Microsoft: Windows 7 tool used GPL code
Microsoft said Friday that its inquiry confirms that a tool aimed to make it easier to load Windows 7 on a Netbook does in fact contain open-source code.
"After looking at the code in question, we are now able to confirm this was indeed the case, although it was not intentional on our part," Microsoft's Peter Galli said in a blog posting. "While we had contracted with a third party to create the tool, we share responsibility as we did not catch it as part of our code review process. We have furthermore conducted a review of other code provided through the Microsoft Store and this was the only incident of this sort we could find.
As a result, Microsoft said it will make available next week the source code for the tool as well as the binaries under the terms of the General Public License (GPL v2). Microsoft will also make the tool again available to customers at the Microsoft store.
Microsoft had pulled the software utility down earlier this week after blogger Rafael Rivera noted in a posting that the tool appeared to use code from the open source ImageMaster project. (Of note, that project is now no longer available on CodePlex, where it had been posted).
Though somewhat arcane, the Windows USB/DVD Tool was Microsoft's answer to a tough problem--upgrading the operating system on Netbooks and other PCs without an optical drive.
Microsoft had been exploring for months different ways to handle the issue, eventually settling on this software program, released last month, which lets users take a downloadable copy of the operating system and create a bootable drive.
Releasing software under an open-source license is not entirely new to Microsoft, although Microsoft typically doesn't do so under the GPL, which it sees as one of the more restrictive of the open-source licenses.
The software maker did release a few Linux drivers under GPLv2, although it may have had its hand forced there as well. Some have suggested the drivers contained GPL code, meaning that they necessarily would have had to be released back under the same GPL license.
Microsoft confirmed on Friday that a tool aimed at making it easier to get Windows 7 on to Netbooks does, in fact, use open source code. As a result, Microsoft said it will make the code for the tool publicly available next week.
(Credit: Microsoft)
They have a known history of doing this.
So, why shouldn't they face criticism for something that appears to be rather systemic? What makes them different from any other company? I mean, sure - good on them for not being stupid, but seriously? Unlike you gents, I have no worship for any corporation. They weren't being contrite - they were saving their skins from a lawsuit that would've damaged their public image. If I were Microsoft I would've done the same thing.
You know? For folks that are eager to let bygones be bygones over this issue, I wonder: What were you two doing/saying back when SCO was rattling its saber, or when Steve Ballmer came out and made the specious claim of "235 patents" and such?
I could guess, but I want to hear it from you before having a little fun and digging into the archives of this fine website's talkbacks...
"I could guess, but I want to hear it from you before having a little fun and digging into the archives of this fine website's talkbacks... "
Oh, I *REALLY* don't think you want to do that, Penguinisto. Anyone can just search on your own prior username here and see exactly what your past history of honesty and integrity are. ;)
At least if you do that, be smart and change your user name to hide from the past. Ah, that's right, you already did that once. I see...
Your history is supect at best. Your opinions based on that history REALLY mean nothing.
If all you folks have are emotional arguments as to why you should cheer Microsoft for essentially pulling an improvised CYA, at least have the courage to say so, but it does expose you as the blind worshippers you truly are. :)
==
"Taking responsibility for a mistake isn't that hard though too few companies do it."
I sincerely doubt that Microsoft would have either, if they could get away with it (again, see their history concerning this). Fact is, they were caught dead-to-rights, so they really didn't have much choice. Let me explain:
Objectively, they did the "right" thing because it did the least damage. If they denied it, word would've spread far further than it did given the stark evidence. If they said nothing and quietly pulled it, they would've had yet another instance of "theft" (not my term, mind) tacked onto their reputation, and it would've reinforced the reputation they have for "stealing" others' intellectual property when it suits them.
In this case, they pulled it and said something in an attempt to spin the perception ('oops, we goofed, we made a mistake, we're sorry, we don't normally do or allow any such thing, etc').
You'll notice that the usual band of Microsoft fanboys and trolls have had to resort to emotional arguments instead of rational ones in this thread... they already know all of this. They can't assail the facts, nor even spin them. It's all they have is to ascribe motives and to ride the ad hominem train for all it's worth... and hope that this quietly goes away, since it disrupts their views of how Microsoft should be viewed by the public.
Now contrast this with the whole SCO mess - Linux 'fanboys' everywhere were screaming to see the evidence, demanding to see what code SCO alleged was "stolen". SCO tried to obfuscate things, and ultimately revealed only one scrap of poorly-encrypted code that turned out to be BSD-licensed. IBM, Novell, and even Chrysler (yes, the car company) destroyed SCO's argument, then watched as SCO imploded. Turns out it was a stock-kiting scam by McBride and the other SCO upper echelon (SCOX started at barely $1/share on the NASDAQ, but rocketed to $25/share at its peak, in mere months. McBride made millions off of it, as did other SCO and Tarantella execs).
If anyone has any evidence of Apple, RedHat, Novell, Solaris, or any similar-sized company swiping code wrongly, I'd love to know about it. AFAIK, only Microsoft has a history of doing it, usually on the down-low. The Microsoft fanboys know this, so this is the best they can come up with: emotional ad-hominem.
Yep, being a hateful person total driven to for the destruction of Microsoft constantly spinning things to the negative dreaming someday of Linux's totral domination, yep you've been pretty consistant. Stupid is a stupid does...
Dan you know Peng will never get it. He needs medical help or something. Reality is just so illusive to him.
The bottom line is this: MS SCREWED UP.
They deserve no praise whatsoever for that, or for being forced to admit it by a third party blogger.
Why, thank you for proving me right, mr. kojacked! No sooner than I say that all you have left is emotion and ad hominem, and you come along at just the right time to post a shining example of why it is that you lot have nothing to stand on here.
