Well the problem for MSFT is that they quit (or never had) a culture of writing good software. They used market muscle and “inevitability” (think of the second place dem candidate for president) to get market share and revenue. Now this is all coming home to roost. Buying Yahoo looks like crying uncle on software completely. I can imagine all the internal politics in re-focusing the company on software.
MSFT is its own worse enemy.
Win7 will be a joke. They should buy RHT, put AERO on top of it, fire 90% of the product managers and marketing MBA types, open source some infrastructure components (exchange?, Share point? SOA?), and acquire a few leading software companies (adobe, SAP, others). And put all the energy into writing really great, consumer and business software.
They are trying to become Google, but if you look at the valuations and 10ks of both even that would be a disaster.
Actually, there’s a better way to do it AND maintain backwards compatibility, but I’m not going to give Microsoft any ideas for free.
I will drop a hint, though. Microsoft should give up on any pretense of in-current-version support of legacy applications and force them to run in a sandbox - like the OS X/Classic relationship.
Paradoxically, most of the decent application software MS has is the product of buy out other software companies. It gets better integated once it comes “in-house” and the programmers have access to the proprietary code they need to make it work with the other MS stuff, but then development go to crap. I remember reading somewhere that the only software application MS ever had that was totally written in house was Bob.