WolframAlpha (Better than Google?)

a reply to: Templeton

All the examples you posted are pre-written user inputs, to which the program has been programmed to answer the adequate pre-written response. As soon as you deviate from the "examples", you can clearly see the limits of the program.

a reply to: swanne

Yes humans are involved. Read the FAQ.

originally posted by: swanne
a reply to: Templeton

All the examples you posted are pre-written user inputs, to which the program has been programmed to answer the adequate pre-written response. As soon as you deviate from the "examples", you can clearly see the limits of the program.

I'm not a paid advocate. Take what you will from it. Your witch hunt is going to let a great tool get away from you though.

a reply to: Templeton

I have nothing against the program, I'm not in a witch hunt either! I actually like these kind of interface.

I'm just giving my opinion as a guy who programs these kind of interface as a hobby.

Sorry if I ruffled some feathers, that wasn't my intention, I swear!

a reply to: swanne

I have a problem with context sometimes. Witch hunt was not what I meant, more like healthy criticisms. My bad, please forgive me. I was having no luck trying to track down his blog post about how their biggest challenge is in interpreting natural language queries, much like a chat bot's would be. Here it is: blog.wolframalpha.com...

Cheers

a reply to: Templeton

Don't worry about it, the fault is mine, I tend to write with a deadpan manner, sometimes making it hard to tell if I'm opposing the topic or supporting it.

Thanks for the link! Hehe, the examples for "artificial stupidity " are cracking me up. The Fall of Troy happening in 6.8 seconds is particularly precious! lol

I'll be reading more about it. Your link helps me deduce better the programming logic behind it, and is even inspiring me to try new things next time I write an AI.

Thanks!