Tuesday, October 21, 2014
Novel insights-the triangle inequality
In mathematics we are often confronted with choices. Which line of attack do we take? How can we approach this problem from a different perspective? Can we combine two approaches and merge them to yield a solution?
In my opinion a great progress has taken place when we combine two or three approaches and bring to bear upon a problem a new approach which would be combinatorial in nature.
We begin by asking a simple question:
Can the sum of any two angles of a triangle be greater than the third side be proved in more ways than one?
Here we are about to follow a slightly different approach than Euclid's 5th notion which be a traditional proof. Hopefully this should get us started on some more different approaches than we may be used to think.
Take a circle and have AB as a diameter (longest chord) and take a point C on the semicircle. Then what we have is that ∡ACB = 90 ͦ (angle inscribed in a semicircle is a right angle)
Then AB is the hypotenuse and AC and BC are arms. So AB^2=AC^2 + BC^2
We now search for Pythagorean triplets like 3, 4, 5 and 5, 12, 13 and in each case if we sum any 2, they are greater than the 3rd side whatever combination we take. So by an inductive argument, we can infer that sum of any 2 sides is greater than the 3rd side.
Now suppose, modify point B so that it lies in the below semicircle such that the Pythagoras theorem no longer holds good. Then in that case, we call this point B’ and we must have AB’ < AB (as AB was the diameter)
So we can write that AB’ + γ = AB (for some finite gamma depending on point B’ location)
If we fix C, then AC has constant length, however length (BC) has changed. It has increased. So we must have BC + α = BC’.
If in the original triangle we had,
AC + BC > AB
This translates to AC + BC’-α > AB’ + γ
AC + BC’-AB’> α + γ
And this is obviously greater than 0 since both α and γ are finite entities.
So we again have AC + BC’ > AB’
Thus, even if the lengths of the original triangle are modified, the property still holds good. Similarly if we move point B to the upper semicircle, keeping A fixed, it would be possible to prove the above.
Subscribe to:
Post Comments
(
Atom
)
No comments :
Post a Comment