PDA

View Full Version : Microsoft swipes some GPL code without permission



phrostbyte
November 8th, 2009, 03:51 AM
Let's be realistic here, it was probably an honest mistake. 8-[

http://www.withinwindows.com/2009/11/06/microsoft-lifts-gpl-code-uses-in-microsoft-store-tool/



While poking through the UDF-related internals of the Windows 7 USB/DVD Download Tool, I had a weird feeling there was just wayyyyyyyyy too much code in there for such a simple tool. A simple search of some method names and properties, gleaned from Reflector’s output, revealed the source code was obviously lifted from the CodePlex-hosted (yikes) GPLv2-licensed ImageMaster project. (The author of the code was not contacted by Microsoft.)

pwnst*r
November 8th, 2009, 06:33 AM
been posted. also, blown out of proportion. typical.

phrostbyte
November 8th, 2009, 06:34 AM
If it was already posted why did you bump this when is was off the main page? [-(

Artificial Intelligence
November 8th, 2009, 06:34 AM
!!!! I can't edit the title, it is suppose to be Microsoft swipes some GPL code without permission

Done.

phrostbyte
November 8th, 2009, 06:35 AM
Done.

Thank you! :p

pwnst*r
November 8th, 2009, 06:38 AM
cuz i felt like it?

actually, didn't realize i was on the 2nd or 3rd. i was looking for something.

phrostbyte
November 8th, 2009, 06:40 AM
cuz i felt like it?

actually, didn't realize i was on the 2nd or 3rd. i was looking for something.

Good enough reason I guess.

pwnst*r
November 8th, 2009, 06:44 AM
there's no guessing. it's a good reason.

pwnst*r
November 8th, 2009, 06:45 AM
oh, and

http://ubuntuforums.org/showthread.php?t=1318255&highlight=slashdot

phrostbyte
November 8th, 2009, 06:50 AM
oh, and

http://ubuntuforums.org/showthread.php?t=1318255&highlight=slashdot

Whatever the hell that was talking about is wrong. If you go to that blog post, it's become clear that isn't just one method (ReadBytes)

The only argument people have is that perhaps the code comes from another mysterious project which Microsoft can legally use code from. But I guess we can not know for sure until the author of ImageMaster speaks up. I hope he is fairly compensated if this turns out to be true. :)

Icehuck
November 8th, 2009, 07:02 AM
Whatever the hell that was talking about is wrong. If you go to that blog post, it's become clear that isn't just one method (ReadBytes)

The only argument people have is that perhaps the code comes from another mysterious project which Microsoft can legally use code from. But I guess we can not know for sure until the author of ImageMaster speaks up. I hope he is fairly compensated if this turns out to be true. :)

Show me the code line for line and I'll believe you. The author of the blog never actually proves it, but say's he leaves it up to the readers to do so.

phrostbyte
November 8th, 2009, 07:04 AM
Show me the code line for line and I'll believe you. The author of the blog never actually proves it, but say's he leaves it up to the readers to do so.

http://www.withinwindows.com/wp-content/uploads/2009/11/example1.png

Notice how that is not ReadBytes

More code:

http://www.withinwindows.com/files/gpl/comparison_1.txt

How much code is needed to prove a violation? I am not a lawyer. :)

Frak
November 8th, 2009, 07:14 AM
http://www.withinwindows.com/wp-content/uploads/2009/11/example1.png

Notice how that is not ReadBytes

More code:

http://www.withinwindows.com/files/gpl/comparison_1.txt

How much code is needed to prove a violation? I am not a lawyer. :)
That's not nearly enough. One or two methods won't be considered enough proof. The rule is if the program emulates the entire logic and goal of another program, then it is theft. Something like this, which is just a simple method, would be considered nothing more than coincidence or canonical methodology.

phrostbyte
November 8th, 2009, 07:15 AM
That's not nearly enough. One or two methods won't be considered enough proof. The rule is if the program emulates the entire logic and goal of another program, then it is theft. Something like this, which is just a simple method, would be considered nothing more than coincidence or canonical methodology.

