MS patent on FAT a problem for ros?

[Z98]

This is becoming more and more of a non-issue with every passing day. None of the developers are concerned about the patents, and most Linux distros, both community centric and corporate backed, don't seem terribly bothered about it either. What does that suggest for the patent?

Our FAT driver is supposed to be compatible with the regular old FAT32 driver in Windows. If it isn't, there's a bug that needs to be fixed. Any suggestion that we aren't being compatible are hypotheticals raised because people keep hammering at this issue, which no one on the development team feels is of major concern.

[marsilies]

This is becoming more and more of a non-issue with every passing day. None of the developers are concerned about the patents, and most Linux distros, both community centric and corporate backed, don't seem terribly bothered about it either. What does that suggest for the patent?

As far as Linux is concerned, Microsoft has said that it won't go after noncommercial developers of Linux.

http://www.news.com/2100-1016_3-6132119.html
"Moreover, Microsoft said it will not enforce its patents against individual, noncommercial Linux developers."

I don't know what this means for the commercial developers. I don't think they have to public announce what licenses they have bought though, so it's possible that aside from Novell, other commercial developers have properly licensed it from Microsoft. Others may be ignoring the patent at their own peril.

http://en.wikipedia.org/wiki/Fat32#FAT_licensing

I think the issue with LFN in FAT is the same as it is with other non-foss. It's like how the LAME project doesn't release any binaries themselves, just the source, due to patent issues.

http://en.wikipedia.org/wiki/LAME#Paten ... gal_issues

The questions are mainly:

• Is ROS free from the risk of Microsoft enforcing its patents on FAT?
• Should ROS utilize something that is patented and thus proprietary, even if there's no enforcement?
• Is it possible to make an implementation of FAT32 that doesn't include the MS LFN support?

[Z98]

I'm going to point out a very simple fact here. ROS likely violates a lot more patents than just the FAT patent. For that matter, Wine probably does as well. And I know that Samba violated at least one in the past, don't know if they changed that algorithm. It happens to be a patent my OS prof filed while working for MS. After all, these days it's next to impossible to implement something without tripping over a patent. So I really don't know why people are hammering the FAT patent issue. It's a drop in the bucket of the whole mess.

If it does come down to a lawsuit, then MS will in effect also be suing Wine, Samba, and any other project that is attempting to reimplement MS technology. We are not alone in having this problem. Can it be solved? Sure, once the Supreme Court gets around to sorting out the whole software patent mess. But again, that's only in the States. The EU doesn't have software patents and it's not like MS coming after the project will mean too much. Development outside the States will continue regardless of what MS tries to do in the States.

[erkinalp]

It does come down to a lawsuit, then MS will in effect also be suing Wine, Samba, and any other project that is attempting to reimplement MS technology. We are not alone in having this problem. Can it be solved? Sure, once the Supreme Court gets around to sorting out the whole software patent mess. But again, that's only in the States. The EU doesn't have software patents and it's not like MS coming after the project will mean too much. Development outside the States will continue regardless of what MS tries to do in the States.

Therefore, in the EU, any developer can develop a FAT32 or exFAT driver despite what I've mentioned above.

[GreatLord]

simple fact why almost everyone do not care about FAT patent
1. FAT have always been free to use (almost always)
2. MS did take some pantent but did let people use it for free
3. after many year around 200x they start take licen money for it
4. legal fight betwin ms and pubpat over FAT, pubpat did win the battle over FAT it was free again then after 1 year they lost in cort again FAT was licen, then again PUBPAT did win so on.
this battle is not over yet.

The question is if the patent is vaild to ask licen money for or not, and that what pubpat win or lost time to time over ms.

So none care really about this licen.

[Arnor]

HPFS ... doesn't support large Volumes

Wikipedia article says that max volume size is 2 TiB. Max file size is 2 GB but this is not limit for HPFS. Max size in this FS depends on sizes of field storing file size and file descriptor (4 bytes in OS/2). So it is possible to overcome this "limit".

HPFS does not support large disks. I'm probably one of the last one still running OS/2 (well, eCS), and my 63GB partition was too large for HPFS to support

EDIT: whoops, didn't know that part of the discussion was so long ago..

