Microsoft wants royalties for use of FAT

[Wierd Wierd]

>   
What is the legal status of HPFS- I know it was
designed by microsoft, but from what I understand, IBM
owns IP rights-- Is the patent on it still valid, and
enforceable? If not, then it might be something to
look at.  You might be able to take HPFS, and modify
it to store DOS short names, for the benefit of the
DOS subsystem later on.  I think IBM would be less
eager to sue our pants off for using an HPFS variant,
than MS would be for using a FAT or NTFS variant.


On Tue, 9 Dec 2003, Vizzini wrote:

>> On Mon, 2003-12-08 at 19:17, Rick Parrish wrote:
>> > Waldo Alvarez wrote:
>> > 
>> >
>   
>> >
>Let's keep with NTFS and FAT and with time
>design a new patent free 
>> > >filesystem. If M$ complaints then we take away
>the drivers and ppl will be able to get them >from
.previous releases. The FSF did that with XMMS >and
patents with MP3s.
>> > >  
>> >
>   
>> > Works
for me. Sort of like limping along on MP3s
>and GIFs until 
>> > patent/license free OGG and PNG support is ready.
>> 
>> In the United States, liability for patent
>infringement exists whether
>> or not you've been notified by the patent holder of
>such infringement. 
>> In other words, if you accidentally step on someone
>else's IP, they will
>> be able to sue you for royalties whether or not it
>was your "fault". 
>> It's a pure liability - no fault is required.

>  In any case the law is already broken. 
> >   Furthermore,
> > courts have a history of stiffer 
> penalites (i.e. higher
> >   royalty payments) for infringers who *knew* they
> were infringing. 
> >   That's why you see "patent pending"
> > on lots of 
> manufactured items. 
>  Bad. That means distributing NTFS and FAT in another
> release is asking for 
> more trouble. 
> >   Finally, as I
> > said before, both developers and end 
> users are liable for
> >   patent infringement.
> >  
> >  With these points in mind, I think we need an 
> official policy of
> >   avoiding patents as much as we possibly can while
> retaining Windows
> >   system compatibility.  If this means (for
> > example) 
> defaulting to a
> >   non-FAT, non-NTFS filesystem, that's what we
> > have 
> to do. 
>  True. You are totally right! 
>  I was thinking this yesterday. What about using an
> NTFS a little bit 
> modified. In a way you can reuse the code written for
> the  drivers already 
> written  and make the conversion easier. It won't be
> NTFS. I don't like 
> too much this but could be a solution. 
>  The other choice I was thinking. What about finding
> some company that 
> already have the license to use the patents and put
> them in charge of the 
> drivers in exchange for some advising in ROS. They
> will not have to spend 
> a cent for it (except for the lawyer maybe) and will
> not have to write a 
> single line of code. I'm sure many will like the
> deal, even if there is 
> trouble in the future because getting in trouble
> could give them more money. 
>  I prefer this choice as there will be not need to
> invent yet another 
> filesystem, ReactOS will be more compatible and
> everything can continue 
> the way it is. 
>  The question is, How legal would it be? 
>  Best Regards
> Waldo Alvarez  
>  >  -Vizzini
> > 
> > 
> > _______________________________________________
> > ros-general mailing list
> > ros-general at reactos.com
> > http://reactos.com/mailman/listinfo/ros-general
> >  

__________________________________
Do you Yahoo!?
New Yahoo! Photos - easier uploading and sharing.
http://photos.yahoo.com/

[Mark IJbema]

On Tue, Dec 09, 2003 at 05:18:16PM -0800, Wierd Wierd wrote:
> >   
>  What is
> the legal status of HPFS- I know it was
>  designed by microsoft, but from what I understand, IBM
>  owns IP rights-- Is the patent on it still valid, and
>  enforceable? If not, then it might be something to
>  look at.  You might be able to take HPFS, and modify
>  it to store DOS short names, for the benefit of the
>  DOS subsystem later on.  I think IBM would be less
>  eager to sue our pants off for using an HPFS variant,
>  than MS would be for using a FAT or NTFS variant. 
That wouldn't help as MS patented the idea of storing long and
short filenames at the same time in a filesystem.

