Linux uses the FAT filesystem only for interoperability with Windows. This is not the native filesystem of Linux, and is not required.
The FAT filesystem is widely used by things other than Linux. For example, my digital camera uses FAT on the removable storage card. I imagine the storage card people would make easier targets than Linux. (Of couse, they could also argue that they used FAT because it was the only thing that Windows supported.)
They only upheld the patent; no case has been settled yet. There is still the issue of latches. Since Microsoft waited so long to enforce the patent, and FAT has now become a de facto standard for interoperability, can they turn around and start demanding royalty payments now? A lot depends on the judge, but there's certainly no guarantee that they would win.
In regards to the long versus short file names. FAT (actually, VFAT) is the only FS I know of that does this. The ext2 filesystem has no use for "short" file names. There are several other File Systems available on Linux, and AFAIK none of them have short and long file names. Short file names as described in this patent are a hack on the FAT file system, and are very specific to DOS and pre-NTFS Windows.
I'm beginning to wonder..... perhaps a community "Prior Art" effort, somewhat comparable to the open source community, is needed.
Identify, research and debunk such absurd/greedy patents (and perhaps eventually much or all of the software patents nonsense) so as to get the patents withdrawn/cancelled.... and/or reassigned to some "open source" holder.
I agree, except that the research for prior art should be done before the patent is issued, not after. This it the job of the US PTO, which IMO, they are doing poorly, as far as software is concerned. (If you want to help, go here and apply for a job.) The patent is supposed to be rejected unless it is "nonobvious to a person having ordinary skill in the area of technology related to the invention". Every software patent I've read has been pretty obvious.
US patent law clearly states, once the patent is issued, it is considered valid, unless prior art can be demonstrated in court. Patent lawsuits are one of the few trials where you are not considered "innocent until proven guilty". You are guilty unless proven innocent.
<soapbox> Which is why software patents should be outlawed in the US. Write your congressman, if you haven't already. The worst part of a software patent, is that it can be granted based on a description of the software, even in cases where the "inventor" didn't implement anything. Writing software is like writing a book. Copyright should apply, not Patent -- see this page. (At one time, I had plans to start my own software company, but now it appears too risky. There is no way that I could possibly know what patents I was infringing. <rant> Unless something changes, open source will cease to exist, and the only software developers left will be Microsoft and IBM, who will cross-license all of their patents to each other. </rant> ) </soapbox>
If you buy a workstation
from
Apple, the difference in price between the 1CPU and 2CPU version is only $600 ($2399 vs $2999).
No, I expect the "base configuration" to contain 4 processors, but probably at the slowest speed. So I would look for something with four 1.6GHz CPUS, and perhaps 2GB of RAM as the base system. (You're probably right about the high-end box costing $10,000+, but that's with probably 32GB RAM and 500GB+ of high-speed disk.) C'mon people, they need to compete with Dell and Sun here, so they're at least going to be in the same price/performance ballpark.
You haven't a clue... you can't build a 1 CPU based off of the next generation PowerPC 970 from IBM. Each chip has two CPUs.
Um, Apple's PPC970s have one CPU per chip. Presumably IBM will be using the same chips.
Actually, it was the IBM Power4 that had two CPUs per chips. The 970 has only one. (See this article)
However, the article does say, "2U two-way and 4U four-way configurations." "Way" in this context usually means CPU, so this means that there will be exactly four CPUS in the 4U configuration. So $3500 sounds like a very competitive price.
So Person A can get a GPL copy of the source code for only EUR 50? He could then release a "derivative" work from the original source (for free -- read the GPL). Everyone else can get their copy from Person A. (Would you pay 50 if you could *legally* get the same thing for free?) So... how does he plan to sell 1000 memberships? It soulds like he wants to charge the developers a "right to work on my project" tax.
