I agree completely. The cost of the CPU is only one factor. Although I tend to look at the higher end of the spectrum, so I'm deciding what CPU makes the most sense to throw in the $1400 system I've built around it. Putting in a $73 CPU really wouldn't make a whole lot of sense.
Unfortunately, they would then have to run the complete set of tests to be accurate with atleast 3-4 different base systems to give a clear picture. El-Cheapo - Using the noname brands of whatever is cheapest, Budget - Use name brand/common parts, Gamer - Same as budget but toss in a high end video card, High End - Same as gamer but toss in a raid array, more memory, maybe SLI.
Priority based I/O (Completely redesigned I/O subsystem) - Running backups, disk defrags, indexing, virus scanning while using your system is almost transparent.
Resource Manager - Get a quick glimpse of CPU usage, Memory Usage, Disk Usage, and Network Usage and be able to drill down to find which processes are using what subsystems easily.
Media Center AND pro features on a single install (Yay!)
DirectX 10 - Better games, shifting some of the burden to the video card.
Redesigned audio - No longer will ultracrappy drivers from creative cause your system to become unstable.
Process View tooltips/task switch - Pick the right instance of a running application every time.
Shadow Copies - Undelete for file content.
IPv6 support
IIS 7 - Complete website configuration in XML (web.config)
Replacable authentication methods in FTP Server
Randomize process offsets - Making it harder for viruses to use stack overflows to compromise your system
I have to say, if you are so concerned about grouping duplicates then you should review your email policies. Afraid you are going to get an ton of complaint emails responses?
Did you bother to even look at the complaints? Most of those are depreciated and for backwards compatability only. To top it off, they were rarely used to begin with. A good portion of those are for backwards compatability with Word Perfect, which isn't a MS product. If Microsoft didn't document them, you'd see people screaming about bracing and extending.
Through the whole document, there are 10ish attributes that if you ignored, all the documents that were written in a eastern asian language using some seldomly used options in word 97 (or earlier) or word perfect 6.0 (or earlier), the spacing might be a little off. That's more than enough than you need to know that you're pretty safe ignoring them (or passing them through). But if you REALLY want to be completely anal about supporting those with pixel-perfect representation as they originally appeared, they even tell you how you could find out what they do. Considering OpenOffice (or any ODF native Office Suite) can't do that at all currently, how is that a step backwards?
Or if Microsoft completely dropped those attributes, would that now classify it as an open format?
It's a shame that ODF can't compete in the marketplace, and has to try and get laws passed for it to even be seriously considered. ODF backers fired the first shot, not Microsoft.
I'm familiar with Groklaw. But NOTHING on groklaw's website does anything to prove that OOXML isn't an open format. In fact quite the opposite. You've just proven that it IS an open format. Groklaw has a list of complaints that they think should be changed in OOXML to make it better. But again, that isn't what defines an open format, so the groklaw list doesn't prove that. In fact, it proves the exact opposite. Groklaw must have been able to read the specification in order to make the above list, and that proves the format is open.
As I said before, you may not like the format. You may think it's overly complicated. You may think it's ambiguous, or parts need better clarification. All of those MAY be valid.
Most people aren't against a free and open format. They are against three things. One being any law that tries to pick a specific technology that must be used. Most of the current proposed laws demand ODF, or something that might as well have just said ODF. That's may be fine for today if that is really what is needed, but it should be something that is done because it's the right choice, not because it's a law. Things change quick, laws don't.
The second being ODF in particular. It's too simple of a format. It's too ambiguous in many areas and/or defers actual clarification to a separate document. There is only *1* office suite that supports it, and it has very little presence in business
The third is that it isn't able to faithfully store the vast amounts of documents that are already created. To convert all the documents that businesses currently have to ODF would be a massive undertaking, one that costs a lot, and has no revenue. Because of that, the largest of businesses will refuse to switch, and we won't end up with the ODF-lovers utopia. We'll now have two formats one of which is seldom used (ODF) to talk to those branches of the government that mandate it, and everyone else. That puts the cost burden on them to constant convert back and forth, or you can argue on the companies they do business with, either way, the government pays for it, and ultimately we all do.
