[clug] Thoughts on Microsoft's Linux patch?

[Alex Satrapa]

On 24/07/2009, at 12:24 , Chris Smart wrote:

> I'm curious to hear what CLUGGers think about the hyper-v code from
> Microsoft for the kernel?

I absolutely trust that Microsoft will continue to innovate in the  
industry.

They'll stick with releasing new versions of Hyper-V drivers for Linux  
until it doesn't suit them to do so. Ask the Plays For Sure consortium  
about Microsoft's commitment to technical standards.

Step 1: Release Hyper-V drivers for Linux that make it work with  
current version of Hyper-V
Step 2: clients adopt Hyper-V for virtualisation, sign site-licences  
which include clauses about paying Microsoft for using Hyper-V on all  
machines that do virtualisation regardless of technology used
Step 3: Upgrade Hyper-V with patented algorithms, do not release  
updated Hyper-V drivers for Linux
Step 4: clients forced to migrate from Linux to something that is  
supported by the "New! Improved!" Hyper-V, or put up with Linux guests  
running at 50% of the speed it was running yesterday, or cancel their  
site licences - thus instigating a complete audit by the BSAA.

Or perhaps they'll be model citizens of the Linux programming world.  
Then they can point out that they've gone to great lengths to make  
sure that their contributions to Linux have been "Best of Breed". Then  
they'll start making noise about other Linux contributors not  
producing code of the same quality as Microsoft. Then they'll upgrade  
Hyper-V with some warning in the same vein as the "DR-DOS is known to  
break Windows" warning of yore. What is your typical trained-monkey  
MCSE going to think of Linux when every attempt to boot a Linux guest  
is accompanied by warnings that, "Linux is not known to be as  
thoroughly tested as Windows. You should really migrate all the  
applications on this guest to a Microsoft Windows platform which is  
supported by Microsoft. If you click 'OK' just be warned that it is  
not outside specification for Linux to cause demons to issue forth  
from your nose"?

Or even worse, what happens when Hyper-V works just fine, Linux as a  
guest works wonderfully, but the licence auditing software continually  
reports, "This guest OS is not licenced by Microsoft"? Will the  
accountants blame Hyper-V or Linux for that warning?

Or perhaps they've cleverly included some bugs in Hyper-V which  
interact with bugs in the Linux drivers so that things work correctly,  
then when some hapless Linux maintainer "fixes" the obvious bug (this  
value in the Kerberos header is wrong), Hyper-V support falls apart.  
Microsoft will then claim a patent on the bug (passing the value with  
the wrong sign means we get 1 more value in the direction that we're  
interested in), such that fixing Linux to work with/around the bug  
will become a violation of the patent. The fact that the patent  
wouldn't stand up in court won't be much comfort to people who can't  
afford to challenge Microsoft in court. How much money did SCO have  
injected into them before they crashed and burned?

I am eagerly awaiting the next episode of Microsoft vs World. Perhaps  
they'll actually innovate and produce some *new* way to piss in the  
collective breakfasts of their friends and customers.

Don't get me wrong, I'm not bashing Microsoft. I'm simply  
extrapolating based on past performance.

I hope youse found this entertaining.
Alex

-------------- next part --------------
A non-text attachment was scrubbed...
Name: PGP.sig
Type: application/pgp-signature
Size: 220 bytes
Desc: This is a digitally signed message part
URL: <http://lists.samba.org/pipermail/linux/attachments/20090724/cf556573/attachment.pgp>