP's, those that are net information sources (inject more packets than they receive), and those that are net information sinks (receive more than they inject) at any peering point. End-user ISP's are therefore usually better off when their users are primarily downloading information. When home users' computers start serving more packets, the end-user ISP is forced to pay more to its provider because it has injected more packets into the adjacent network(s) at the peering point. Hosting company ISPs (hosting web servers, for instance) pay significantly more because they are net information sources, and inject far more packets into the network than they receive.
Granted, this is a vast oversimplification of what is a very complex topic that not many people are familiar with, but in my opinion, it explains why it has traditionally always been cheaper to obtain download bandwidth than upload bandwidth: peering points generally "charge" based on packets sent. Anyone who knows differently can correct me... I'm still looking for the paper on BGP peering that I read that brought this all to my attention.
OK. Let me get this straight. A private company introduces software that basically introduces built-in encryption for word documents, spreadsheets, and email. This technology is designed to allow companies to prevent emails and documents from accidentally "leaking" to the press or into the hands of corporate spies. This won't even affect the home user AT ALL because home users don't have the necessary software to make use of IRM anyway (it requires a separate Windows 2003 Server in addition to MS's Information Rights Management software).
And the availability of this product is somehow an example of "blatant abuse of the law"? I think some people here are suffering from some kind of paranoia.
Well, that is a strange view of PCB design. After many years in the industry, I don't share that view. That would imply that everyone who uses chipsets is simply "buying a design."
In the case of a USB wireless adapter, there are some significant translation layer issues with converting the 802.11x layer to USB that are not solved by the Atheros chipset in question. I'm not saying that this problem hasn't already been solved by several other add-on chipsets, but I'm pointing out that there is at least one more layer that must be designed into the board. And this is not performed by outsiders.
In the industry, "buying a design" implies contracting out the entire board design to meet your specs. Buying a chipset to use in your own design is most certainly not "buying a design." It's just common sense system integration.
Look, I'm not saying that MS had a hard time designing this box. And I understand that as chipsets continue to evolve and BiCMOS featuresets expand, PCB designs (generally) get much easier. But PCB layout isn't a pushover by a long shot, and a well designed board can spell the difference between a buggy piece of crap that's in your face all the time and a well-behaved utility that quietly goes about its business.
Are we in agreement there at least? :-)
AOL has already changed their protocol on several occasions specifically to break the clients. This is nothing new.
I don't understand the big deal here. The MSN Messenger servers are Microsoft property. If they want to charge 3rd party clients to use them, that's their prerogative. And it seems to be a perfectly legitimate business move, unless you're of the persuasion that believes the public is "entitled" to use these servers in any way they choose. I disagree, however, and so do private property laws in the US.
Nowhere in the article does it say anything about "buying a design." MS has excellent hardware engineers.
Furthermore, if you're referring to their using the Atheros chipset, I don't know of any companies that roll their own wireless chipset. So basically, I don't understand what you're referring to at all.
There are at least two significant factors that would tend to offset the speedup obtained by adding AMD64's 8 extra general purpose registers: higher average instruction length (contributing to higher instruction read I/O), and 64-bit cache pollution.
Higher average instruction length is a factor because each 64-bit instruction (requiring use of 64-bit registers, or simply using any of the new "extended" general purpose registers) must be prefaced by a "64-bit mode" byte. That adds 8 bits to every 64-bit instruction. Increasing average instruction length has several consequences. First, the increased absolute size of code generates more instruction fetch traffic, using more memory bandwidth. Second, larger average instruction length decreases the effectiveness of the instruction cache because more space is needed to store the same number of total instructions (decreasing locality).
64-bit cache pollution is similar to the problem with the instruction cache, but generalized to include data. Using 64-bit data types and pointers increases (often unnecessarily) the size of each piece of data brought into a cache line. Once again, this increased size of each unit of data decreases data locality and increases the likelihood of capacity cache misses. The primary way to account for this is to simply increase the size of the data cache.
So in other words, it is not altogether obvious that the extra GPRs alone will compensate for these two issues that plague 64-bit computing. (For those who don't know why extra GPR's would speed things up: they offer the compiler some extra temporary storage space when performing computations, thereby reducing the number of data load/stores to main memory which are orders of magnitude slower than keeping things in registers.)
Now, don't get me wrong. I'm excited about these new AMD64 systems too. But I'm waiting for the legitimate 3rd party performance tests to come out so that I can verify the truth of these claims.
MS already warns you when you try to install a driver that has not been signed by the WHQL team that it has not been certified to work with Windows. To me, this is a *good thing* and is what has been sorely lacking on the Windows side of the house for years now: 3rd party driver quality control.
With more than half of Windows crash reports attributed to faulty 3rd party drivers, it seems that forcing hardware vendors to work more closely with Microsoft can only benefit the end result. Microsoft is not going to use its vendor relationships to push one piece of hardware over another, because Windows thrives on offerring support for a HUGE number of cool devices.
