After 2 years of examination the U.S Patent and Trademark Office has reversed its two earlier unofficial decisions and decided that Microsoft’s File Allocation Table file system constitutes a “novel and non-obvious” system enabling it to be patented. This coupled with Microsofts plans to charge licensing fees for use of the system could cause many problems for open-source operating systems that implement the file system, or even to mp3 players. Elsewhere, APCMag.com has an interview with Microsoft’s “open source point man” Martin Gregory.
U.S Patent Office Awards FAT Patents to Microsoft
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
Sorry for the language folks, but this is just one small example of what a f–king joke the US patent situation really is.
>Sorry for the language folks,
No need to apologize for accuracy, Laz
I agree completely. Claiming that a file allocation table is “novel and non-obvious” speaks volumes of the tiny intellect of the patent offiser in charge of this investigation.
Anybody know how we can contact this guy? I’ve got a lot of “novel and non-obvious” ideas that need patenting, such as a “hinged door on refridgerators, allowing owners to retrieve contents of said refridgerator”, or perhaps I should patent a way to style a comb-over or a way to swing on a swing… Oh, wait, those two last ones are already patented… (Patent numbers 4,022,227 and 6,368,227, if anybody is wondering.)
Yeah, I totally agree with you. The United States patent system is a “f-king joke” because the company that invented and maintained a technology is receiving the proper institutional recognition and legal control over it. What a “f-king” joke.
It’s not really a technology as such. It’s merely a connection of several algorithms. It has to be combined with other things before it becomes technology in it’s real sense.
But this is one critical difference between USA and Europe and the main reason why people and companies in Europe are fighting software patents.
Like patenting moving a pointer over a button and the action of clicking it?? Or using a link on a web page? Specially when someone else invented it?
Yes, it is a f–king joke! It halts innovation, being able to patent simple virtual actions that mimic everyday actions. Imagine if someone patented pushing a real button. Now what are VCR manufacturers going to do? Pay royalties??? Hell, even being able to manufacture a mechanical typewriter it would be probably be necessary to pay royalties!! Yes, a f–king joke!
Then they should have patented it in 1980 (or when was it first introduced). Not 25 years later, when everybody and their dog is implementing and using it.
I’m still not in clear, whether they’ve patented the whole FAT system, or just the long-names extension. However, even in the second case, they should have first patented it, then introduced it and protected the patented algorithm during it’s whole life time. Not making it almost public domain thing, having everyone use it and _THEN_ claiming a patent and licensing fees for it.
The point which makes this a f–king joke (no point in not writing “f–k”, when I mean “f–k”) is, that they want money for something what was available and unprotected for years….
I agree with this point. I had assumed that you couldn’t patent something *after* it had come into common usage. I thought that you had to protect the patent.
Can’t mod > 5.
when the invention becomes public, there’s one year grace period to patent it; FAT became public in 1981 and later versions became public the day they released; it’s disturbing that patent office is not honoring this rule in this case.
1981? I though it was part of MS Disk BASIC in 1977.
http://en.wikipedia.org/wiki/File_Allocation_Table
even more reason to use ext2, now that free ext2 filesystem drivers are available for windows: http://www.fs-driver.org/
i use it on my 160 GB external USB harddisk, no problems whatsoever
FAT is too much of a hack anyway, i don’t understand why people still use it
Every computer has it. It’s great on a pen drive.
There are several alternatives to fat32 on a pen. The problem is lack of Windows drivers for these file systems. At least we have a good freeware driver for ext2fs (and ext3fs – mounted without journalizing).
FAT is a joke. And the patent is more of a joke.
The problem is that even if there’s a driver it won’t be installed when pen drives are useful: When you are at a machine you’ve never seen before.
Then you install the driver, if you have any chance to do so – spreading the word
But so far it’s not a problem to use FAT32/VFAT.
And it’s not going to. The patent will not survive in court, and diminishes the chances for software patents in Europe.
I’m happy I’m a danish citizen
Oh yes… we all have administrator rights to every computer we encounter…
And what if the computer has no (or slow) internet access? Do we store the driver on the USB pen? How?
Interesting, but does it work with ext3? I tried a different ext2 driver for Windows and it would corrupt files that are edited (by changing the file size to some rediculously high number – you could get the file back with tail). If this works OK I will consider replacing all my FAT32 partitions with ext3. Freedom!
Anybody wanna dig up the list of prior are for me? I’m too lazy to do it at the moment.
I think this is a better demonstration of why software patents are a bad thing than it is anything else. The technology should actually be long out of patent realistically, I mean, it wasn’t yesterday’s technology: It was last month’s technology. Should it still be patented?
Not to mention, their patent, while not through the office, was out of enforcement for how many years?
I hope the courts don’t hold up any charges they seek.
“The FAT file system, a common means of storing files, was originally developed for Windows”
not very true, it was developed for DOS and it can’t store long filenames – only if you use ugly hacks.
the FAT horror caused me a lot of trouble, happy using NTFS / ext3 / reiserfs now 🙂
FAT may be available on every computer, but that does not imply that FAT has quality. we should strive for quality and progress – using a dump filesystem because everyone uses it is kind of …. dumb.
Ugly hacks? The way Microsoft implemented LFNs on top of the existing FAT was ingenious. They did it in such a way that maintained compatibility with DOS and any other OSes that used FAT, but allowed OSes that understood LFNs to take advantage of them.
Their solution was much better than just all-out breaking the format, such as is common in the Linux world.
Exactly. Because as we all know, ext3 isn’t backwards compatible with ext2.
You know I’m not talking about those. Open up your freaking mind.
Microsoft’s success can largely be attributed to how much effort they put into backwards-compatibility.
Like Office 2000 not capable of reading Office 95 format properly, or Office 97 incapable of reading files written with Office 2003.
Not to mention Publisher 97 incapable of using documents from Publisher 98.
Wow yeah. Great backwards compatibility
And it’s this lack of backwards compatibility that has ensured MS’s success, not the other way around, as LIP contends.
Him telling someone to open their minds is insanely funny.
