Whats the scam?

Post Reply New Topic

Usually a move like this means they are up to no good.

http://www.microsoft.com/presspass/feat ... nuxqa.mspx

I wonder how hopelessly out-of-date, buggy and bloated the code is?

"The code, which includes three Linux device drivers, has been submitted to the Linux kernel community for inclusion in the Linux tree. The drivers will be available to the Linux community and customers alike, and will enhance the performance of the Linux operating system when virtualized on Windows Server 2008 Hyper-V or Windows Server 2008 R2 Hyper-V."

I've just noticed the arrogance. It goes from "submitted ... for inclusion" immediately to "will enhance..." without any thought that Linus could just reject it all.

Also note that the ONLY stuff they are releasing is code to run Linus on a virtual machine on top of MS OSes.

There also seems to be some confidence on their part that: "Without this driver code, Linux can run on top of Windows, but without the same high performance levels."

I wonder how they can make this assumption - that the thousands of Linux developers are incapable of improving on the code Miscosoft's Hyper-V team have managed to cobble together.

Well, I note that they are releasing under GPL v2 rather than 3, which makes me wonder if they caught a loop hole in the licence. A little leverage?

Even if the code is out of date, its open enough for GPL v2, so it can be improved. Likely what they are doing is fishing for improvements on something which will be the basis for a closed product. Since they are the originators, they can assign a dual license to their own work.

Just because a programmer is employed at Microsoft, does not necessarily mean (s)he is incompetent. And of course the code they submitted is intended to serve a purpose that will benefit Microsoft. They are a business, not a charity. I'm sure the kernel developers will look at the code, make any improvements they feel are necessary, and decide whether or not it is appropriate to add this code to Linux.

Scam? Aren't you guys overreacting? It's just drivers for their virtualized hardware. Last time I checked, almost all distros include the equivalent drivers for virtualbox. So what's wrong with M$ releasing their drivers with a license most distro should be comfortable with?

It doesn't seem like you get what these drivers are for. Without these guest drivers, an emulator has to emulate every details of the hardware. All those I/O are costly to emulate. These drivers can usually do a DMA transfer with a round trip of context switches. But each emulator has their own way of invoking those functions, that's why they need their own drivers.

There's actually an interesting trick that Hyper-V does to reduce emulation overhead. It needs guest OS support so I don't think it works with linux but it's interesting none the less. Normally, the guest drivers still operate like it's communicating with some hardware. The guest OS still perform all the low level functions until it reaches the driver. With Hyper-V, the drivers can operate at a higher level, bypassing some of these overhead. On Windows, this is done by making some of them as filter drivers. It's kind of like partial-paravirtualization.

Microsoft did this because they legally had to; a network driver in Microsoft’s Hyper-V used open-source components licensed under the GPL and statically linked to binary parts. The GPL does not permit the mixing of closed and open-source elements.

They only contributed 14K lines of code, far from 20K.

The drivers were going to be dropped out anyway in Linux 2.6.33 due to lack of support. Microsoft developers stepped up to maintain it again, so they're still in 2.6.33 and 2.6.34, but we're looking at 2.6.35 dropping it due to the developers dissapearing again.

The code "will enhance the performance of the Linux operating system when virtualized on Windows Server 2008 Hyper-V or Windows Server 2008 R2 Hyper-V.". It sounds less like a bash against Linux and more like an enhancement to Linux's kernal for those who decide to run Server 2008 Hyper-V or Server 2008 R2 Hyper-V virtualized.

Thats why I started the topic. It seems innocuous and legit. On the other hand, they are conceding ground, making linux work better, enabling it to be run more smoothly. This is tantamount to Ford Motor Co. assisting Chevrolet.

They're doing it in exchange for not getting sued due to breaking the GPL.

The GPL doesnt say they have to write drivers. They could let support die forcing people to run windows products under hypervisor. The GPL only says that if they write drivers, they also have to be GPL.

http://www.osnews.com/story/21882/Micro ... _Violation

Data collection? It's kinda useful to business people. I'm not sure exactly how it works, but if they can use this to observe web traffic on windows servers, then it can be used to help specialize further products and market others.

I think what Jookia said is that they are caught releasing a driver with GPL components, so they are forced to release the code.

There's someone from Novell blogged that they had contacted M$ before they made the code public. Someone from M$ also blogged that it's not the reason they opened the code. Anyway the legal side of it depends on the interpretation of derivative works. The court has so far not needed to rule on it so it's anyone's guess how it's applied. The FSF's position is that any linking is a derived work, so any program that uses GPL library has to be in GPL. That's is also why LGPL is conceived because it's not the intention of a lot of projects. In the end, that's just FSF's position, it has not been tested in court. The FSF has clearly failed to act in cases that involved libraries only. For example, Asus has been using GPL code in their drivers for years and the FSF clearly know about it. So their position isn't really strong if they decided to sue M$.