A bit OT, but this issue about FAT patented by MS is not a real one to be worried about, some experi

  • Thread starter Thread starter Jan Panteltje
  • Start date Start date
J

Jan Panteltje

Let me evaluate this a bit deeper:
The issue is all the small cameras.
Video and photo cameras.
When I look at the back of my ImationFlashGo it clearly states:
Made In China.

Now if the Chinese did not want to pay license cost to MS,
then they could standardize ext2 as file system for their
cameras.
This would make sense because already China has adopted Linux
as an OS.

But then, to sell in China, as the US and other want, they would have
to provide at least an ext2 option on their products.

But for MS this would mean INCLUDING ext2 in MS windows (so you
can use explorer).
But if MS includes open source soft, then they must release source.

So this will finally do MS in :-)
 
Jan said:
Let me evaluate this a bit deeper:
The issue is all the small cameras.
Video and photo cameras.
When I look at the back of my ImationFlashGo it clearly states:
Made In China.

Now if the Chinese did not want to pay license cost to MS,
then they could standardize ext2 as file system for their
cameras.
This would make sense because already China has adopted Linux
as an OS.

But then, to sell in China, as the US and other want, they would have
to provide at least an ext2 option on their products.

But for MS this would mean INCLUDING ext2 in MS windows (so you
can use explorer).
But if MS includes open source soft, then they must release source.

So this will finally do MS in :-)

You do not seem to understand software development or the GPL license very well....

First, let me make it clear that I am a firm supporter for non-Microsoft
platforms. And I provide several free open source code projects of my own.

Now, about Microsoft producing an ext2 driver... Who says they would need to use
the Linux source code to do that? There are full descriptions published that
would allow an independent version to be created. It is no different than the
MANY features that have been added to Linux through the years without copying
source code...

Also, and this is just conjecture, but if there is a version of ext2 that has
the LGPL license then Microsoft could legally create a Windows DLL and only
provide the source for the DLL. For a real example, Sony Entertainment is using
my HawkVoiceDI project to provide speech compression in PlanetSide. HawkVoiceDI
is LGPL, and they distribute the unmodified DLL, so they simply provide a link
in their documentation to my site so that users can download the source code.
 
You do not seem to understand software development or the GPL license very well....

First, let me make it clear that I am a firm supporter for non-Microsoft
platforms. And I provide several free open source code projects of my own.

Now, about Microsoft producing an ext2 driver... Who says they would need to use
the Linux source code to do that? There are full descriptions published that
would allow an independent version to be created. It is no different than the
MANY features that have been added to Linux through the years without copying
source code...

Also, and this is just conjecture, but if there is a version of ext2 that has
the LGPL license then Microsoft could legally create a Windows DLL and only
provide the source for the DLL. For a real example, Sony Entertainment is using
my HawkVoiceDI project to provide speech compression in PlanetSide. HawkVoiceDI
is LGPL, and they distribute the unmodified DLL, so they simply provide a link
in their documentation to my site so that users can download the source code.
OK, you are right, but STILL the ability of MS windows OS to then read ext2 may
somewhat cause frowns in Redmond
On the pure technical side, I think ext2 is perhaps better suited for these cards.
Cards get larger all the time.
I tried, could not get reiserfs on a 32 MB SmartMedia, but ext2 did fit.
I am very happy to hear Sony is using your LGPL soft.
Regards
Jan
 
Jan said:
On a sunny day (Wed, 10 Mar 2004 19:00:43 GMT) it happened "Phil Frisbie, Jr."


You do not seems to provide a Linux version, only a windows delphi and mac
version in
hawk/hawknl1.68/samples/
?

All those samples compile fine on Linux. I have just need too busy/lazy to
create a makefile to automatically compile them ;)

How about this... After you run 'make', change to the /samples/ directory and to
compile the first sample use this command line:

gcc -O2 async.c -o async /src/LibNL.a -lpthread
 
All those samples compile fine on Linux. I have just need too busy/lazy to
create a makefile to automatically compile them ;)

How about this... After you run 'make', change to the /samples/ directory and to
compile the first sample use this command line:

gcc -O2 async.c -o async /src/LibNL.a -lpthread

That did not do it, but this did:
gcc -O2 async.c -o async ../src/libNL.so.1.6.8 -lpthread

OK i did them all, ran some tests on my local http server (not online now).

../getfile localhost /panteltje/gallaxy.gif q1.gif

galaxy.gif is
-rw-r--r-- 1 root root 43482 Aug 24 2001 gallaxy.gif
After the above getfile i get:
Server: Apache/1.3.19 (Unix) (SuSE/Linux) mod_throttle/3.0 mod_layout/1.0 mod_fastcgi/2.2.2 mod_dtcl
Last-Modified: Fri, 24 Aug 2001 19:14:13 GMT
ETag: "bb7da-a9da-3b86a785"
Accept-Ranges: bytes
Content-Length: 43482
Connection: close
Content-Type: image/gif
-rw------- 1 root root 43483 Mar 11 02:20 q1.gif

So one extra FREE byte, but the file is corrupted and does not display.
You can see the same online at
http://www.home.zonnet.nl/panteltje/gallaxy.gif

The site displays OK, so does my local server.
Just testing :-)
 
Let me evaluate this a bit deeper:
The issue is all the small cameras.
Video and photo cameras.
When I look at the back of my ImationFlashGo it clearly states:
Made In China.
Why is this an issue at all ? If the manufactor of that camera
did not buy a license its their problem, not the problem of
the users/customers.(Only SCO belives its that way..)
 
Jan said:
That did not do it, but this did:
gcc -O2 async.c -o async ../src/libNL.so.1.6.8 -lpthread

OK i did them all, ran some tests on my local http server (not online now).

