"... what we're seeing actually is that with a lot of data, you ultimately see things that seem intelligent even though they're done through brute force ..."
This simple idea applied to problems like spelling (and obviously search) has an orders of magnitude improvement over previous hacks. So while MS Word might append those squiggly lines underneath your name when you type it into word, spell checking your name in google is more than likely to show your name and links to it. Check this simple python spelling example by Norvig ~ http://norvig.com/spell-correct.html
Add a bit of Maths (probability theory) and loads of data that Google collects and Google spell correction kicks MS Words F7 butt.
Yes, this is an old post but one thing that strikes me about the semantic web vs. brute force approach is that the semantic web approach is (assuming it works) much cheaper for startups, whereas the brute force practically dictates having tons of cash and an existing infrastructure.
Taken from this perspective, it's easy to see why the big players favor the brute force approach.
This simple idea applied to problems like spelling (and obviously search) has an orders of magnitude improvement over previous hacks. So while MS Word might append those squiggly lines underneath your name when you type it into word, spell checking your name in google is more than likely to show your name and links to it. Check this simple python spelling example by Norvig ~ http://norvig.com/spell-correct.html
Add a bit of Maths (probability theory) and loads of data that Google collects and Google spell correction kicks MS Words F7 butt.