Mark

[Frank D. Engel, Jr.]

Then calculate short names dynamically at runtime, rather than storing
them?

--- Mark IJbema <mark(a)ijbema.xs4all.nl> wrote:
>   On Tue, Dec 09, 2003 at 05:18:16PM -0800, Wierd Wierd
> wrote:
> > >   
> > What is the legal status of HPFS- I know it was
> >  designed by microsoft, but from what I understand, IBM
> >  owns IP rights-- Is the patent on it still valid, and
> >  enforceable? If not, then it might be something to
> >  look at.  You might be able to take HPFS, and modify
> >  it to store DOS short names, for the benefit of the
> >  DOS subsystem later on.  I think IBM would be less
> >  eager to sue our pants off for using an HPFS variant,
> >  than MS would be for using a FAT or NTFS variant. 
>  
>  That wouldn't help as MS patented the idea of storing long and
>  short filenames at the same time in a filesystem.
>  
>  Mark
>  _______________________________________________
>  ros-general mailing list
>  ros-general(a)reactos.com
>  http://reactos.com/mailman/listinfo/ros-general 
=====
=======
Frank D. Engel, Jr.

Modify the equilibrium of the vertically-oriented particle decelerator to result in the
reestablishment of its resistance to counterproductive atmospheric penetration.

__________________________________
Do you Yahoo!?
Protect your identity with Yahoo! Mail AddressGuard
http://antispam.yahoo.com/whatsnewfree

[Mark IJbema]

On Wed, Dec 10, 2003 at 06:40:55AM -0800, Frank D. Engel, Jr. wrote:
[ms patented storing long and short filenames at the same time]
>   Then calculate short names dynamically at runtime,
> rather than storing
>  them? 
But how to do this? It isn't as easy as it looks, since each long
filename should give the same short filename all the time. It could be
done by using some hash, but then unique names aren't guaranteed.
Another alternative would be calculating the name based on the creation
date of the file, but then files get the creation date they're created
on the local fs, which isn't the case normally under windows.

Mark

[Andrew "Silver Blade" Greenwood]

Doesn't FAT32 use that annoying ~1 for long filenames? For example:
"A Text File With A Long Name.txt" == "ATEXTF~1.TXT"

Although you'd need to be careful to avoid collision with other files
with similar names.

However, this would break compatibility with DOS and Windows 3.1 (if
anyone still uses that!) and would only really be useful for DOS and
Win3.1 app compatibility.

Hmmmm...



On Fri, 2003-12-12 at 13:48, Mark IJbema wrote:
>   On Wed, Dec 10, 2003 at 06:40:55AM -0800, Frank D.
> Engel, Jr. wrote:
>  [ms patented storing long and short filenames at the same time]
> >   Then calculate short names dynamically at
> > runtime, rather than storing
> >  them? 
>  
>  But how to do this? It isn't as easy as it looks, since each long
>  filename should give the same short filename all the time. It could be
>  done by using some hash, but then unique names aren't guaranteed.
>  Another alternative would be calculating the name based on the creation
>  date of the file, but then files get the creation date they're created
>  on the local fs, which isn't the case normally under windows.
>  
>  Mark
>  _______________________________________________
>  ros-general mailing list
>  ros-general(a)reactos.com
>  http://reactos.com/mailman/listinfo/ros-general

[KJK::Hyperion]

At 15.36 12/12/2003, you wrote:
>  However, this would break compatibility with DOS and
> Windows 3.1 (if 
> anyone still uses that!) and would only really be useful for DOS and 
> Win3.1 app compatibility. 
it's not just "useful": some recent programs *require* it. I tried 
disabling short filenames, but several programs (installers in particular) 
didn't like it at all