Torvalds Invalidates Microsoft FAT Patent

[Belazor]

Yeah, when I had to get files off a ext4 drive I ended up just making myself an Ubuntu LiveCD and copying files via LAN :p

Slower, but it worked :)

Thanks for the suggestion of ext2fsd btw, appreciated :D

[simplezz]

Yeah, when I had to get files off a ext4 drive I ended up just making myself an Ubuntu LiveCD and copying files via LAN :p

That can do the job too. I also like to use WinSCP sometimes for transferring ext4 files from a Linux machine running sshd.

Thanks for the suggestion of ext2fsd btw, appreciated :D

Not as good as proper ext4 support, but it's about as good as you'll get on Windows :(

[Mr. Gibs]

Source

FAT is better supported across platforms than NTFS. It's also simpler and more compact. Ext4 is its true successor though.

Sorry a discussion on a newsgroup doesn't really count as printed publication. If it did, then I claim that in the future spaceships that can travel at the speed of light by capturing it and then releasing it. If in the future spaceships do end up doing that, then I want $600 billion from each engine manufacturer of implementing something I said on this forum.

Rather stupid example, and probably completely inaccurate, sure but you see what I mean? You really think if in the future that came to happen, GE, Rolls Royce etc would end up paying me royalties of $600 billion just because of something I said on this forum in 2012?

[The_Decryptor Veteran]

Torvalds post describes implementing long file names by stuffing it into multiple directory entries in the file system, so an OS capable of understanding them would show the new file name vs. the old one.

Microsoft's patent describes implementing long file names by stuffing it into multiple directory entries in the file system, so an OS capable of understanding them would show the new file name vs. the old one.

See the overlap? Torvalds didn't just claim "In the future we can use long file names!", he described an actual method of doing it before Microsoft patented a similar method.

[Mr. Gibs]

Torvalds post describes implementing long file names by stuffing it into multiple directory entries in the file system, so an OS capable of understanding them would show the new file name vs. the old one.

Microsoft's patent describes implementing long file names by stuffing it into multiple directory entries in the file system, so an OS capable of understanding them would show the new file name vs. the old one.

See the overlap? Torvalds didn't just claim "In the future we can use long file names!", he described an actual method of doing it before Microsoft patented a similar method.

Hm I see. The other thing is OS/2's (which was co-developed by Microsoft) file system HPFS supported LFN and that came out in like 1989 or something. I don't know how it was implemented, so it could be different, but can't Microsoft claim they were working on it since that (and have proof of it)? Just because the patent wasn't filed till 1995 doesn't mean it wasn't worked on.

[The_Decryptor Veteran]

HPFS might have just supported long file names from the get go though, the issue is with FAT because it only supported 11 characters max, they worked around that by splitting the name over multiple entries.

[TEX4S]

So do "poor man's patents" no longer work for prior art ?

Design something, sign it and then mail it to yourself & dont open it - once through US mail it becomes a federal document..... this no good anymore ?

[rfirth]

So do "poor man's patents" no longer work for prior art ?

Design something, sign it and then mail it to yourself & dont open it - once through US mail it becomes a federal document..... this no good anymore ?

No. Your prior art has to be public knowledge. The whole point of the patent system (other than to grant a legal monopoly) is to disclose inventions.

[Gerowen]

A judge in England has little power over international proceedings other than to voice opinion, and maybe establish some precedent, with regards to foreign patent(s).

Last I heard, FAT was a Microsoft patent in the USA; if USA judges uphold the patent, UK, and elsewhere must enforce the judgment.

Why is that? Are we special and get to tell other countries what to do, but don't have to listen to them?

[TEX4S]

Why is that? Are we special and get to tell other countries what to do, but don't have to listen to them?

No, but different countries have different laws, the US Patent & Trademark Office only deals with U.S. patents.

BUT, if other countries all think 1 way on a patent claim, it might show case precedent & affect U.S. interpretation of the patent.

For instance, some companies have different interpretations about a monopoly, or a S-Corp - so what happens here, stays here so to speak.

If you have a U.S. Patent but want to nail someone for infringement, who is wholly, owned & operated in EU - you couldnt do anything about it, without taking extra steps.

Has nothing to do with us being special. But since you asked - Im special. Just like everyone else.

[BajiRav]

I don't think I know any FS driver (other than NTFS and FAT) that works properly (and seamlessly) on Windows :ermm:

Apple's HFS+ from bootcamp works pretty nicely.

[greenwizard88]

It's a stretch to suggest that it took three years to create / get round to filing it. I don't know the exact rules, but I was under the impression the idea has to be presented or used publicly, not just internally.

Even if it was "public", you would then have to proove that back in the late 80's, that this mailing list was both public and well known in the OS/File system world.

Rather irrelevant since patent rules now say first to implement, and MS not already has the patent, they WHERE first to implement.

Either way, wouldn't a statue of limitations be run, so you can't challenge the patent anymore?

[rfirth]

Either way, wouldn't a statue of limitations be run, so you can't challenge the patent anymore?

The entire patent should expire pretty soon.

[DeadEndAccount]

cool...so now does android manufacturers have to drop paying MS license fee?

open source can never fail.

For larger storage cards they'll need to use exFAT but from what I understand the cost is pretty modest as far as patents go - you can either pay per unit up to a certain amount then it switches over to a flat rate per year.