On topic, this really won’t effect anything. Personally, I eliminated FAT a loooong time ago for my uses, and if MS wanted to enforce this, it would be easy for everyone else to do so as well. USB key makers and the like could ship EXT2 read/write drivers with their products pretty easily.
If you had actually bothered to look in the Save As menus for any Office program, you would see that you have the option of saving files in a previous version of the format.
No shit Office 97 can’t read Office 2003 if you’ve saved the files with Office 2003 compatibility only. Sure, blame Microsoft for your ignorance. Tool.
I know about “Save as…” and using it regularly.
But already here we have broken backwards compatibility.
It’s only backwards compatible if no conversion is required. Therefore the applications do not maintain backwards compatibility.
And the conversion is not at all good. Complex documents tend to screw up when saved in older formats.
If Microsoft were not using newer and better formats with new releases of their major product, then the anti-MS camp would be whining that they are backwards and clinging to ancient formats. There’s just no winning with you people.
Making Office 2003 the default save format, but providing Office 97/2000 options, is a perfectly sound configuration. If you have no specific needs for backwards compability, use the newest format with its advantages. If you NEED the compatibility, you do a bit of extra work to get it. This is no different than anything out there, whether hardware, software, or not even related to computers.
Get real.
I actually wouldn’t be complaining. Of course, some open source fanatics would probably be. But as you can see in some other of my posts, MS is not always behaving badly.
Maintaining backwards compatibility while handling new functionality is easy if the file format has been designed properly. Office 97 has major issues in regard to that one.
But as an OpenOffice.org user (apart from Access and Visio) I don’t have that problem anyway.
That’s great for you.
“The way Microsoft implemented LFNs on top of the existing FAT was ingenious.”
Sigh!
No it wasn’t. It was just one method of about six or seven equally obvious methods that could have been used.
“They did it in such a way that maintained compatibility with DOS and any other OSes that used FAT”
Yes. Could have been done in a number of ways. There were even dicussions on the internet debating the various methods … no clear winner.
“, but allowed OSes that understood LFNs to take advantage of them. ”
Sigh!
The one and only purpose of getting a patent is to **NOT ALLOW** other parties to take advantage of the method you have patented.
not very true, it was developed for DOS and it can’t store long filenames – only if you use ugly hacks.
Since the patent covers vfat then that is exactly true
.
That aside I’ve seen far worse patents than this, after all MS did invent it first, and it is an actual working “product”. It’s not like they’ve gone and patented a broad obvious technology for which they don’t actually have any working code.
I was under the impression that an eight-bit FAT was used on CP/M and Atari ST machines, and that this was increased to 12 and used in DOS (I read somewhere that floppy disks are, or were, formatted as FAT12), then 16 bits, and finally to FAT32, the 32-bit implimentation.
Not only is this example of why this is bad because they CAN patent it, but it’s also a good example because they almost HAVE to patent these kinds of things, to stop some other company from doing it and trying to sue them.
Microsoft has always been pretty good about not suing over patents, AFAIK. We’ll see what they do with this..
They are even better at patenting other persons and companies inventions. Like grouping taskbar buttons
But no wonder you’re protecting MS. You are the primary MS apologist in here
I don’t see why it’s good MS can patent it. The whole problem is they can patent it. Nobody else can patent it, because MS can prove it’s been in MSDOS since the very beginning.
1. I said it’s a problem. But it’s also a problem for MS, because the patent office is a bunch of morons and will grant a patent to someone else even if they don’t deserve it.
2. Can you reply to any of my posts without insulting me? (yes, calling me an ms apologist is an insult and you know it)
Why don’t you try sticking to replying to my points instead of getting into semantic bulls–t.
I would also normally disagree with you as you always blindly defend Microsoft, no matter what…. How about the WMF debarcle ?
Anyway, I never insult you the way Mr Jones does, that is a waste of typing.
I agree with you this time. This whole episode just shows how silly the whole papent system is in the US. I am a UK citizen and we are part of the EU. We voted against software patents as they are plainly ammunition for a company if they were ever to go up in court against another.
FAT patents ? Hmmm. Almost every peripheral with storage built in uses FAT. Cameras, MP3 players, pen-drives, removable backups etc etc
If Microsoft decided to charge a fee for using FAT, this would either force the manufacturers to use an open system like ext2, or else charge an increased amount on purchase, (which the consumer is unlikely to support), or go out of business.
There is no point saying the Microsoft simply patented FAT in case someone put in a suit. Nah, the best thing Microsoft could have done with FAT is put it in the public domain. Everyone uses it anyway.
The patent covers the encoding of long file names. It most certainly was not available in the original MS-DOS.
I don’t see why it’s good MS can patent it. The whole problem is they can patent it. Nobody else can patent it, because MS can prove it’s been in MSDOS since the very beginning.
What the hell is your point? Of course they can patent it, they invented it.
I didn’t say they can’t patent it. I merely stated I could not see what’s good about it.
And no. In Denmark they cannot patent it. According to danish laws software is algorithms, and as such not patentable.
The point is, and I will break it down into easily digestible parts so DUMB AS f–k US patent officials can understand it.
YOU
CANNOT
PATENT
AN
IDEA
OR
A
THOUGHT
Which is exactly what all computer programs are made up with.
Ignoring all the issues about software patents or non-obviousness.
How can they patent something which has been in a shipping product for 20+ years? You can’t do that with an actual invention, but you can with software? Why?
http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d…
Microsoft didn’t have to seek a patent for this.
>Microsoft has always been pretty good about not suing over patents, AFAIK. We’ll see what they do with this..
True, but Microsoft usually rules by fear of force rather than force itself. They sling the FUD of patents all the time.
This IS bad news for nearly all niche operating systems. Because FAT is *the* most widely used portable filesystem. Want to share files between various OSes? Use FAT. No more. And I’m thinking it will take a while before all these tiny niche OSes (not talking about the bigger ones) will have replaced FAT with ext2, if they do so at all.
This means OSes like FreeDOS are in serious trouble; they can’t run without infringing on a MS patent.
