I've mentioned this before, but I agree wholly. Google's product "strategy" gets shit on all the time (killedbygoogle.com), but the real issue is that the company is structured as if that's what they want to have happen. It's a loose association of warring VPs trying to establish big enough fiefdoms to be able to buy a third house. There's no real benefit to maintaining existing products for them. They just want to reorg things so they can amass more and more reports and justify their next enormous equity refresh. Killing old products to "make room" for new hotness is a pretty good way to do that.
It took them the failure to dominate the next two new markets and the fear of remaining rich but becoming marginal.
With the internet they had a success with IE but saw companies developing their services with new non Visual*/.Net languages and running them on Linux and MySQL.