I'll have to disagree for now.

Frak
November 8th, 2009, 07:23 AM
I'll have to disagree for now.
Code theft is incredibly hard to prove in court. CherryPC was the last project that could, without a doubt, be proven in court.

Chronon
November 8th, 2009, 07:28 AM
That's not nearly enough. One or two methods won't be considered enough proof. The rule is if the program emulates the entire logic and goal of another program, then it is theft. Something like this, which is just a simple method, would be considered nothing more than coincidence or canonical methodology.

I don't see how that can possibly be the case. Emulating what another program does is totally legal and legit (barring patents or the like). Copyright only protects the form, not the function of a program's code. I do agree that existence of a few common strings is probably not enough to demonstrate a copyright violation.

Frak
November 8th, 2009, 07:30 AM
I don't see how that can possibly be the case. Emulating what another program does is totally legal and legit (barring patents or the like). Copyright only protects the form, not the function of a program's code. I do agree that existence of a few common strings is probably not enough to demonstrate a copyright violation.
By emulate, I mean entire theft of code.

phrostbyte
November 8th, 2009, 07:30 AM
Code theft is incredibly hard to prove in court. CherryPC was the last project that could, without a doubt, be proven in court.

If there is reasonable assumption that they may have swiped this gentleman's code, I think Microsoft would be generous enough to take steps to obtain a license and compensate the author for his efforts and in proportion to their mistake. Despite Microsoft's disrespect of the GPL, I do believe they respect copyright law and the right of a software developer to control unauthorized use of their work, which this may be just one example of.

Frak
November 8th, 2009, 07:36 AM
If there is reasonable assumption that they may have swiped this gentleman's code, I think Microsoft would be generous enough to take steps to obtain a license and compensate the author for his efforts and in proportion to their mistake. Despite Microsoft's disrespect of the GPL, I do believe they respect copyright law and the right of a software developer to control unauthorized use of their work, which this may be just one example of.
I'm going to go with the 3rd party studio on this one. I doubt Microsoft would swipe code. There's massive code audits on internal code every so often. I'll go with the same thing that happened with the Windows XP sound files, Microsoft handed the duty to a 3rd party studio and they decided to take the easy way out and swipe code.

I still say this is blown out of proportion.

lethalfang
November 8th, 2009, 09:42 AM
If there is reasonable assumption that they may have swiped this gentleman's code, I think Microsoft would be generous enough to take steps to obtain a license and compensate the author for his efforts and in proportion to their mistake. Despite Microsoft's disrespect of the GPL, I do believe they respect copyright law and the right of a software developer to control unauthorized use of their work, which this may be just one example of.

Except that, once you have released your code under GPL, you cannot un-GPL it.

Frak
November 8th, 2009, 09:47 AM
Except that, once you have released your code under GPL, you cannot un-GPL it.
Current releases, no, future releases, yes.

phrostbyte
November 14th, 2009, 03:05 AM
New news on this. :)

Microsoft has admitted that they took GPL code without permission and has in turn released their entire codebase for this tool under the GPL:

http://community.winsupersite.com/blogs/paul/archive/2009/11/13/rafael-is-vindicated-microsoft-did-steal-open-source-code-for-usb-dvd-tool.aspx

Sealbhach
November 14th, 2009, 03:06 AM
New news on this. :)

Microsoft has admitted that they took GPL code without permission and has in turn released their entire codebase for this tool under the GPL:

http://community.winsupersite.com/blogs/paul/archive/2009/11/13/rafael-is-vindicated-microsoft-did-steal-open-source-code-for-usb-dvd-tool.aspx

Looks like Microsoft is getting more and more contaminated by freedom.

.

phrostbyte
November 14th, 2009, 03:07 AM
Looks like Microsoft is getting more and more contaminated by freedom.

.

Indeed. :lolflag:

Frak
November 14th, 2009, 05:03 AM
Looks like Microsoft is getting more and more contaminated by freedom.

