The trends of this october are about some upcoming products.
A clear analisys of QuarkXPress failure si sketched in roughlydrafted site.
I do not think the same consideration can be applied to Vista.
As Joel said, M$ can throw away much money before only starting to see its market reduced.
This lead me to the (hoped?) big fear of M$ for open source.
It is a dream, baby.
Open Source in Italy has success because of high costs of developement library.
When I tried to look for a commercial application to delivery some web graphic stuff, the prices was crazy.
The Open Source solution needed us only to buy a book(!).
Open source deliver a sufficient quality at low mantenance costs.
A smart developer in a 3-people team can set it up and deliver a small project.
If really M$, Sun, IBM'd start fear open source, they can cut their prices of 70% and all the drama will end quickly.
But the truth is Linux is not evloving: it is growing in the SysAdmin area only, popping out old dinosaurs like Sun Servers, replaced by cheap PC.
Sun Servers needs custom monitor and mouse (!) and their price is high compared to equally fast high-end pc.
The server segment is so over-priced that even Apple has succesfully proposed its over-priced servers (this is very fun!).
In the last five years I have seen no new on operating systems, or strong standard evolution.
EJB specs seems quite dead, hibernate is old, and the only "new" ideas are RubyOnRails and... PHP5.
On programming languages python seems growing well.
By the way there is much hypo on Ruby, but its base class library is very small compared to python or perl.
Ruby community is very well organized and can eventually bring Ruby to success, but for the meantime I see more success on PHP if you need a rock solid road.
Finally I have started to work on Squeak Smalltalk Weekly News, and I suggest you to look at it sometimes.