And, at a base level, you don’t want Microsoft to be excluded simply for ideological reasons?
No. And I guess the open source debate has been very emotional and had a lot of combative language in it, and I think the approach we’ve been on for the last 18 months has been to encourage people to make it a data-driven decision.
It might be emotional. But you just took a valid political and legal issue and failed to refer to it as such but referred to it as a side effect it often causes (emotion). A classically deceptive statement indeed.
If you think data excludes the legal circumstances you’re in for quite a surprise later on in life. For a company which, originally, made itself by a legal idea (licensed software) it’s sick for them to pretend that licensing is unimportant and not to be considered.
I’m not saying you should never use software over the license. But here’s a blatant example of where licensing has to be weighed in: Educational software pretty consistently says you cannot sell anything you create with it. Obviously you don’t want to buy photoshop educational for your ad agency!
You lose badly in one area, but instead of admitting it you use deceptive language to pretend it’s a non-issue. I just can’t describe how much I hate it when people do that!
“Educational software pretty consistently says you cannot sell anything you create with it.”
Yes, but has it ever been tested in court in the EU? Is there a case where someone has sold something, and then been sued, and had anything awarded against him/her?
Doubt very much if it would hold up if it was tried. Regardless of what the Eula says.
The patents in question cover VFAT, not FAT. The old MS-DOS file system, with the names restricted to eight characters and a three-character extension, is not covered.
LOLOLOLOLOLOLOLOLOLOLOLOLOLOLOLOLOLOLOLOLOLOLOLOLOL
Hahahaha… what a perfect joke. Apparently 2006 is going to no better than 2005. What a perfect ridicolous beginning.
First WMV and now a f–king FAT32-license hahahahahaha
Even MS isn’t foolish enough to ever consider enforcing this patent…. CALM DOWN…
Even MS isn’t foolish enough to ever consider enforcing this patent…. CALM DOWN…
Agreed, I do think Microsoft shows considerable restraint with patent enforcement; if anything, they’re usually on the defensive.
My concern, though, is the impact this could have on OSS projects like linux. While I don’t think MS would necessarily enforce the patent, the mere fact that it now exists is complication enough.
Then again, MS has been collecting liecnsing fees AFAIK from some flash card manufacturers using FAT, so who knows. It’s hard to collect licensing fees from one party while looking the other way with others.
Who knows.
It will not have any effect on Linux, as they do not use FAT for anything except reading DOS and Windows disks.
A fully functioning Linux system can be built without any DOS/Windows compatability in it, and there would be no loss of functionality.
You see, the point is this. Microsoft employees think the computer world revolves around the company. to a certain extent they are correct. However, in the not to distant future, Microsoft, its lawyers and its employees would have themselves put the company out into the cold. Alone.
We dont need to go online and try and destroy Microsofts reputation, they are doing a pretty good job of that themselves.
Then they f–king don’t need that patent, do they!
Whatever, it’s only a problem for US citizens. Clean up your own mess.
Code is algorithms is math which cannot be patented.
Code is algorithms is math which cannot be patented.
True. By the similar token, though, the space shuttle is machinery is molecules is atoms which cannot be patented 🙂
No, the fundamental bases for code are not patentable, but that doesn’t mean the finished product isn’t. It’s when it comes to stories like these that it becomes debatable whether or not that’s just.
If the finished product is a hardware device than go patent it. I don’t care.
The software behind it though. That cannot be patented. It’s nothing but an algorithm.
Sometimes Denmark is wonderful country, though somethings are rotten
No point arguing with a US citizen. They get indoctrinated from they first attend school to believe everything their governments decide is true and just.
Look at the way they are forced to pledge allegience to the flag every morning.
In fact, how many US citizens here do not support software patents for any company ?
I agree with you, apart from the “pledge allegience to the flag”-part. That doesn’t seem so bad to me. The trick is to distinguish between the nation, and the leaders and laws of said nation.
Here’s one! Hi from the States…
“Then they f–king don’t need that patent, do they!”
YEAH they f–kin DO……
It’s to keep some dickhead from suing them. This more a matter of self-defense than anything else.
Suing them over what?
Apparently it can be in the US.
Yup, and their patent system is broken – even in the eyes of Microsoft.
Some months ago Microsoft released a paper with many good ideas on how to improve the situation for patents in USA. Only on one point did I disagree with MS, basically because that part would made things even worse than today. But all the other ideas were really good.
This is just another example of software patents being bad news.
It’d be interesting to hear from Red Hat on this issue, I imagine that this news will mean VFAT will have to be removed from Fedora kernels.
This is a double edged sword. The less available FAT is, the better. It can hurt a lot of projects and small companies (which is horrible), but the long term net effect for users is that we move on from an ancient technology that wasn’t even great in its own time to something better.
The less available FAT is, the better.
Perhaps. People appreciate the plug-and-play aspects of FAT-formatted devices today – no drivers necessary. Now peripheral makers are reduced to come pleading to MS if they want their devices to “just work” with the today’s dominant OS. Microsoft now has a potential stranglehold on both sides of the equation. How this plays out in the end is anyone’s guess.
Actually, all the peripheral manufacturers need to do is provide their devices unformatted… The Windows formatter will do the rest…
For those of you who think they will enforce it or scare people with it, can you show prior cases where they have done such a thing when it was completely unjustified?
Microsoft has already entered into licensing agreements with third-parties. For all you know, Microsoft simply intends to extract license fees from commercial users of vfat and have the rest of the effect of the patent be a spectre over competition that supports long file names, like Linux. They don’t ever have to file a lawsuit.
I really wish I could edit that.
And for all you no, they don’t.
You’re right, they could look to actually sue.
Or they might not.
Which is what I said.
Or maybe you didn’t…
I seem to recall stating that they need not file a lawsuit. If that wasn’t me, then who’s been inserting memories into my head?
I was just playing around
With my memories? I think I’ve seen that movie.
effect of the patent be a spectre over competition that supports long file names, like Linux.
Linux use of long filenames is nothing to do with FAT/VFAT or in fact any Microsoft product.