There are many more reasons, like trying to keep idiology (which in many cases has similiarities to religion) from being mandated anywhere. It's neither the governments or business's primary (or secondary, or tertiary) role to use a particular format (open/closed/free/commercial). For publicly traded companies, they already try to maximize profits, and I'm personally against any legislation at all that tries to force companies to use/pay for something that boils down to ideological opinion. If it's more profitable, cheaper, or easier to use a tool, then it should be used. Not because someone says that it conforms to some standard that makes a few people all warm and fuzzy inside.
Please, go and write some multi-threaded applications as I have for the past 10 years. Then you can refute your own posts.
I am fully aware of how thread context-swithcing affects software peformance as I have not only used many of the common ones out there, but I have even written a few of my own. While I agree that problems that can be parallelized are often better if they are written as such, it is important that you always keep in mind the performance hit you take for doing that. Taking a simple application and spawning off 6,000 threads just because it can be parallelized to that extent is a bad idea, and will have serious performance issues.
Well, yes. Quantum mechanics is also such a misunderstood term. Nevertheless, both "open format" and "quantum mechanics" have important, well defined meanings, and the responsibility is on you to understand and use them correctly. I agree, unfortunately, your definitions are wrong. "Open formats" is well-defined, but it has nothing to do with copyrights or patents. "Open formats" means what it always has, no matter how hard people like you try to redefine it to be what you want it to be. "Open formats" means that the format is documented and available to third parties. That is IT. It doesn't mean that it's free. It doesn't mean that it's not copyrighted or patented. Those things are covered by a GPL, or other similiar arrangement.
ODF is an open format, OOXML is a closed, proprietary format. Again, your bias shows through. OOXML is open. It's fully documented. It's been approved as an ANSI standard. It's even free to use (although that goes beyond just being open). It's also used by more than just microsoft. So it's neither closed, nor proprietary as you claim.
There is a performance (and complexity) hit when you run many more threads than you have cores. Each thread as it is switched in and out of the core has some pieces of data that also must be saved and restored to get the core back into the state it was in when that thread was last executing. At a very minimum it would be things like the instruction pointer, the processor flags, and the processor registers. So switching between threads unnecessarily will cause the process to actually run slower than if it was single (or less) threaded.
The most efficient code will use one thread per core, and each thread will use the processor at near maximum. However, if you can not get a thread to use the full core, then you are better off having another thread that can be switched in when another is in some kind of wait state. It's a balancing act between being able to use the full potential of multi-core processors and still running efficiently on single-core (or less cored) machines.
Because ODF is too simple of a file format to represent (or store), even older versions of word, excel, powerpoint, or access data.
It'd be similiar to saying that we should all convert to a new and great "open" computer interface. Then trying to push the calculator as the interface, trying to force laws that government agencies toss out all their PC, and standardize on calculators only and wondering why PC users don't want it.
For example, if you don't specify if you want next day am, next day, or bulk (QoS) then they hold the package forever. Then they charge you differently for each packet based on QoS and size. Oh wait, that is what you are arguing against.
I haven't heard of baseband being defined that way either, but broadband isn't that hard to figure out.
Diferent frequency ranges are broken up into bands. Anything that uses more than one of these bands is "broadband".
But then again, I have no problem understanding that the term broadband no longer refers to a technical description of how something works, and refers to a high speed internet connection now. Times change, and so do the meanings of words.
It all depends on what you are trying to do, and with whom. When I was having some serious problems with my cable company (comcast), they finally moved me to a different node, and accidentally left me uncapped. I was able to hit speeds of 40Mb/sec consistantly from the servers I use, and I use them a lot. Many websites can't send data that fast (MANY), but the ones I use most could.
No offense, but google is your friend. ODF is a very simplistic document format. Which is great for simple documents, but there a lot of things that it doesn't handle. OOXML isn't simple, but it does a lot more.
You would think so, but my experience is quite different. I have had 2 windows machines on the internet, both with no firewall at all for the past 10 years or so. I have had 2 linux machines on teh internet, again no firewall for the past 2 years.
1 Windows machine (Currently running windows 2003 server) got hit once via the Code Red virus a number of years ago. Even that was minor -- It failed to deliver it's payload, and never had escalated privs.
