VISUAL MATHEMATICS
"I see dead easy proofs"
Joe Hurd
@gilith
http://www.gilith.com/
Ignite Portland 9
23 September 2010
_________________________________________________________________________
BIOGRAPHY
Joe works for Galois, Inc. in Portland, OR, where he uses functional
programming languages to construct correctness proofs of software.
He studied at the University of Cambridge, receiving degrees in
mathematics and then computer science as he made the realization
that it was easier for him to program computers to find proofs than
to do them by hand. In his free time Joe contributes to open source
projects, carries out independent research and regularly gives talks
to audiences at all levels.
_________________________________________________________________________
ABSTRACT
It is a sad fact that many people believe that mathematics is about
doing arithmetic and memorizing formulas. However, if these people
were to ask a mathematician what their profession is about they
might be surprised by the answer. "A mathematician, like a painter
or poet, is a maker of patterns. If his patterns are more permanent
than theirs, it is because they are made with ideas." So said the
number theorist G. H. Hardy in his 1940 essay 'A Mathematician's
Apology', and in this talk I will attempt to convey this alternative
point of view using a series of examples of visual mathematics. Each
example will take the form of an easy-to-explain statement about
numbers or geometric shapes, followed by a purely visual proof of
the statement. The success of the talk may be judged by the number
of 'Eureka!' moments it generates in the audience.
_________________________________________________________________________
SCRIPT
Hi, my name is Joe Hurd, and I'm going to show you some visual
mathematics.
At school mathematics is presented as calculations, algebra, and
(worst of all) memorizing formulas, but in truth mathematics is no
more about formulas than astronomy is about telescopes.
Astronomers have to use telescopes to see and measure the cosmic
events they're really interested in, and in the same way
mathematicians have to use formulas to understand and prove properties
of mathematical objects. In this talk I'm going to show you some
visual proofs that don't need formulas.
Let's warm up with a puzzle. If you have a bunch of 2x1 dominoes you
can perfectly tile a regular chess board, but can you do the same if
you remove two squares at opposite corners?
It turns out you can't, because if you cover the board with a checker
pattern then each domino must cover one light square and one dark
square, but you've removed two squares of the same colour, making any
tiling impossible.
You can use the same kind of parity argument to prove that there is no
knight's tour on an odd-sized chess board that ends on the same square
it started, because when a knight moves it alternates between light
and dark squares.
OK, now we've warmed up, let's look at a classic proof of geometry.
Suppose you have a semi-circle, and you draw two lines from the
corners to a point on the circumference, they will always meet at 90
degrees.
To prove this, all we have to do is rotate the whole picture 180
degrees around the centre of the circle. Each line becomes two equal
parallel lines, and these form a four-sided shape. Since the two
diagonals have the same length, the shape must be a rectangle, so each
corner is 90 degrees.
And now for something completely different: number theory! A lot of
proofs begin life when someone notices a pattern and wonders if it
will continue indefinitely.
For example, adding the first n odd numbers together results in the
nth square number. Why should this always be true?
The proof involves constructing squares in layers, where each layer
contains an odd number of points! In this picture we can immediately
see why the pattern is true, and that it will carry on forever.
What is the result of adding up the first n numbers, or alternatively,
how many blocks are there in this diagram?
It's hard to count them like this, but using your tetris skills you
can neatly fit two such diagrams together, and then it's easy to see
the result must be half of n * (n + 1).
Let's finish with a proof from classical times: Pythagoras' Theorem.
Given a triangle where one of the angles is 90 degrees, called a
right-angled triangle, extend each side of the triangle to a square.
Pythagoras' Theorem says that area of the two smaller squares is the
same as the area of the big square, and we're going to prove this
non-obvious fact.
This is an IKEA proof, and for assembly we will need: a square box
where each side has length (a + b), and four identical copies of the
right-angled triangle.
Now we simply put the four triangles into the box in two different
ways. In the first way we arrange them into two pairs, and the
left-over space is the area of the two smaller squares.
In the second way we arrange them along the sides of the box, and the
left-over space in the middle is the area of the big square. The
left-over space must be the same in both ways, so we have proved
Pythagoras' Theorem.
That's all the visual mathematics I have for you today: thank you for
your attention.
_________________________________________________________________________
IMAGE CREDITS
http://www.flickr.com/photos/thearchigeek/99162829/sizes/l/in/photostream/
http://xkcd.com/774/
http://www.flickr.com/photos/judepics/449405116/sizes/l/in/photostream/
http://www.flickr.com/photos/badastronomy/4179773665/sizes/o/in/photostream/
http://www.flickr.com/photos/ogcodes/377800888/sizes/o/in/photostream/
http://www.flickr.com/photos/dullhunk/426622486/sizes/o/in/photostream/
http://www.flickr.com/photos/myklroventine/2332789392/
http://en.wikipedia.org/wiki/File:Kapitolinischer_Pythagoras_adjusted.jpg
http://commons.wikimedia.org/wiki/File:Paradoxical_decomposition_F2.png
http://www.flickr.com/photos/9729909@N07/4706488737/
http://commons.wikimedia.org/wiki/File:Lorenz_attractor.svg
http://cookiemagik.deviantart.com/art/Love-for-Tetris-64805505