Who says apple hasn't built a viable alternative already? 12 billion might just be the cost of keeping them out of the search market.
It's not 2000 anymore. Building a web scale search engine no longer requires novel distributed systems work, and the most viable search algorithms have strong open source implementations. It's still a very hard problem to get right, but you can create an "interesting" poc in a few weeks. Recent work on unspervised content based recommenders can help reduce click stream data requirements.
It's not 2000 anymore. Building a web scale search engine no longer requires novel distributed systems work, and the most viable search algorithms have strong open source implementations. It's still a very hard problem to get right, but you can create an "interesting" poc in a few weeks. Recent work on unspervised content based recommenders can help reduce click stream data requirements.