Microsoft's focus in the vendor arena is helping vendors producing a wide variety of devices with the best possible quality drivers. That is the path to making more $$. And that is the path that MS will take.
Spell it with me: T - E - L - N - E - T
If that is too many letters, try this: F - T - P
And who, exactly, wasn't sending passwords in clear text?
At least get your facts straight before you make outrageous claims like that.
That idea isn't politically correct, but it may be the ONLY way to get people to pay attention to the fact that their computer can cause huge amounts of problems if it's neglected.
In some states, if you leave a firearm unlocked and someone steals it and uses it to commit a crime, you can be charged with the crime of "Failure to Secure a Firearm."
For example, Massachusetts General laws Chapter 140-131L requires that firearms MUST be secured in a locked container or equipped with a tamper resistant mechanical lock. Failure to secure a large capacity weapon is a felony with a minimum of one year imprisonment and/or $1000 fine. Failure to secure a non-large capacity firearm is misdemeanor.
Granted, we're not talking about life/death stakes here, but such unprotected systems can cause HUGE damages through denial of service attacks, as well as points of launch for worms and e-mail viruses. The lost time and productivity around the globe usually adds up to a significant amount of $$. That alone should be worth at LEAST a misdemeanor or citation with a fine!
I spent about two weeks trying to help a small private high school (about 200 students) near Memphis, TN get their systems online because the school year had already begun and they couldn't get their new infrastructure to work.
The school was tech-heavy. Every student was issued an IBM Thinkpad with WinXP, much to the chagrin of the school's IT director, who was an Apple "distinguished educator" (she had requested iBooks but the Board of Directors had vetoed her, citing numerous parental complaints).
But here's the curious part. When it came to servers, she decided to go with 3 OSX Server machines, to the tune of almost $5000 each. They were to run file shares, email servers, a web server, and some library database software the school already had.
To make a long story short, it was a mess. She and her assistant had no clue how to administer the Windows laptops. They had no idea what a domain was. Their experience setting up Appleshare networks with AtEase for authentication just didn't prepare them for administering a real heterogenous network.
The file sharing was the real problem. The IT director eventually broke down and bought a Win2k domain controller for the Windows machines for authentication and print services. But the OSX file server could not seem to use the Windows domain controller for authentication. The students not only needed a domain account, but also needed separate OSX accounts just to get to their file shares. The problem lay in Apple's LDAP authentication implementation, which did not integrate with Samba very well. Authenticating the OSX users over LDAP worked, but it refused to authenticate the WinXP users. And this was a problem because the students needed to access the file shares from their laptops. Apple did not have a nice, GUI way to configure such a situation.
The whole time this was happening, I just kept thinking how her $15,000 could have been much better spent buying a few PCs and a few copies of Win2k Server. They probably would have had a few $thousand leftover to send her assistant to some Windows administration classes. Her decision to buy Apple OSX Servers was more like a stubborn attempt to get back at the Board of Directors. At each problem that came up, she always blamed the Windows machines because "she'd never run into such problems in her old Apple lab." Nevermind that her old Apple lab was trivial...
Exactly! But what more can you do? I mean, if I get root access to a Red Hat box, I could corrupt the RPM database just as easily as a Windows virus could corrupt the registry. It's just a programmer's API, any way you look at it. If you have the intention and the permissions, you can screw up any OS.
The registry is protected with ACLs just as well as your average access-controlled filesystem (NTFS), so complaining about it being "easy to modify" is irrelevant. Files are easy to modify too, if you have the right permissions. But you EXPECT the permissions to block stupid programs from messing with your files.
Now I'm assuming someone is going to say, "But Windows users run as Administrator!" Well, if that's the case, then running a trojan horse or spyware app is their own damn fault. Running as Administrator all the time basically makes your NT system as secure as Windows 9x was.
Let the Zionist infidels bask in their illusion, but I triple guarantee you: there are absolutely NO power problems at all in the Five Boroughs.
I am not scared, and neither should you be! We will defeat the enemy spreading such rumors and use their own shoes to butcher them!
Real Iraqi Information Minister QuotesIf these companies were to execute such a strategy, would they not be guilty of criminal conspiracy to engage in anticompetitive practices in the first degree? (collusion to put a specific company out of business?) Isn't this just as illegal as what Microsoft has been charged with?
Call me idealistic, but I don't see how two wrongs would make a right here...
Let them fight Microsoft by making superior products that make desktop users or corporate IT departments want to switch. THAT is what is good for consumers! Don't force them to switch just because a bitter CEO issued a directive ordering the switch for political reasons...
Agreed. But then again, the law doesn't really have to be on the side of the party that "wins". After a mountain of legal fees and paperwork, some projects simply fold, even if they eventually win in court. I still have an Aureal sound card around here somewhere as a reminder to what can happen...