Linux, being UNIX based, has used long filenames since inception. UNIX has also used long filenames since long before DOS1.
If there is a patent for long filenames, it will be held by some UNIX company, (maybe SCO?) who at this minute could be getting the lawyers warmed up for a fight with Microsoft.
The whole thing is silly.
“If there is a patent for long filenames, it will be held by some UNIX company, (maybe SCO?)”
Two problems with this thought.
(1) Long filenames is the default. It is only itsy-bitsy old floppy-based operating systems (like CPM and MSDOS) that had the 8.3 filename restrictions in the first place.
(2) SCO does not have any patents. Try AT&T, USL, Novell, Sun, IBM or some Unix company (almost any Unix company other than SCO) that does actually hold some IP.
If you read the actual patent it’s not about the concept of long filenames. It’s about the implementation of long filenames within FAT.
Do you have some allergy to being informed?
you sir, are a f–ktard.
I mentioned long filenames as a REPLY, to a PREVIOUS POST……
knock, knock… can you hear some sense trying to get into your thick head ?
wanker
“For those of you who think they will enforce it or scare people with it, can you show prior cases where they have done such a thing when it was completely unjustified?”
Getting this patent in the first place is completely unjustiified.
The patent covers the manner in which “long filenames” were grafted onto the FAT32 filesystem. There is nothing at all innovative about it. There were many alternatives on how to do this … Microsoft just chose this particular scheme (out of many posibilities all of which would have done just as good a job) … then they patented this particular scheme.
There can be only one reason for doing this … to screw everyone else, to attempt to eliminate competitors, and to limit cross platform interoperability.
Microsoft does not do interoperability. In fact, Microsoft deliberately does non-interoperability in an attempt to force everyone to use their system.
It is not valid to offer as an excuse … but someone else would have patented it, and then sued Microsoft. The reason why that is not valid? … Microsoft could easily have shown prior art.
This whole exercise has “Microsoft monopoly protection behaviour” written all over it.
You’re making a lot of assumptions about what they are planning to do.
NO, suing people is not the only possible reason. You may say it is the most likely reason if you wish, because that can be argued. But only? I don’t think so.
Does this apply to all versions of FAT (FAT12, FAT16, FAT32)?
Whether or not you open-source fanboys agree with the patent is irrelevant — the patent covers the design of the file system, not any particular code as far as I can see. The simple truth of the matter is that FAT/VFAT *is* a definitive design, and as such, Microsoft has every right to patent it.
Patenting the idea of a “file allocation table” is asinine, but as long as the patent only deals with the exact design and specs of FAT, it is valid.
Despite providing a link for the patent in question, no one seems to want to read it.
I’d tend to agree with your statement.
It is however unfortunate that FAT/VFAT is seen as a standard in order to communicate to most electronic devices, and such should be covered under some sort of agreement to permit its free use.
I don’t have a problem allowing MS to ask for a once off license fee for its use, it was something along the lines of 250k for each factory.
Although from a Linux perspective this certainly does put into question how the vendors will handle it.
As someone already said, Microsoft tends not to enforce its patents, and I doubt they would now, seeing as how they want to give off a “we’re friendly” image.
In any case, the only reason FAT is such a standard is because it’s extremely easy to implement it. I’m sure some CS graduates could write a FAT implementation with relative ease.
Yes, it’s very easy to implement. For one of my projects (building a GPS device) it was easier to reimplement FAT than try to port an existing implementation.
To implement a read-only FAT12/16/32 with no VFAT it took less than two days by one coder (me), and there were only two major bugs. (Read only with no VFAT was all that was needed for our project.)
They have the right to it in USA, because your system is so f–king screwed up. You can patent math!?
It’s so silly you have to be one of the “The war is all about oil and nothing else”-antiwar hippies to see any logic in patenting math.
What devices cannot be described using mathematics?
Any device can be described using math, but the description is irrelevant.
A software patent is a patent on an algorithm.
A hardware patent is a patent on physical device. The mathematical description is irrelevant since we are patenting the hardware and not the algorithm describing it.
The description is not patented.
I’ve read a lot of your posts in this thread and I came to the conclusion that you are retarded. If you need any help seeking prefosseional help email me.
“I’ve read a lot of your posts in this thread and I came to the conclusion that you are retarded. If you need any help seeking professional help email me.”
Everyone gets a chance to make a fool of themselves here, some just have more practice than others.
Retarded? Well, the border line between genius and madman is pretty thin. Perhaps I’m a madman after all. Wouldn’t be so mad.
If you think you can help me, then please feel free to do so.
In the meanwhile I’ll be happy knowing I’m not like the rest of you.
There is a lot of good things to be said about being a madman
The fact that your post can be modded up to 2 despite the fact it is offensive, says a lot about those who voted.
They can’t even vote according to the simple rules of this forum. That could be considered retarded behaviour. And so could your post, since you are insulting another person on the personal level. Simply because you do agree.
But feel free to call me a madman. I prefer that when it comes from you.
The description of the operation of any physical device is an algorithm. You are not patenting the physical device, you’re patenting the invention of collection of properties and algorithms performed by the device. You don’t secure protection for your ice cream scoup, you secure protection for the ideas contained within your ice cream scoup. All of which can be expressed in terms of mathematics. In fact the entire purpose of the patent was to make public all of that information you find so irrelevant in exchange for a government-protected monopoly.
You seem to misunderstand me.
What I was trying to say was that it’s not the description which is being patented. Therefore it’s not algorithms that are patented in the case of hardware devices and their functioning (the idea behind), but the idea it self. The fact it can be described as an algorithm is irrelevant in this discussion. The question is whether the patented idea in itself is an algorithm or not.
Software is nothing but algorithms. An idea for a hardware device is described by algorithms but is not in itself an algorithm. There is a difference here, whether you patent the idea of an algorithm or patenting the idea of a hardware device described via algorithms.
The first one is an algorithm, the latter one isn’t.
