3 ms·
It's hard to see how a computer would produce a proof of the Banach-Tarski "Paradox" ( http://news.ycombinator.com/item?id=411043 http://news.ycombinator.com/i
by RiderOfGiraffes 18y ago
It's hard to see how a computer would produce a proof of the Banach-Tarski "Paradox"
( http://news.ycombinator.com/item?id=411043 http://news.ycombinator.com/item?id=411043 )
without going through effectively the same steps as a human. It doesn't seem likely that it will be reduced to calculations, or simple formula manipulations such as can be performed by Mathematica.
And in case you think that B-T is irrelevant, informal reasoning about infinities produced all sorts of results that caused people problems. Formal reasoning about infinities gave us calculus for real, that works wherever it's known to work. And I use infinities on a daily basis designing and implementing systems to assist with tasks similar to air traffic control and automated target tracking.
Computers can be used to explore and experiment with numbers, geometry and some structures, and will probably one day be able to assist with exploring more abstract concepts, but as things stand today, there's a great deal of mathematics that no one can see how to do with computers.
To say otherwise is to demonstrate ignorance of both fields.