Kill Vista and start the a new operating system that focuses on creating a lightweight core you can grow and expand for the next 20 years. This means stripping

Kill Vista and start the a new operating system that focuses on creating a lightweight core you can grow and expand for the next 20 years. This means stripping the OS to it's bare essentials and allowing modules to implement different features. Kill the visual effects and focus on stability, speed and security. This should be as small as possible, without loosing any of the items just mentioned. Ship a version of Office lite, that strips out many of the extra features from Word/Excel/Powerpoint and provides the basics. Ship stripped down copies of Office '07 if necessary. Be working on an online version of office. It's not important now--but in 5 years when Google's stolen your Office clients you'll be wishing you had a competitor. Make syncing to/from Office and Web Office seamless. This means I can write my document in Word on my desktop and with no effort edit the document online. Leverage the user base here. I don't see users shifting away from Office completely--more than likely there will be a balance between desktop and web users. Microsoft can and should dominate both. Give up on the Yahoo! deal and partner with more content producers--say AOL, by providing advertising in order to expand your reach. Your Facebook/Digg deals just aren't doing a lot here--focus on vertical niches that are profitable and then slowly fill the cracks. Make advertising rates extremely competitive--perhaps even consider taking a loss until you build momentum in the market. Adopt a pay-per-action model for search advertising instead of pay-per-click. There are some issues with this--but it's not impossible. Advertisers are only paying for leads and sales now, merging affiliate marketing and search engine marketing--two fields that have been closely related for years. You're not going to beat Google by copying them, you've got to change the game somehow. Re-focus your efforts on search. MSN is the easiest engine to game right now--without decent search results, search advertising will never work. Consider search engine acquisitions and heavily investing in researching new ideas (think AI, Semantic, Social, Personalized searching). Do what you did with Halo again. That seems to have worked well for you. Just a few thoughts. I do believe the Yahoo! deal will not help Microsoft. Sure the reach Yahoo! has helps Microsoft--but there are other (better) ways to achieve this. Not to mention without the core (good SERPs) search advertising just won't be effective. Much of what I've said can be summarized in focus on the core and strip everything else. It's weighing you down.