You should read some actual patents.
http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d…
http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d…
http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d…
Patents are informal specifications for ‘inventions’ that typically perform a set of operations in a certain manner. The operation of the device is an algorithm. Whether this device happens to be a compressor or a computer configured to perform some operation. You are restricting the use of the ideas behind this ‘invention’ which include any relevant algorithms it embodies.
If you want to hand-calculate the resolution of long file names performed by Microsoft’s procedure, then have at it. If you want to use the algorithm to open an automatic door in patterns, enjoy. But that isn’t what you want, and it’s not what you’re complaining about. Well, maybe some people don’t want the way their compressors can operate to be restricted, either.
Those 3 patents are not software patents. And as such I do not oppose to them.
No, they aren’t. Yet you don’t seem to understand what patents really patent.
The point is that software isn’t something described using mathematics, it IS mathematics.
1. I don’t live in the US.
2. The war is all about oil, but if you ask many Americans, they’ll swear that Iraq was the attacked on 9/11.
The patent is for the specifications and design of the FAT file system, and probably the official Microsoft code for it. They have as much right to patent FAT as a car manufacturer does some specific type of combustion engine enhancement.
1) Lucky you :p
2) And I’ll say the opposite.
Software is algorithms and math cannot be patented. There FAT cannot be patented. FAT is nothing but algorithms.
I’m so happy I live in Denmark. We aren’t encumbered with software patents (that’s why I legally can use FreeType2 with BCI enabled).
Software patents are a crime against humanity. If you think otherwise then please pay me, and I’ll give you a license for use of my patented add-algorithm.
2+2=4 and it’s my invention. Though of it my self. Please pay me 100000000€ and put them on an account in Nigeria.
You have a very narrow-minded view of things.
2+2=4 is not an algorithm. It’s a statement. And I’ve said it once, but I’ll say it again for your benefit: There are specifications and design that go into a file system. Those have been patented, and with every right.
Saying that FAT is an algorithm and merely resting your entire argument on that is weak, because everything can be described through math. That Pentium 4 processor? Yeah, algorithms … just implemented with transistors instead of pencil, paper, and mind.
Are you going to tell me that a processor can’t be patented now because it’s just math as well? Go away.
Of course a processor can be patented. It’s not an algorithm. It can be described via algorithms but is not in it self an algorithm.
Software is algorithms described by algorithms. Specification of algorithms are not patentable. And is not by any right.
This is a key difference between USA (for an example) and most european countries.
FAT is merely a combination of several algorithms and as such cannot be patented. FAT is not a physical device with a functionality described by algorithms, but merely a combination of algorithms.
A design in it self not patentable (at least not in societies with a free market). Ideas aren’t property.
A processor is just as much an algorithm as a file system is, if not more so. The processor, before transistors are placed on silicon, is a set of ideas, specifications, and algorithms. A file system, before the code is written and a volume formatted with the file system, is a set of ideas, specifications, and algorithms.
Please, don’t be so narrow-minded. I’ll explain for the millionth time: There is more to a file system than just algorithms. Lots of thinking, planning, and testing goes into file system design. The final product of a file system design can merely be a very detailed document detailing how, where, when, and what data is stored, how, where, and when it is to be manipulated, etc.
You could hand such a document to any number of coders and have them produce code which does those exact things — but the document itself, and ideas, and designs, and specifications detailed within would be patentable. A patent does not have to apply to a tangible, physical something.
Microsoft isn’t patenting FAT. They patented what they believe to be a novel approach to supporting long filenames while remaining compatible with FAT. The whole discussion about algorithms and whatnot sort of sidesteps the issue. It’s easy to program a general-purpose computer to perform an arbitrary set of processes, that if there were no general-purpose computers, would be performed by a bunch of specialized devices. It would be sort of like if you had some magic ball of goo, that you could instruct on how to become a new kind of door lock, or a new kind of automatic vacuum cleaner, or a new kind of airplane wing. You would instruct this ball of goo, using algorithms, to take on the properties and perform the necessary operations for completing a task. So everyone has these balls of goo in their home, and whenever they want a new kind of fabric, or a new kind of airplane wing they download descriptions that permit the goo to do whatever they need.
So are the devices that this magic goo transorms into still novel-enough to warrant patent protection? Does your Never Squeeks(tm) Hinge lose its novelty, because everyone has magic goo fabrication in their house?
Does your Never Squeeks(tm) Hinge lose its novelty, because everyone has magic goo fabrication in their house?
Yes.
Microsoft isn’t patenting FAT. They patented what they believe to be a novel approach to supporting long filenames while remaining compatible with FAT.
Like someone pointed out, that method already exists in UNIX long before Microsoft themselves existed thus nothing new.
No one has pointed that out. Someone simply stated that UNIX supported long filenames. So what? Read Microsoft’s patent. It doesn’t cover the concept of long filenames. It patents the manner in which long filenames are patented within FAT. If there exists prior art that makes their approach not novel (which was one of the reasons cited when Microsoft was initially denied these patents) that could be presented, but no one here seems to bother learning-enough about a subject before commenting to provide that.
That should read, “are implemented.” Patent patent patent. I don’t even like patents, and I apparently have the word patent on the brain.
“They patented what they believe to be a novel approach to supporting long filenames while remaining compatible with FAT.”
The problem with this thought is that Microsofts approach is not “novel”.
Microsoft’s approach is just one of a number of fairly obvious and straightforward possibilities that could have been used. Microsoft chose this one.
There is no point in Microsoft wanting to patent it, other than it lets Microsoft make it more difficult for competitiors to be compatible with Microsoft software.
The validity of the patent has long since ceased to be the issue here.
A proccessor is a physical device that implements certian algorithms.
Software is algorithms, no matter how much thought and testing went into developing those algorithms.
It should not be possible to patent algorithms.
FAT is merely a specification for arranging information. Translation of long file names by encoding them in a manner compatible with this specification is a process. A computer that performs this process is a device.
BTW: I never said 2+2=4 was an algorithm.
You were talking about algorithms not being patentable, then provided “2+2=4” as a cheap example.
Nice try.
The way dylansmrjones keeps referring to algorythms and descriptions of algorythms baffled me a bit too. But I think that he’s trying to make the point that it’s weird that you patent specs about a software rather than the software implementation itself.