.
Looks like that 3rd party studio is about to feel the iron palm of Microsoft's lawyers.

earthpigg
November 14th, 2009, 05:27 AM
Looks like Microsoft is getting more and more contaminated by freedom.

ammunition for future claims that the GPL is a 'cancer' that 'infects' software.

wilee-nilee
November 14th, 2009, 05:39 AM
You can't get the link to the dvd-thumb anymore but just yesterday I was able to download into another computer with the link I saved. I haven't tried to see if it still works even though I have a W7 DVD my thumb is loaded already with the original use I am not going to wipe it to see if the program still works.

phrostbyte
December 7th, 2009, 06:24 AM
Another update on this (a bit late):

http://port25.technet.com/archive/2009/11/20/update-on-the-windows-7-usb-dvd-tool.aspx

Microsoft seems to be delaying the release of this source code. They give no indication when it will actually be released.

Frak
December 7th, 2009, 06:27 AM
Another update on this (a bit late):

http://port25.technet.com/archive/2009/11/20/update-on-the-windows-7-usb-dvd-tool.aspx

Microsoft seems to be delaying the release of this source code. They give no indication when it will actually be released.
It's a conspiracy!

phrostbyte
December 7th, 2009, 06:27 AM
It's a conspiracy!

Probably.

alexfish
December 7th, 2009, 06:41 AM
History / part1

The next time Bill Gates sends an e-mail through Microsoft’s shiny new Wireless LAN it will be passed through a behind-the-scenes Linux-based network appliance.
Earlier this year Microsoft and Aruba Networks jointly announced the two companies will work to replace Microsoft’s existing Cisco wireless network with Aruba’s centrally-managed infrastructure, which eliminates the need for individual changes on the access points.
Aruba Networks was selected to provide the networking equipment for what is considered to be one of the world’s largest next-generation wireless LANs, serving more than 25,000 simultaneous users a day in some 60 countries. According to an Aruba press statement, Microsoft’s new WLAN will be deployed in 277 buildings covering more than 17 million square feet using Aruba mobility controllers, mobility software and some 5000 wireless access points.


So Linux Works / Wonder if they made a donation / and have copies of the GNU General Public License

history/ part 2

Seems they don't have them

so posted this on their new site Forum
http://www.linux.org/info/gnu.html

History /part 3

the story behind part 2

Linux /mobile broadband / site restricted / used friends Phenom X4 9850 system / xp

site not restricted / posted // then had coffee / Smashed the dongle

phrostbyte
December 10th, 2009, 10:35 PM
They finally released the code.

http://wudt.codeplex.com/

Unfortunately it contains unmanaged (C++) libraries which are written to be heavily dependent on Win32. I didn't look at it so hard, but it's probably difficult to port to Linux.

It might however be useful as a ISO -> USB Drive creator for Windows. I have to commend Microsoft, it has a very nice UI. This could even replace the Windows-based USB tool that is bundled with Ubuntu. :)

Frak
December 10th, 2009, 10:52 PM
Unfortunately it contains unmanaged (C++) libraries which are written to be heavily dependent on Win32. I didn't look at it so hard, but it's probably difficult to port to Linux.

CoolStoryBro

phrostbyte
December 10th, 2009, 10:55 PM
CoolStoryBro

Do you have anything productive to add to this thread?

Frak
December 10th, 2009, 10:55 PM
Do you have anything productive to add to this thread?
No, other than:

"Wow, Microsoft having code that is heavily tied to Windows? That's new!"

murderslastcrow
December 10th, 2009, 10:58 PM
I wonder why this hasn't been marked as recurring discussion. ^_~ LOLZ!

The day to hide behind your dominance is ending fast for Microsoft. I can see it all around me.

phrostbyte
December 10th, 2009, 11:09 PM
I wonder why this hasn't been marked as recurring discussion. ^_~ LOLZ!

The day to hide behind your dominance is ending fast for Microsoft. I can see it all around me.

Actually I'd like this to be moved to Programming Talk, if possible, now that Microsoft released the source code.

phrostbyte
December 10th, 2009, 11:11 PM
The code does a total of six native calls.