1 Linux machine has been rebuilt multiple times because of it getting hacked.
1 Windows machine (Currently running Vista) has never been hacked, cracked, or had any virus on it (And I use it daily).
1 Linux machine has only been hacked twice.
You need to get your head out of the sand. Talk to any large ISP technician. Go browse the forums of linux forums that belong to server based products. Linux servers get hacked ALL the time.
Because a large portion of slashdot, and a very large portion of the vocal slashdot, DON'T MAKE A LIVING at all. So free is the only thing they can afford.
Slashdot Mirror
I agree completely. The cost of the CPU is only one factor. Although I tend to look at the higher end of the spectrum, so I'm deciding what CPU makes the most sense to throw in the $1400 system I've built around it. Putting in a $73 CPU really wouldn't make a whole lot of sense.
Unfortunately, they would then have to run the complete set of tests to be accurate with atleast 3-4 different base systems to give a clear picture. El-Cheapo - Using the noname brands of whatever is cheapest, Budget - Use name brand/common parts, Gamer - Same as budget but toss in a high end video card, High End - Same as gamer but toss in a raid array, more memory, maybe SLI.
I'll take 48kHz and stable vs 96kHz and blue screening any day. I'll even take 4kHz if necessary.
I have to say, if you are so concerned about grouping duplicates then you should review your email policies. Afraid you are going to get an ton of complaint emails responses?
Did you bother to even look at the complaints? Most of those are depreciated and for backwards compatability only. To top it off, they were rarely used to begin with. A good portion of those are for backwards compatability with Word Perfect, which isn't a MS product. If Microsoft didn't document them, you'd see people screaming about bracing and extending. Through the whole document, there are 10ish attributes that if you ignored, all the documents that were written in a eastern asian language using some seldomly used options in word 97 (or earlier) or word perfect 6.0 (or earlier), the spacing might be a little off. That's more than enough than you need to know that you're pretty safe ignoring them (or passing them through). But if you REALLY want to be completely anal about supporting those with pixel-perfect representation as they originally appeared, they even tell you how you could find out what they do. Considering OpenOffice (or any ODF native Office Suite) can't do that at all currently, how is that a step backwards? Or if Microsoft completely dropped those attributes, would that now classify it as an open format?
It's a shame that ODF can't compete in the marketplace, and has to try and get laws passed for it to even be seriously considered. ODF backers fired the first shot, not Microsoft.
I'm familiar with Groklaw. But NOTHING on groklaw's website does anything to prove that OOXML isn't an open format. In fact quite the opposite. You've just proven that it IS an open format. Groklaw has a list of complaints that they think should be changed in OOXML to make it better. But again, that isn't what defines an open format, so the groklaw list doesn't prove that. In fact, it proves the exact opposite. Groklaw must have been able to read the specification in order to make the above list, and that proves the format is open.
As I said before, you may not like the format. You may think it's overly complicated. You may think it's ambiguous, or parts need better clarification. All of those MAY be valid.
Wait for the cost/benefit ratio to be more favorable.
There is no proven (and open) IPv6 stack. The one in linux has had numerous bugs reported so far.
You say stuff, but nothing on how it's either closed or proprietary. I get it, you don't like OOXML, but that doesn't make it closed or proprietary.
Most people aren't against a free and open format. They are against three things. One being any law that tries to pick a specific technology that must be used. Most of the current proposed laws demand ODF, or something that might as well have just said ODF. That's may be fine for today if that is really what is needed, but it should be something that is done because it's the right choice, not because it's a law. Things change quick, laws don't.
The second being ODF in particular. It's too simple of a format. It's too ambiguous in many areas and/or defers actual clarification to a separate document. There is only *1* office suite that supports it, and it has very little presence in business
The third is that it isn't able to faithfully store the vast amounts of documents that are already created. To convert all the documents that businesses currently have to ODF would be a massive undertaking, one that costs a lot, and has no revenue. Because of that, the largest of businesses will refuse to switch, and we won't end up with the ODF-lovers utopia. We'll now have two formats one of which is seldom used (ODF) to talk to those branches of the government that mandate it, and everyone else. That puts the cost burden on them to constant convert back and forth, or you can argue on the companies they do business with, either way, the government pays for it, and ultimately we all do.
