So it’s been a few days since Microsoft announced their offer to purchase Yahoo, and since then there’s been endless speculation on motivations and possible outcomes.
Except nowhere have I seen the obvious conclusion, which means I’m either a complete moron, or some sort of freak genius. And I’m not betting on the latter, but wanted to get this out there first just in case I happen to be right.
If you look at all the issues MSFT will face in assimilating YHOO (sorry, couldn’t resist that one) a couple seem to be repeated over and over: culture differences and technology differences.
YHOO runs a large part of their infrastructure on BSD. To an engineer used to working in that sort of environment, porting to a Windows stack would be, to say the least, painful – unappetizing to the point a career chance would be in order – there go all those developers that SteveB wants so badly.
On the other hand, MSFT can’t afford to have a large part of their infrastructure running on something that isn’t Windows – you know, dogfood and all that stuff. And while all this is going on, there’s that wonderful success known as Vista.
So what’s the obvious answer to the dilemma?
Simple – Microsoft needs to do what they’ve always done in the past: copy from someone else, in this case Apple.
Gut the Windows OS, port the GUI and Win32 API to run on top of BSD, and there’s the solution to the takeover, dog food, and by the way fixes the whole Vista mess.
Seriously, is there another company in the world that MSFT could buy that has the experience and expertise with BSD that YHOO has? And if you’re a BSD expert, is there anything in the world that would be a bigger, more rewarding challenge than putting Windows out of everyone’s misery?
I mean, it can’t just be about eyeballs, can it?
GronkX's Weblog 
February 6, 2008 at 1:17 am |
And why exactly do they need to spend $44B to build a better window manager?
February 6, 2008 at 3:39 am |
They obviously don’t need to spend $44B to build a better window manager. If they’re trying to compete in the online advertising world – basically going after Google – it’s an added bonus, and it deals with the issue of converting Yahoo’s infrastructure.
I think that only way they’re going to raise the stock price is by doing something drastic, and by releasing a BSD-based OS, they could start with a clean slate – get rid of all the compatibility cruft that’s accumulated over the years in windows, re-do the security model, and move to a much more modular architecture.
Longer term, they should also be able seriously reduce headcount – which is about the only way they’re going to be able to bump the stock price.
February 6, 2008 at 4:56 am |
Let me try to put this kindly:
Your theory makes no sense. Vista’s problem isn’t its kernel. The NT kernel is rock solid. Replacing it a different kernel, no matter how good, won’t improve Microsoft’s situation significantly. The problems with Vista are in the device drivers and the higher level code, which would be just as cumbersome and buggy on a new kernel.
Or, to put it another way, Microsoft switched from their buggy DOS kernel to NT in Windows XP, around the same time Apple did.
February 7, 2008 at 6:06 pm |
Microsoft has always counted on increasing chip speed to bail them out – sure, new releases were slow, but a year or two down the road we had much better performance and the system became usable.
That trend has pretty much stopped, and the existing code does not handle multiple cpus/cores as well as other OSes.
Plus, it’s very expensive – look at what Apple has done in the last 5 years compared to MSFT, and then compare the number of engineers they’ve done it with.
February 21, 2008 at 1:36 am |
Problem is, Yahoo! isn’t running on BSD anymore; they’re in the process of migrating everything over to RedHat Enterprise. All the new stuff is already on RHEL.
I doubt after all the badmouthing that MS would embrace Linux.