In your own example: it’s ok if Intel provides the detailed description of a chip or a part of it, let’s say an ALU, and patent their implementation. It’s not ok if they patent all conceiveble ALUs that go under those specs. It’s ok, though rather silly, if Apple patents _their_ code to implement spring-loaded folder. It’s not ok to patent the _idea_ of spring-loaded folders.
In the same way, one might say that it’s ok if MS patents their implementation of a filesystem with those specs, not the specs themselves. That’s how patents are treated in most european countries, as far as I know.
The point is debatable of course, because you have to walk the thin line between encouraging evolutionary innovation, that just like in academia needs open concurrency of ideas and implementations, and rewarding great, new ideas.
The point is that patents really come from a simpler time when they basically applied to novel mechanical applications of _known_ physics laws. Today patents are applied to anything from software to molecular chemistry, and they are as good at that as fish are good at riding bikes. They werent born and did not evolve for that.
Do you remember the big fuss about patents and costs of AIDS treating meds in third world? Several industries out of US started producing patented substances. In their countries that was alright, because in their legislation you can patent the _process_ that gives you a given molecule, not the substance itself.
To sum up: patents on software are controversial. Just saying that you have the right to patent any set of numbers you come up with, or any general idea, or any natural law or process or mathematical algorythm you find is silly. Saying that patents should be downright abolished for software only is also silly, given the economy of the world we live in. Let’s deal with the shades of grey, and hope a decent treatment of software patents is soon estabilished, to guarantee all of us from the preemtpive carpet-bombing patenting and constant legal litigation that only huge US companies could sustain.
Edited a few typos
Edited 2006-01-11 07:54
Though base-10 arithmetic is a set of algorithms. That’s basically where the word stems from.
Is this decision challengeable in the US Federal High courts? It is clear that Microsof placed undue pressure on staff at the USPTO to get the decision in their favour. And – this decision only applies to the US, and countries which abide by US style patents. Europe simply should not recognise this as a valid patent/reject it in direct defiance of Microsoft and the USPTO. The time has come to start punishing the US government economically, since it’s clear that it only cares about US businesses interests and will twist and turn to keep rich businesses rich.
Dave
The time has come to start punishing the US government economically, since it’s clear that it only cares about US businesses interests and will twist and turn to keep rich businesses rich.
So enlighten me, who should the US government care after?
And what kind of punishment do you propose?
is a link to the actual patent…
Browser: w3m/0.5.1
Does this mean fat32 support will be taken out of future linux kernels?
Not that it’s a huge deal for me because I haven’t had Windows installed on any of my home machines for quite some time, but I could possibly see this as an inconvenience for a linux newb, and the last thing a linux noob needs is another inconvenience.
Anyway, this kind of sucks in the short term (maybe) but in the long term I think it’s a good thing to put the FAT file system in the grave where it belongs.
I believe there should be a standard filesystem that works with most if not all os’s (for portable filesystem purposes: pen drives, portable hd’s etc) and it should be under bsd, gpl, or some similar liscensing that windows, mac, linux, bsd, and all the little guys can deal with.
Edited 2006-01-11 02:45
I believe there should be a standard filesystem that works with most if not all os’s (for portable filesystem purposes: pen drives, portable hd’s etc) and it should be under bsd, gpl, or some similar liscensing that windows, mac, linux, bsd, and all the little guys can deal with.
Probably every OS already supports FAT. The OS devs will possibly have to pay some licensing fee to use it. What’s wrong with that?
A lot.
But you are so demented you cannot see that (If I’m retarded by your standards, you must surely be demented by my standards).
They should have patented it when they first discovered the solution, rather than several years later.
It should at least be free for all existing implentations since they have been created before the patent came to exist.
It’s purely MS’ own fault they did not get the patent earlier.
And the fact alone VFAT is so wide spread the patent should not be granted – get the patent _before_ you publish anything. And not afterwards. Apart from that it’s merely algorithms.
People who like patents are people who like controlling other persons. And as such patents are generally bad, making competition more difficult, basically turning the free market society into a communist-styled planned economy society.
They should have patented it when they first discovered the solution, rather than several years later.
And spring the trap early? Surely you jest.
From your previous posts, it appears that you are the retarded one around here, or are a Microsoft employee.
Probably every OS already supports FAT. The OS devs will possibly have to pay some licensing fee to use it. What’s wrong with that?
Clearly this shows through all your anti-linux posts that you have never actually USED linux before.
FAT by default is not part of a Linux setup. It can be added to provide compatability, but it definately is not needed. Why should a Linux developer pay Microsoft to use its patented components, when linux does not actually need to use them ?
FAT by default is not part of a Linux setup
Huh? It’s part of the kernel, virtually 100% of distributors choose to have it enabled. Not part of a Linux setup?
No, FAT under Linux is actually a kernel module. The kernel module does not have to be added and you are incorrect saying that virtually 100% of distributers have it enabled.
The “vast majority” do in fact enable this module, but it is not needed to have a fully functioning system. It is only needed for cross platform compatability.
sorry for being so pedantic, but clearly, these things need to be clarified, or there will be loads of patent suits in the future.
Clearly this shows through all your anti-linux posts that you have never actually USED linux before.
Wow. Just wow.
Why should a Linux developer pay Microsoft to use its patented components, when linux does not actually need to use them ?
Bingo!
> Probably every OS already supports FAT. The OS devs will possibly have to pay some licensing fee to use it. What’s wrong with that?
The problem is that software patents make the copyright obsolete. The patents are there to protect inventions which were NOT protected before. Just think of the hovercraft – can you copyright it? I didn’t see any real replicators in-action until today (but I’m sure atom chip has one in their labs
).
Software is already protected by the copyright. What is the point of copyright on software if you can just protect it by patents? Actually (software) patents are something synthetical developed by big concerns to “protect” their already protected property to get more fees. The software patents are not about actually DEVELOPING something, it is about taking your juridifical vocabulary and writing down your ideas in such a language that those simple thoughts seem to become very complicated.