There are many more reasons, like trying to keep idiology (which in many cases has similiarities to religion) from being mandated anywhere. It's neither the governments or business's primary (or secondary, or tertiary) role to use a particular format (open/closed/free/commercial). For publicly traded companies, they already try to maximize profits, and I'm personally against any legislation at all that tries to force companies to use/pay for something that boils down to ideological opinion. If it's more profitable, cheaper, or easier to use a tool, then it should be used. Not because someone says that it conforms to some standard that makes a few people all warm and fuzzy inside.
Please, go and write some multi-threaded applications as I have for the past 10 years. Then you can refute your own posts.
I am fully aware of how thread context-swithcing affects software peformance as I have not only used many of the common ones out there, but I have even written a few of my own. While I agree that problems that can be parallelized are often better if they are written as such, it is important that you always keep in mind the performance hit you take for doing that. Taking a simple application and spawning off 6,000 threads just because it can be parallelized to that extent is a bad idea, and will have serious performance issues.
ODF is an open format, OOXML is a closed, proprietary format. Again, your bias shows through. OOXML is open. It's fully documented. It's been approved as an ANSI standard. It's even free to use (although that goes beyond just being open). It's also used by more than just microsoft. So it's neither closed, nor proprietary as you claim.
There is a performance (and complexity) hit when you run many more threads than you have cores. Each thread as it is switched in and out of the core has some pieces of data that also must be saved and restored to get the core back into the state it was in when that thread was last executing. At a very minimum it would be things like the instruction pointer, the processor flags, and the processor registers. So switching between threads unnecessarily will cause the process to actually run slower than if it was single (or less) threaded. The most efficient code will use one thread per core, and each thread will use the processor at near maximum. However, if you can not get a thread to use the full core, then you are better off having another thread that can be switched in when another is in some kind of wait state. It's a balancing act between being able to use the full potential of multi-core processors and still running efficiently on single-core (or less cored) machines.
God, it's "doesn't".
So you're saying, use linux it's just like windows except the software you want to run costs 100 times as much?
Because ODF is too simple of a file format to represent (or store), even older versions of word, excel, powerpoint, or access data. It'd be similiar to saying that we should all convert to a new and great "open" computer interface. Then trying to push the calculator as the interface, trying to force laws that government agencies toss out all their PC, and standardize on calculators only and wondering why PC users don't want it.
Actually they both work because "FLOP" (Floating Operations) is also a valid abbreviation.
I agree. They should act more like FedEx.
For example, if you don't specify if you want next day am, next day, or bulk (QoS) then they hold the package forever. Then they charge you differently for each packet based on QoS and size. Oh wait, that is what you are arguing against.
I haven't heard of baseband being defined that way either, but broadband isn't that hard to figure out.
Diferent frequency ranges are broken up into bands. Anything that uses more than one of these bands is "broadband".
But then again, I have no problem understanding that the term broadband no longer refers to a technical description of how something works, and refers to a high speed internet connection now. Times change, and so do the meanings of words.
It all depends on what you are trying to do, and with whom. When I was having some serious problems with my cable company (comcast), they finally moved me to a different node, and accidentally left me uncapped. I was able to hit speeds of 40Mb/sec consistantly from the servers I use, and I use them a lot. Many websites can't send data that fast (MANY), but the ones I use most could.
No offense, but google is your friend. ODF is a very simplistic document format. Which is great for simple documents, but there a lot of things that it doesn't handle. OOXML isn't simple, but it does a lot more.
You would think so, but my experience is quite different. I have had 2 windows machines on the internet, both with no firewall at all for the past 10 years or so. I have had 2 linux machines on teh internet, again no firewall for the past 2 years. 1 Windows machine (Currently running windows 2003 server) got hit once via the Code Red virus a number of years ago. Even that was minor -- It failed to deliver it's payload, and never had escalated privs. 1 Linux machine has been rebuilt multiple times because of it getting hacked. 1 Windows machine (Currently running Vista) has never been hacked, cracked, or had any virus on it (And I use it daily). 1 Linux machine has only been hacked twice. You need to get your head out of the sand. Talk to any large ISP technician. Go browse the forums of linux forums that belong to server based products. Linux servers get hacked ALL the time.
Because a large portion of slashdot, and a very large portion of the vocal slashdot, DON'T MAKE A LIVING at all. So free is the only thing they can afford.