Revenue expectations (July-December 2002) - Initial funding (community, e-shop, sponsoring): 100.000 - Member License subscriptions: 1000 in 6 months, 50.000 - E-shop revenues general products: 20.000 - Product License subscriptions: 10 x 5k = 50.000 Total: EUR 220.000
(See here)
Now... Where, exactly, does this "Initial funding" come from? I question these figures. How much money do similar sized *GPL* project bring in over 6 months time? However, Tom does make one more good point in his proposal:
Doing nothing at all, leaving all IP locked in the Holding, will simply downgrade the value of Blender to close to zero within 12-18 months.
This program will be utterly useless in 12 months if development does not continue. So, even though this is a risky venture, it is probably the best chance for the shareholders.
In his paper, he uses the basic COCOMO model for estimating the cost. This model, quite frankly, sucks. Boehm's book even states, more or less, that the COCOMO model is only accurate to a factor of 10.
I have the COCOMO II book, and I have used the COCOMO model for certain projects. I agree that it is not appropriate here. COCOMO was designed with a narrow focus in mind, and applied best to repeatable projects in a structured work environment. It requires you to estimate parameters for factors such as "Programmer Unfamiliarity", "Precedentedness" "Development Flexibility", "Team Cohesion", "Process Maturity", "Multisite Development", etc. Each of these fudge-factors makes it extremely difficult to correctly apply the model to someone else's work.
Also, each of these factors is likely to be different for each major component.
"I was unable to find a publicly-backed average value for overhead, also called the 'wrap rate.' This value is necessary to estimate the costs of office space, equipment, overhead staff, and so on. I talked to two cost analysts, who suggested that 2.4 would be a reasonable overhead (wrap) rate."(from here)
He is using an average overhead rate for a large corporation. He forgot to take in to account the fact that Open-Source developers (generally) don't get office space or health insurance or secretaries. They use their own equipment in their own homes. So a more reasonable overhead rate for this project would be close to 0.1.
So taking all of this in to account, he's probably off by a factor of more than 100. (If you want to know how accurate he was, compare his estimate to the actual cost of developing a Linux distro...;)
While it might have made interesting headlines, I see little value in the actual number.
The gist of this technology is that my next computer will allow only signed code to run. Of course... This is the next logical step after SSL, which allows only signed applets to run from the internet. The difference is that now they are hiding the decoder key in a piece of hardware. This hardly seems like the type of novelty that deserves a patent. (Reminds me of the Clipper Chip)
This technology means that if I want to upgrade my BIOS, I can't just download the latest BIOS image from DELL. Now, DELL needs to jump through the M$ hoops and get the code signed by a private key.
Sure, you might assume that many vendors will hold a signing key, and that they will all be able to sign their own code. However, you only need to compromise one master key to defeat the system. It is unlikely that there will be more than a handfull of signing keys (One each for Microsoft, Intel, AMD, and IBM).
This means than any software vendor who wants to write code that runs in kernel space (device drivers... goodbye Ethereal) or handles digital media (music, video, images... goodbye Mozilla, GIMP, etc) or interacts with any of these things, will need to be signed. Once they've won the initial battle, they sign IE, WMP, and Office, but create a bureaucracy to sign competitor's products. (You thought OS licensing was expensive?) Also, every new release will need to be signed. Sound like an open-source killer? Sounds like blatant anti-trust to me.
You might think this is not a big deal, because open source media players suck anyway. But if this catches on, then a few years from now, you won't have a choice of media players (or web browsers or whatever...) Not that it would matter which one you picked because M$ gets royalties for them all... And since they can sign a piece of code with an expiration date, they can always force you to upgrade to the latest version!!! (They call it a "feature" which requires you to apply security patches...)
The best part of all this is that it's now enforced by Hardware! Which means that you won't be able to turn it off. It could prevent you from booting from a (non-signed) floppy disk. Goodbye Linux... Goodbye BSD...
They should involve a PowerPC/Mac and some other CPUs/OSes in the fray. THAT would make an interesting tournament...
Not a very impressive article
on
Is Linux Dead?