Or better tell me: what’s the difference between a double-click on a mouse and a double-click on your audio recorder to jump to the next song instead of seeking? I bet Microsoft was developing the double-click since their first day of existance… Actually we were already doubleclicking in computer-games long before microsoft “invented” the double-click. The device was … aah, I don’t remember the right name … something you surely can see in a museum … aah, now it is here: keyboard (or optionaly a gamepad/joystick).
Compression, protocols and everything else is nothing but maths or text. Remember: I filed a patent on that comment since it is an idea and I doubt the copyright is enough to protect it.
Patent 10,283,495,192,189384593.12:
a communication method where two parties are aware of same information and one party has more than one information for the same purpose to make a choice of. The other party sends a special information derived from its own choice to its party. The packet’s purpose is to initiate a choice of the right information on receiver’s side from its information pool.
Protocol:
<word>={[a…zA…Z]+}
<comma>={,}
<EndOfMessage>={?}
<Space>={ }
<Convict>={<comma><word>+<EndOfMessage>}
<Information>={<word>[<space><word>]*<Convi ct>}
Notice: Convict is a simple stack of words with no special meaning. They are only there to create an <EndOfMessage>.
OR SAID VERY SIMPLE: IT IS A RHETORICAL QUESTION.
Same for all those software patents which “protect” ideas and no inventions.
Edited 2006-01-11 12:26
The most common theory is that they want to get royalties from the FAT patents…
Royalties on Flash ? No, they will not get any. From now on, Flash drives will come unformatted, manufacturers will leave that to the user.
However, dual-booting Linux/Windows users are likely to be affected by this. The “normal” way of sharing data in a dual-booting workstation is to have a FAT32 partition used as a buffer between a NTFS 2000/XP install and a ReiserFS/JFS/ext3fs/whatever Linux system. Users can copy data to the FAT32 partition to have it accessible to both Windows and Linux. The reason is that Windows does not support anything else
than NTFS and FAT/FAT32, while Linux still has uncomplete NTFS support (creating files still does not work).
It seems like Microsoft is aiming at those dual-booting users, making using Linux alongside Windows, or switching to Linux even more difficult.
NB : you may replace Linux by *BSD at your leisure
A UDF volume on disk, and native write support from Windows would make everyone happy.
Actually there are some really good freeware drivers for Windows which enables read/write support for ext2 and ext3 (the latter one mounted as ext2 – e.g. no journalizing).
Nope, there are a few device drivers for Windows that lets you look at ext2/ext3 filesystems.
Microsoft is the one who never plays ball
If the finished product is a hardware device than go patent it. I don’t care.
The software behind it though. That cannot be patented. It’s nothing but an algorithm.
Personally, I think I’m of similar mind. I think.
Sometimes Denmark is wonderful country, though somethings are rotten
Your statement would be more modular if you replaced Denmark with any variable country. And just as accurate 🙂
Now the U.S. Government is helping Microsoft to further their monopoly by awarding them this patent.
what’s next Bill Gates for President.
I’m convinced I will be spending an equal amount of time worrying about Bill Gates running for President, and Microsoft being a monopoly.
It would put an end to high-tech warfare at the very least.
No cruiser missile utilizing MS software would be capable of hitting anything
Is that it is both patented, and copyrighted. Right now software is the only thing I can think of that is able to be protected simultaneously by both. It gives too much power to the holder of the ip, especially when the patent is on ‘any device or mechanism that…’, or when the patent covers the use of a general mathmatical concept such as a fourier transform in regard to a certain task.
One question, does microsofts patent start from today (expiring in 2021), or does it start from when they initially ‘invented’ the method of using long file names on fat? (expiring in 2009 or so)?
It seems a ridiculous amount of time and money spent if they’re only getting 3 years of protection for it.
“It seems a ridiculous amount of time and money spent if they’re only getting 3 years of protection for it.”
It is probably worth it from Microsoft’s point of view if they can make other systems less interoperable with Microsoft. AFAIK modern Microsoft OSes only support two type of filesystems: VFAT and NTFS. NTFS is “protected” from other systems being interoperable with it by virtue of being a trade secret. Up until now, VFAT has not been protected. This means that other OSes are able to read and write VFAT disks … up until now.
Microsoft is hoping to get a bit more “lock-in” out of this. They don’t want people to be able to set up dual boot systems with a common VFAT partition to store data which can be used by any of the OSes (even non-Microsoft ones).
Hehe, this may be a bit farfetched, but i hope the EU manages to force MS to allow other 3rd party partitions to be included in windows.. As in, to stick the drivers for ext2/3 into windows. Something similar to what they’re doing with their media player.
But then again, i wonder if that’s really such a good idea . Will it encourage open source and attract people to *nix, or will it excite users to stick to windows since it’s starting to support more linux technology..
Anyway, just a thought.. back to work i go.
As has been pointed out DOS/FAT filesytem is a very simple one. If we want to replace it, what is the next simplest filesystem to write to add to alt OSes?
People suggest ext2 and ext3, but are these the simplest filesystems to write that will also do the job, or would developing a still simpler FS be worth while considering?
The development of a simple filesystem is not difficult. The difficulty is providing interoperability with other systems. FAT has taken on the role that ISO9660 and UDF fill by providing platform-neutral filesysem access for a variety of devices. So you can develop a simple filesystem, and it doesn’t decrease the headache for people that can’t use it because Windows doesn’t ship with support for it. You can write and distribute a version of the filesysem for Windows, but then people have to know enough to need to want to get it.
If Microsoft supported writing to UDF volumes in Windows, that would go a long way toward making this all less important. Considering that FAT is seriously inadequate for large volumes, they should do this anyway.
All you have to to understand:
Microsoft is not evil company!
Never they would enforce patents onto deserving loyal customers.
Its only to protect them from evil companies that sue with patents like IBM!
Microsoft does all hard work and innovation. They want protect not only them but you as user too!
Never you can get sued by folks like SCO that just claim patent!