[Blackcrack]

Hy,

why such a wrangling around the "Fat" file system, to shold be programming the drivercache btw.
the enability so fast as possibility to use some other file system drivers, .. it is not true/ isnt it ?

A dispute of the rank the filesystem driver is useless.
Rather, one should look that is possible is made possibility use other file system driver !
To come in consens on ROS, it is a must be a filesystem who support Userrights, like NT (MS-Standart)
therfor we need a FS to support that and ext3/4 do it. ReiserFS too.. and it's faster and if any not like RFS
so take XFS and all be frendly..

Therefor mus be finisht the driverport.. imho.. and then it isnt a problem the Fad Patent..
then we dont mus use the Fat-fs it is wait manny Programmers/ or be at start if it finish
the driver-port so whant they are write/porting FS-drivers for Reactos.org
Why ? They are show what it is behind Reactos... a Open Source and "timeless" consens.
Not so like Microsoft to bring every new standarts, like Win95/98/Me and XP , Vista and updatet like W7
New Systems who must be buy/have with manny munny and later gone so software who
have funtions in versions befor... nahh.. it's true.. isnt it ?!


that it is important that we finally have the driver port...
To make possible to let make/porting diffrents Filesystems for Reactos ...

Alone these links it's give a very,very small look into the future ! :
http://www.acc.umu.se/~bosse/
http://www.ltr-data.se/opencode.html
http://ashedel.chat.ru/source/

best Regards
Blacky

[Haos]

Blackcrack, you obviously have no bloody idea why do we ever need vfat driver. Without details too technical, without it, working properly, there will be no Mm, thus no Cc working correctly, thus no chance to support any other FS, ever. So please spare us, and get yourself a break from posting...

[Blackcrack]

Haos,

please, i know, the api mus be finishing to use other FS, i know it..
Haos, if you every blow hot wind .. so it's hot wind, please,
stop and wait for the supporting some other FS,
i say it simply for you Haos, i know it.. Dude..
and.. leave me Haos, you be the one who nerve..
and stay at topic..

chears
Blacky

[Haos]

Sorry, reading your gibberish posts is quite hard, but from what i gathered, you asked why devs work on FAT filesystem and not others, like ntfs, ext, or reiser...
Its because kernel components, vital to any FS system, Cc that is, needs a lot of work just to work correctly. Cc on the other hand is heavily knit to Memory Manager (Mm) and both, to be developed, require a working, stable base FS. Here is where Fat comes in handy.

[Blackcrack]

api = application programming interface, it was meant it with Memory Manager too.
how ever, you be a nuisance , you go me on..

[Haos]

API is a user, public interface. NTOSKRNL code is mostly internal and private. No, its not Win32 API.

[Heis Spiter]

Wow... Easy!

We need to refactor our FAT driver because the one we use now, has been pretty badly written (to support lacks of features). Then, Cc and Mm have been hacked to support that driver properly. Writing a new driver will help us to get rid of all that hacky stuff. Moreover a good FSD can be used as a test suite for kernel.

[andrewweb]

Microsoft has their patent up held in court for FAT
The problem is noone and I mean noone has ever acknowledged they had one.
Now Microsoft has put up a page saying they want money for the use of FAT.
Almost all usb drives and memory sticks/cards use FAT
Linux supports fat . In fact ost OSs use or supprt FAT
No need to worry, for a low fee of $250,000 you can use it all you want with your product.

From my two business law class that I have taken (yes, that does not make me a lawyer), MS has to go after anyone that abuses their patent or else it is the same as void it.

Yes.. there are millions of mobile manafacturers, portable manafacturers who make products using sd cards and the like with the Fat filesystem... and none of them ask microsoft about IP first... and MS hasnt sued them yet, (I dont believe theyve sued linux yet, but they may have). so there.

[lukves]

Its hard to implement JFS ??? I read about it on eComStation OS (os2) and i use it in Linux.. Is very good filesystem.. i think FAT and NTFS are MS like filesystems. and when we wants make ReacTOS Win more compat. OS windows the nwe must have FAT and NTFS, but in the future ReactOS CAN NOT be depend on MS becouse MS is not very good.. its evil.. and MS patents are problems... I vote for IBM JFS filesystem.. its used in OS2 and i think is good choose for reactos.. ext3 is more linux and unix like