3 ms·
People used it to search for information. You would ask it a question, and it would immediately give you the answer from huge repositories of information
by dstein 16y ago
People used it to search for information. You would ask
it a question, and it would immediately give you the
answer from huge repositories of information it had gathered
I found this part rather funny. They blatantly lied to this man. Google doesn't give you direct answers immediately, and they never will because that's not how Google makes moey. Instead they present you with dozens of SEO optimized links and paid advertisements that might (but usually don't) contain relevant information.
- orangecat 16y agoGoogle doesn't give you direct answers immediately, and they never will because that's not how Google makes moey. http://www.google.com/search?q=90210+weather http://www.google.com/search?q=90210+weather http://www.google.com/search?q=2%2B3 http://www.google.com/search?q=2%2B3 http://www.google.com/search?q=aapl http://www.google.com/search?q=aapl
- dstein 16y agoI obviously wasn't referring to Google's canned responses which offer no real insight or understanding. The description they gave make it sound like futuristic AI, when clearly its not. Let me know when Google approaches IBM's Watson level. http://www.google.com/search?q=is+five+dollars+more+than+8+cents&ie=UTF-8&oe=UTF-8&hl=en&client=safari http://www.google.com/search?q=is+five+dollars+more+than+8+c...
- nostrademons 16y agohttp://www.google.com/search?q=5+dollars+in+cents http://www.google.com/search?q=5+dollars+in+cents BTW, there are actually queries that give you exact answers which are not specifically programmed in, eg: http://www.google.com/search?q=empire+state+building+height http://www.google.com/search?q=empire+state+building+height
- Dn_Ab 16y agoThat is like saying a racecar tuned for closed tracks is more impressive than a general purpose modern car because its faster. Both leverage the same principles and similar components but with different focuses and optimize for different results. Watson is not futuristic when compared to what Google does, its mainly its packaging that makes it so readily impressive. Google are focused on different goals. Watson is optimized to list a set of confidence weighted answers/hypotheses. Google is optimized to surface a list of relevance and authority weighted documents. A Watson like google is orthogonal and could essentially be another UX on top of regular google that would completely transform the product to a different set of use cases. Your example is easy when run through an engine that's tuned for it. But it does look like google is experimenting. http://www.wolframalpha.com/input/?i=is+6+dollars+less+than+5+pounds http://www.wolframalpha.com/input/?i=is+6+dollars+less+than+... http://www.google.com/search?q=who+produced+all+of+the+lights http://www.google.com/search?q=who+produced+all+of+the+light... http://www.wolframalpha.com/input/?i=who+produced+all+of+the+lights http://www.wolframalpha.com/input/?i=who+produced+all+of+the...
- thamer 16y agoRe: “is 6 dollars less than 5 pounds”, this is how Wolfram Alpha interpreted it: “TrueQ[$6 (US dollars)<5 lb (pounds)] ”. It interpreted pounds as a unit of mass, not as the currency and gave “False” as a result. You have to pick “British Pounds” from a list of other interpretations in order to get the right answer, “True”. I don't know what it says about your argument, other than it's hard to tune the engine.
- muyuu 16y agoWorked for me (interpreted "pounds" as "British pounds")