This is great news, me as user is protected, you as user are protected! Microsoft is all about protecting its users!
what a f–king sheep
You do realize Microsoft is financing SCO/Calderas ligitation game?
All you have to to understand:
Microsoft is not evil company!
Never they would enforce patents onto deserving loyal customers.
Its only to protect them from evil companies that sue with patents like IBM!
Microsoft does all hard work and innovation. They want protect not only them but you as user too!
Never you can get sued by folks like SCO that just claim patent!
This is great news, me as user is protected, you as user are protected! Microsoft is all about protecting its users!
Dont you think your sarcasm is a little over the top ?
Speaking of pen drives, wouldn’t shipping them unformatted sidestep the issue? I mean, even floppy disks started coming unformatted in the beginning
Memory cards on the other hand could also come unformatted, leave it to the camera, mobile phone, whatnot format it, ofcourse then the royalty whould shift to the other side, but you would only pay it once
Number one, we don’t compete with open source … as a development model or a technology model.
sounds like he is really saying. We will start patent everything that is related to software and charge the open source a huge license fee, or we will sue them with the help of our 50billion dollar bank account.
Some remarks
– AFAIK, the patent concerns the handling of long filenames in FAT filesystems, not the FAT filesystem itself.
– The principle of operation of FAT is quite good for size limited devices, it is superior to ext2 for example, except for the broken filename handling and file ownership attributes.
– Maybe a filesystem targeted to FLASH medias should try to balance the number of writes in the whole area because FLASH/EEPROM cells are write/erasure cycles limited. The FAT Table is modified after all writes in the FAT filesystem which is no good for FLASH ( but maybe actual flash disk chips have provisions for enabling extra writes in that region, I don’t know )
– The FAT filesystem was derived from Digital Research CP/M filesystem, from day 1 the limited filename length was a problem ( back at that time, 25+ years ago, Apple ][ DOS wasn’t limited ). Many ways to overcome that problem was developped, for example OS/2 and the CDROM “RockRidge ISO9660” used an extra file with long filenames or filenames with Upper/lowercase caracters, spaces and accents. There are still problems between Windows and Linux concerning the range of allowable characters and filenames on CDROMs.
– There were many previous works of dealing with FAT problems anterior to Microsoft’s solution, these ways weren’t mainstream because of 2 reasons :
– Microsoft did never support these extensions.
– The best way to handle that was to slightly alter the format of directory sectors, and Microsoft could have developped a new version of DOS that could totally break the compatibility with ‘enriched content’ directories : The separate file concept of ‘RockRidge’ allows full backward compatibility whereas altering directories can break it if Microsoft wants it.
What’s impressive is that Microsoft delivered a broken format, everyone knew it, everyone could have developped a new format or alter that format easily but only Microsoft could fix it in a fully Microsoft-compatible way.
– I think that some flash card vendors have officially licenced the FAT filesystem to Microsoft ( Sandisk )
There is a general threat with formats controlled too heavily by a single entity, for example, more and more PDF files ( datasheets, … ) only works with the latest version of Adobe Acrobat Reader : You will be forever allowed to produce PDF files but with the complexification of the PDF format, it will be more and more difficult to process these files. In such domains, the differences pushed by RMS and the FSF between free and ‘libre’ are cleary shown : Basically FAT and PDF are free use, but not ‘libre’, a malicious corporation can screw you at any time.
Implement a FAT32 filesystem that writes ONLY the long file name – it does not also try to fill out the 8.3 part of the filename using “DOCUME~1.DOC” – perhaps it just uses “00000012.DAT” or whatever – not at all the same method that Microsoft has patented.
Since no-one uses the 8.3 version of the filename anyway – could this approach form the basis of a cross-platform compatible but non-infringing implementation?
Why won’t our government do anything to protect us from Microsoft’s monopoly? They have failed us at every opportunity. Something has to be done! If government will not, the people must! We must have techno-economical freedom!
So far, but with people like Bent Bensen in goverment we are still in danger for having swpats in the future. Especially with EU.
In a completely un-related story they handed out some FAT checks to a whole bunch of lobbyists…
I guess some would like this announcement to read “US Patent Office awards Microsoft a license to print money”. Bit I’m not so sure it will come to that.
As we know from recent spats with the EU and South Korea, Microsoft faces plenty of states and outfits around the world that would love to hit them up for a few hundred million or inundate them with bad publicity. Getting heavy with the FAT stuff would be a great excuse, so the chances are Microsoft will be smart and not get provocative.
Looks to me as if this is more a shift of emphasis as computing moves into a host of peripherals and smart devices. Control the file system and access to it on those devices – whether a cellphone, a USB stick or a media center – and you start to control the market. The money isn’t in exacting 25c patent royalties. It’s in using that control to leverage getting your software and OS into those devices.
Watch out.
Why does it matter? I dont use fat except for my camera and usb stick and they will have to take that up with canon and PNY… Patent away MS just means companies will start using alternatives! Shooting yourself in the foot AGAIN!
Think this all the way thru and you will realize this is not going to benefit MS at all, in the long or short run!
This wont apply to much anyway. I have to pay a extra quarter for my two hundred dollar camera – whoppeee
If it does not benefit Microsoft, why did USPTO grant VFAT patent to Microsoft 20 years later not earlier?
this has nothing to do with another OS being able to read/write fat…
this patent has to do with DEVICES that are FAT devices, that use fat for its filesystem. media, cameras and so forth…not linux, OSX, BSD, or any other OS!
The company that makes these devices will have to license FAT, not me, not linux! And last I checked LEXAR already had a license and have you seen outragous prices, nah! I think it is a .25 per device to license it.
Wake up and stop arguing over nothing this is nothing adn as I said will not benefit MS at all…
we ought to license clues… get one people
Amomg all this wondrous invective hasn’t a Great Big Point been missed?
If MS start making too much of a fuss about, for examole, FAT32 surely IBM or (insert tech company of your choice) will sue them for some-patent-or-other on a tit-for-tat basis?
All this would achieve is that we would end up with an IT industry that has ground to a standstill and is wholly owned by lawyers – then things would get REALLY bad. So they (probably) wont.