And yet, here you are. Keep proving me right - this is actually quite enjoyable. :)
At least the 'fessed up to it. It's good to see the GPL working as it should - keeps folks honest (BSD's licensing style invites too much loopholing...)
The Linux kernel code is out there, for anybody to see. No secrets whatsoever. The beauty of the GPL...
Good luck getting MS or Apple to release their kernel source... I'll see pigs fly before that happens.
@JoeF2: The OSX kernel is open-source as well, see also Darwin. You're right in your first premise though - the open and publicly-viewable nature of Linux and OSX is exactly why you don't see news stories about Apple or Linux pulling a stunt like the one in that article up there. :)
While Darwin is Open Source, it is not licensed under the GPL.
Unlike Linux vendors who modify the plain vanilla Linux kernel, Apple is under no obligation to provide the OSX kernel sources.
And indeed, it seems that they do not provide core components of the kernel. Otherwise, this whole thing about the latest OSX kernel not running on Atom processors would be a non-issue. Just remove the offending code, recompile and be done with it...
So, until Apple releases the full OSX kernel source code, they are not any better than MS, at least in that respect.
As you yourself said: BSD's licensing style invites too much loopholing...
Apple doesn't open Aqua (the UI) or other modules and subsystems that run atop the kernel, but the kernel itself is indeed open source, is bootable, and does run as advertised. Darwin itself can run atop the Atom processor, so the blockage was put in place elsewhere. You can download, modify, etc the Darwin kernel straight off of Apple's website.
It would be the equivalent of XOrg and/or X11 being closed, IMHO.
The fact that Darwin isn't released under GPL doesn't mean it isn't open source. The loopholing in BSD's license relates more to companies swiping the code, locking it under proprietary license, and only having to give credit (cf. Microsoft burying author credit in a .dll file for the TCP/IP stack they swiped and put into earlier versions of Windows. Perfectly legal, but seriously unethical).
Well, people have reported that they were able to run the OSX upgrade with the previous kernel on Atom. So, clearly, it is a kernel issue, which implies that the full and 100% complete kernel source is not available from Apple. Darwin apparently is not the same as the OXS kernel.
I suppose I can see the other inference, but I sincerely doubt that was intended. While I know of more than a couple of journalists at cnet (especially a couple of the bloggers) who would purposely make a misleading headline, Ina Freid isn't one of them.
It was never a secret how to boot from a USB stick. The software is GPL. It was out there for everybody to see.
Microsoft just was unable to write similar code on their own. Would have taken longer, and they outsourced it to the cheapest bidder, who apparently also was only able to do it by relying on other people's intellectual property.
The bottom line: audit your code before you release it so that you can make sure you don't violate other peoples' copyright.
MS outsources a lot of non-essential work and there is a thriving micro-economy surrounding Microsoft. (Pun not intended.)
I'm guessing the vendor responsible for using the GPL code will be red-flagged or even lose their vendor status as the SOW/MSA contracts clearly outline IP rules that are meant to prevent this sort of thing. (That, and one or more of the developers at that vendor company may be out of a job??)
Of course, it was an honest mistake. It always is.
It owuld be interesting to see a legitimate review of all 7's subsystems...
The 'San Francisco Canyon Company' affair (which Microsoft had to pay dearly for just as Steve Jobs returned to Apple, if memory serves).
The Windows 2000 networking stack (though that was BSD licensed, and they did give credit... buried deep in a .dll file).
Message Passing Interface in Windows Server 2003 Compute Cluster Edition : http://www.eweek.com/c/a/Windows/Open-Source-Code-Finds-Way-into-Microsoft-Product/
It goes on and on, really... all honest mistakes of course. Anyone could make them. Really.
(...anyone want to buy some oceanfront property in Utah?)
Sad footnote about Stac v. MS DoubleSpace (later DriveSpace): The early version of DoubleSpace had some nasty data loss bugs, a problem that Stacker never suffered from.
See, it can happen to anyone. Now other companies did work with Apple and licensed the technology. Apple? No, Nokia has had to take them to court for this.
Meh, no big deal except to trolls who hate Microsoft. Let's see if we can spot them by their comments.
It is often drop-easy to stumble across someone else's patent - there are literally millions of them out there, and one can credibly argue against any given patent (see also 'prior art', 'overly broad' patents, 'submarine' patents, etc).
Violating copyright however (as Microsoft did) is mostly intentional, because you need someone else's code to copy from.
The most you might need to do is prep/format the drive in advance, which is done with 7 or 8 commands in diskpart.exe. I can't remember the last time I installed Vista/7 from a DVD, even on machines with one. It's simply much much faster from a 30+ MB/s high-quality thumb drive.
Let's not name names though. Their comments will make it apparent enough without anyone pointing fingers to them.
@Dan: I doubt it has anything to do with hate. If you ever do programming professionally, you'll understand that violating copyright on code is not a good thing, no matter who does it.
They got caught, and after verifying that they indeed were using pirated software, they agreed to abide by the license that the software they pirated comes with. With that, it is no longer considered pirated.
Bad Micro$oft - bad monopolist!
Quote: "...we share responsibility as we did not catch it as part of our code review process..."
My @ss! How naive to they think we are?
They tried and failed! As simple as that! Well Micro$oft can say now, for 5 minutes they had their hands on decent software.
:-)
- by giant_david November 16, 2009 5:41 AM PST
- Microsoft is using third party to build components of Windows. We can't be sure that Microsoft is reviewing the code accordingly. No one else can review, since it is close source. Can someone tell us how reliable this scenario is?
- Like this Reply to this comment
-
(49 Comments)