MainForm.cs/NativeMethods
-------------------

/// <summary>
/// The send message method for allowing the window to be dragable.
/// </summary>
/// <param name="hWnd">The window handle.</param>
/// <param name="msg">The message to send.</param>
/// <param name="wParam">The wParam value.</param>
/// <param name="lParam">The lParam value.</param>
/// <returns>The hResult of the call.</returns>
[DllImport("user32.dll")]
public static extern IntPtr SendMessage(IntPtr hWnd, int msg, IntPtr wParam, IntPtr lParam);

/// <summary>
/// Releases the window caputre.
/// </summary>
/// <returns>True on success.</returns>
[DllImport("user32.dll")]
[return: MarshalAs(UnmanagedType.Bool)]
public static extern bool ReleaseCapture();

/// <summary>
/// Sets the execution state for allowing the tool to disable standby (Vista and higher).
/// </summary>
/// <param name="esFlags">The flags indicating the thread execution state.</param>
/// <returns>The execution state that was set.</returns>
[DllImport("kernel32.dll")]
public static extern EXECUTION_STATE SetThreadExecutionState(EXECUTION_STATE esFlags);

/// <summary>
/// Gets the integer value for the given window message.
/// </summary>
/// <param name="msgString">The window message to lookup.</param>
/// <returns>The integer value for the message.</returns>
[DllImport("user32", CharSet = CharSet.Auto)]
private static extern int RegisterWindowMessage([In, MarshalAs(UnmanagedType.LPWStr)] string msgString);

USBDriveService/NativeMethods
-----------------------------

/// <summary>
/// Sets the active partition.
/// </summary>
/// <param name="drive">The root path of the drive to update.</param>
/// <returns>Returns 0 for success. See System Error Codes for possible error values.</returns>
[DllImport("IoWrapper.dll", CharSet = CharSet.Auto)]
public static extern int SetActivePartition([In, MarshalAs(UnmanagedType.LPWStr)] string drive);

/// <summary>
/// Formats the drive.
/// </summary>
/// <param name="drive">The root path of the drive to update.</param>
/// <returns>Returns 0 for success. See System Error Codes for possible error values.</returns>
[DllImport("IoWrapper.dll", CharSet = CharSet.Auto)]
public static extern int FormatDrive([In, MarshalAs(UnmanagedType.LPWStr)] string drive);


The obvious question would be:
Does anyone know what native calls or Mono methods to replace these methods, so that it will compile under Linux?

red_Marvin
December 10th, 2009, 11:18 PM
Except that, once you have released your code under GPL, you cannot un-GPL it.
AFAIK the GPL does not stop the original developer of a piece of code to dual license it etc.

Frak
December 10th, 2009, 11:33 PM
AFAIK the GPL does not stop the original developer of a piece of code to dual license it etc.
Exactly. The original developer could have said "It is OK for Microsoft not to release the source" and that would be the end of the discussion. Microsoft could have paid the developer to let them go without releasing the source code, which I don't doubt Microsoft has done in the past. They didn't, which was wise. The project didn't have any real value to Microsoft, but even then, it's cheaper to buy working code than to build it again yourself.

wilee-nilee
December 10th, 2009, 11:41 PM
It has been re-released available at the MS store and here.
http://wudt.codeplex.com/

phrostbyte
December 10th, 2009, 11:50 PM
I managed to get the tool to accept an Ubuntu ISO. :guitar:

Don't get to excited though, I did it in a hackish way (changed something from "false" to "true"). I want to figure out the 'right' way. :)

I've pinpointed the issue so far to this block of code


VolumeTag tag = new VolumeTag();
if (tag.Parse(0, buffer, buffer.Length)
&& tag.Identifier == (short)VolumeDescriptorType.AnchorVolumePtr)
{
result = new UdfFileExtent();
result.Parse(16, buffer);
}


Not sure what VolumeTag.Parse or VolumeTag.Identifier is suppose to mean in this context.