The M$ attack is mostly based on the decentralized nature of Linux (which is also it's secret weapon against MS)..
This attack might just get some mind-share with many CxO types, who are used to centralized controls being a generally good thing.
As an example
Does Red Hat, for example, take responsibility? It cannot, as it does not produce the Linux kernel. It produces one distribution of Linux.Yeah-- the Red Hat distribution Kernel. -- and they have the full source code. This is rather different than someone who is producing a Win-CE device and may have access to the source code for certain parts of the OS but are not likely to be able to do more than minor tweaks without running up against legal problems with MS.
Since everybody has access to the source code, however, if Red Hat were to do something nasty with it, their changes would be open for anybody else to recognize and complain about. When you're dealing with a binary-only distribution, it's a lot easier to hide nasty code (whether MS wanted it in there, or it was inserted by s hacker who got quiet access to their systems.
They then attack things like the number of people posting to the Kernel -- claiming that MS has more 'real' coders 'really' working on Windows than the Linux kernel does. (room for all sorts of statistical and semantic slight of hand in that statement).
Then, at the end, he talks about how Linux has "no single development environment". -- Well, to the extent to which having only one development environment could be seen as a good thing, my response would be that Linux is the development environment. This has been a semantic slip for both Unix and Linux for years. It is developed by it's users, and used by it's developers. Unlike Windows, there is no need to separate the two. Where a tool has been missing for powerful development, it has been added to Linux, itself, rather than added so some mythical 'development environment'.
If you think that Linux isn't a powerful platform for development, then you should ask yourself how it's gone from a hobby OS to the biggest threat to Microsoft's multi=billion=dollar=a=year monopoly in the space of only a few years.
This then reaches into their other 'myth' -- That Linux is missing critical parts... He talks about how various companies have done pilots and found that certain critical pieces were missing from Linux. This is probably quite true. As few as 2 or 3 years ago, Red Hat itself was discouraging companies from looking at Linux as a complete desktop solution. In the last couple of years, this has changed, and most of the critical missing pieces have been developed.
IBM's challenge to 'eat it's own dog food' and move general users to Linux is both proof of how far Linux has gone to being a capable replacement for MS-Windows and part of that process. Moving a company like IBM from Windows to Linux is not a minor undertaking and would not have been started lightly.
That having been said, IBM's move to the desktop is going to find hiccups. This goes back to the previous arguments. Linux IS the development environment. It is developed by it's users, and they all have access to the source code (some are capable of using that source code. Others are capable of hiring people that can).
If you don't think that a company like IBM is capable of fixing the few holes that it's going to find in the process of switching over, then I would suggest that you turn around and take a good--