·
· Score: 1
Agreed. Though I didn't see that much praise. He started out by saying "Just a few years ago...", but then he goes on to wonder why Linux hasn't changed the world yet. How he heard a "promise" from some guy at a PC conference saying that Linux would liberate the world "from the burden of paying for software". "What's taking them so long?" He whines... Quiet you impatient fool. If you can't find a good linux distro, then you shouldn't be writing about it.
He also mentioned that Open Source based companies like RHAT have seen big drops in their stock prices since last year.
He goes on to say, "Created by Finish college student Linus Torvalds...", incorrectly using "finish: to come to an end" instead of "Finnish: of, relating to, or characteristic of Finland" (see www.m-w.com). Not very impressive.
He also says that "94%" of the market goes to MS for desktop and laptops, "according to IDC". Hmmm... I couldn't find much information about this IDC research firm, but 94% sounds suspiciously like 100% minus 5% for the Macintosh. That can't be right... Oh, wait. Maybe they're talking about new PC's, which are bundled with an OS. What about dual-boot machines? What about that old machine that won't run MS anymore, but for which Linux has breathed new life? Since there are no license fees for Linux, how exactly do you track the number of installed users? It just doesn't add up.
I think this guy should be doing more research, and less writing.
It's not the nature of the audience, but the size which makes the biggest difference. If your music appeals only to technical people, then you have a much smaller audience of potential customers.
Slashdot Mirror
A few things to keep in mind:
In regards to the long versus short file names. FAT (actually, VFAT) is the only FS I know of that does this. The ext2 filesystem has no use for "short" file names. There are several other File Systems available on Linux, and AFAIK none of them have short and long file names. Short file names as described in this patent are a hack on the FAT file system, and are very specific to DOS and pre-NTFS Windows.
I'm beginning to wonder..... perhaps a community "Prior Art" effort, somewhat comparable to the open source community, is needed.
Identify, research and debunk such absurd/greedy patents (and perhaps eventually much or all of the software patents nonsense) so as to get the patents withdrawn/cancelled.... and/or reassigned to some "open source" holder.
I agree, except that the research for prior art should be done before the patent is issued, not after. This it the job of the US PTO, which IMO, they are doing poorly, as far as software is concerned. (If you want to help, go here and apply for a job.) The patent is supposed to be rejected unless it is "nonobvious to a person having ordinary skill in the area of technology related to the invention". Every software patent I've read has been pretty obvious.
US patent law clearly states, once the patent is issued, it is considered valid, unless prior art can be demonstrated in court. Patent lawsuits are one of the few trials where you are not considered "innocent until proven guilty". You are guilty unless proven innocent.
<soapbox> Which is why software patents should be outlawed in the US. Write your congressman, if you haven't already. The worst part of a software patent, is that it can be granted based on a description of the software, even in cases where the "inventor" didn't implement anything. Writing software is like writing a book. Copyright should apply, not Patent -- see this page. (At one time, I had plans to start my own software company, but now it appears too risky. There is no way that I could possibly know what patents I was infringing. <rant> Unless something changes, open source will cease to exist, and the only software developers left will be Microsoft and IBM, who will cross-license all of their patents to each other. </rant> ) </soapbox>
If you buy a workstation from Apple, the difference in price between the 1CPU and 2CPU version is only $600 ($2399 vs $2999).
No, I expect the "base configuration" to contain 4 processors, but probably at the slowest speed. So I would look for something with four 1.6GHz CPUS, and perhaps 2GB of RAM as the base system. (You're probably right about the high-end box costing $10,000+, but that's with probably 32GB RAM and 500GB+ of high-speed disk.) C'mon people, they need to compete with Dell and Sun here, so they're at least going to be in the same price/performance ballpark.
You haven't a clue... you can't build a 1 CPU based off of the next generation PowerPC 970 from IBM. Each chip has two CPUs.
Um, Apple's PPC970s have one CPU per chip. Presumably IBM will be using the same chips.
Actually, it was the IBM Power4 that had two CPUs per chips. The 970 has only one. (See this article)
However, the article does say, "2U two-way and 4U four-way configurations." "Way" in this context usually means CPU, so this means that there will be exactly four CPUS in the 4U configuration. So $3500 sounds like a very competitive price.