./getfile localhost /panteltje/gallaxy.gif q1.gif

galaxy.gif is
-rw-r--r-- 1 root root 43482 Aug 24 2001 gallaxy.gif
After the above getfile i get:
Server: Apache/1.3.19 (Unix) (SuSE/Linux) mod_throttle/3.0 mod_layout/1.0 mod_fastcgi/2.2.2 mod_dtcl
Last-Modified: Fri, 24 Aug 2001 19:14:13 GMT
ETag: "bb7da-a9da-3b86a785"
Accept-Ranges: bytes
Content-Length: 43482
Connection: close
Content-Type: image/gif
-rw------- 1 root root 43483 Mar 11 02:20 q1.gif

So one extra FREE byte, but the file is corrupted and does not display.
You can see the same online at
http://www.home.zonnet.nl/panteltje/gallaxy.gif

The site displays OK, so does my local server.
Just testing :-)

Thank you for bringing this bug to my attention! I know that getfile has worked
on binary files before, and after looking at the source I see where the problem
is. I have fixed the current source, and here is the patch if you want to
further test for yourself:

Modify line 177 to this:
write(f, &buffer[i+1], count - i - 1);
 
Why is this an issue at all ? If the manufactor of that camera
did not buy a license its their problem, not the problem of
the users/customers.(Only SCO belives its that way..)
We probably agree that SCO is not in touch with reality,
or at least try to make their own.
But I think the idea that MS has, is to wait until every camera and consumer system
uses FAT with long filenames, and THEN charge everyone (or sue them for patent
violation).
China could prevent this by going for ext2, and then the ball would be theirs.
A differrent game, without MS.
Given that MS finances SCO, and SCO is playing this thing, the sooner everyone
changes to ext2 the better they are protected against the vultures.
JP
 
We probably agree that SCO is not in touch with reality,
or at least try to make their own.
But I think the idea that MS has, is to wait until every camera and consumer system
uses FAT with long filenames, and THEN charge everyone (or sue them for patent
violation).
China could prevent this by going for ext2, and then the ball would be theirs.
A differrent game, without MS.
Given that MS finances SCO, and SCO is playing this thing, the sooner everyone
changes to ext2 the better they are protected against the vultures.

But don't ya know, SCO owns ext2! Since ext2 is a derivative work of
the Linux kernel, and the Linux kernel is a derivative work of AIX and
AIX is a derivative work of Unix and SCO kinda-sorta-maybe owns Unix
copyright. Therefore all your file systems are belong to SCO!

At least in that odd version of reality known as SCO-world. :>
 
The world rejoiced as Jan Panteltje said:
Given that MS finances SCO, and SCO is playing this thing, the sooner everyone
changes to ext2 the better they are protected against the vultures.

I think I'd rather it be the Minix filesystem, or some other such
filesystem that is _simpler_ than ext2.
--
If this was helpful, <http://svcs.affero.net/rm.php?r=cbbrowne> rate me
http://www.ntlug.org/~cbbrowne/x.html
"Es is nicht gasagt das es besser wird wenn es anders wirt. Wenn es
aber besser werden soll muss es anders werden. (Loosely translated:
Different is not necessarily better. But better _is_ necessarily
different.)" -- G. Ch. Lichtenberg
 
Jan said:
That did not do it, but this did:
gcc -O2 async.c -o async ../src/libNL.so.1.6.8 -lpthread

OK i did them all, ran some tests on my local http server (not online now).

./getfile localhost /panteltje/gallaxy.gif q1.gif

galaxy.gif is
-rw-r--r-- 1 root root 43482 Aug 24 2001 gallaxy.gif
After the above getfile i get:
Server: Apache/1.3.19 (Unix) (SuSE/Linux) mod_throttle/3.0 mod_layout/1.0 mod_fastcgi/2.2.2 mod_dtcl
Last-Modified: Fri, 24 Aug 2001 19:14:13 GMT
ETag: "bb7da-a9da-3b86a785"
Accept-Ranges: bytes
Content-Length: 43482
Connection: close
Content-Type: image/gif
-rw------- 1 root root 43483 Mar 11 02:20 q1.gif

So one extra FREE byte, but the file is corrupted and does not display.
You can see the same online at
http://www.home.zonnet.nl/panteltje/gallaxy.gif

The site displays OK, so does my local server.
Just testing :-)

Thank you for bringing this bug to my attention! I know that getfile has worked
on binary files before, and after looking at the source I see where the problem
is. I have fixed the current source, and here is the patch if you want to
further test for yourself:

Modify line 177 to this:
write(f, &buffer[i+1], count - i - 1);
That causes even more problems:

-rw------- 1 root root 47255 Mar 11 23:41 q1.gif
My suggestion to you is to try it out first one some downloads.
The gif is not only far too long now, but also still corrupted.
Hope you find the bug,
Jan
 
Christopher said:
I think I'd rather it be the Minix filesystem, or some other such
filesystem that is _simpler_ than ext2.

I agree with that. I don't know any filesystem that
is simpler than minix. It is impressive that the
source for the minix implementation in Linux is
only half the size of the fat implementation. And
yet minix has features that fat lacks.
 
On a sunny day (Thu, 11 Mar 2004 17:40:11 GMT) it happened "Phil Frisbie, Jr."

Thank you for bringing this bug to my attention! I know that getfile has worked
on binary files before, and after looking at the source I see where the problem
is. I have fixed the current source, and here is the patch if you want to
further test for yourself:

Modify line 177 to this:
write(f, &buffer[i+1], count - i - 1);
That causes even more problems:
Eh, that was my mistake, I added that line and forgot to comment out the old one.
Seems to work OK now :-)
JP
 
Back
Top