I have the COCOMO II book, and I have used the COCOMO model for certain projects. I agree that it is not appropriate here. COCOMO was designed with a narrow focus in mind, and applied best to repeatable projects in a structured work environment. It requires you to estimate parameters for factors such as "Programmer Unfamiliarity", "Precedentedness" "Development Flexibility", "Team Cohesion", "Process Maturity", "Multisite Development", etc. Each of these fudge-factors makes it extremely difficult to correctly apply the model to someone else's work.
Also, each of these factors is likely to be different for each major component.
"I was unable to find a publicly-backed average value for overhead, also called the 'wrap rate.' This value is necessary to estimate the costs of office space, equipment, overhead staff, and so on. I talked to two cost analysts, who suggested that 2.4 would be a reasonable overhead (wrap) rate."(from here)
He is using an average overhead rate for a large corporation. He forgot to take in to account the fact that Open-Source developers (generally) don't get office space or health insurance or secretaries. They use their own equipment in their own homes. So a more reasonable overhead rate for this project would be close to 0.1.
So taking all of this in to account, he's probably off by a factor of more than 100. (If you want to know how accurate he was, compare his estimate to the actual cost of developing a Linux distro... ;)
While it might have made interesting headlines, I see little value in the actual number.
SCSI vs Firewire
This technology means that if I want to upgrade my BIOS, I can't just download the latest BIOS image from DELL. Now, DELL needs to jump through the M$ hoops and get the code signed by a private key.
Sure, you might assume that many vendors will hold a signing key, and that they will all be able to sign their own code. However, you only need to compromise one master key to defeat the system. It is unlikely that there will be more than a handfull of signing keys (One each for Microsoft, Intel, AMD, and IBM).
This means than any software vendor who wants to write code that runs in kernel space (device drivers... goodbye Ethereal) or handles digital media (music, video, images... goodbye Mozilla, GIMP, etc) or interacts with any of these things, will need to be signed. Once they've won the initial battle, they sign IE, WMP, and Office, but create a bureaucracy to sign competitor's products. (You thought OS licensing was expensive?) Also, every new release will need to be signed. Sound like an open-source killer? Sounds like blatant anti-trust to me.
You might think this is not a big deal, because open source media players suck anyway. But if this catches on, then a few years from now, you won't have a choice of media players (or web browsers or whatever...) Not that it would matter which one you picked because M$ gets royalties for them all... And since they can sign a piece of code with an expiration date, they can always force you to upgrade to the latest version!!! (They call it a "feature" which requires you to apply security patches...)
The best part of all this is that it's now enforced by Hardware! Which means that you won't be able to turn it off. It could prevent you from booting from a (non-signed) floppy disk. Goodbye Linux... Goodbye BSD...
Guess I'll be getting an Apple...
$450 is a little high for a box with only 64 MB RAM, don't you think? It wouldn't even run windoze.
They should involve a PowerPC/Mac and some other CPUs/OSes in the fray. THAT would make an interesting tournament...
He also mentioned that Open Source based companies like RHAT have seen big drops in their stock prices since last year.
He goes on to say, "Created by Finish college student Linus Torvalds...", incorrectly using "finish: to come to an end" instead of "Finnish: of, relating to, or characteristic of Finland" (see www.m-w.com). Not very impressive.
He also says that "94%" of the market goes to MS for desktop and laptops, "according to IDC". Hmmm... I couldn't find much information about this IDC research firm, but 94% sounds suspiciously like 100% minus 5% for the Macintosh. That can't be right... Oh, wait. Maybe they're talking about new PC's, which are bundled with an OS. What about dual-boot machines? What about that old machine that won't run MS anymore, but for which Linux has breathed new life? Since there are no license fees for Linux, how exactly do you track the number of installed users? It just doesn't add up.
I think this guy should be doing more research, and less writing.
It's not the nature of the audience, but the size which makes the biggest difference. If your music appeals only to technical people, then you have a much smaller audience of potential customers.