Posted by: Dave Richeson | February 25, 2016

Good activity for an Introduction to Proofs class

I just read this post at Futility Closet. (Spoiler: Don’t click the link unless you want to know the punchline.) Perhaps the result is well known, but I hadn’t seen it before.

The post made me think of a neat project for an “Introduction to proofs” class. I’ll have to save it for the next time I teach that class (Discrete Mathematics at my college).

Start with the first n prime numbers, p_1,\ldots, p_n. Divide them into two sets. Let A be the product of the primes in one set and let B be the product of the primes in the other set (the product is 1 if the set is empty). For example, if n=4 we could have {2,3,7} and {5}, so A=42 and B=5. What can we say about A+B and A-B? Form a conjecture and prove it.

I like the problem because I could imagine the students will go through a process of (1) I have no idea. (2) Aha! I’ve figured it out. How do I prove it? (3) Wait, I found a counterexample. (4) Ahhh! I see what’s going on, here’s the theorem, and here’s the proof. (I think they would come up with a different statement of the theorem than the one in the Futility Closet post.)

Posted by: Dave Richeson | October 10, 2015

Editing a Very Poorly Written Proof

I’m teaching Discrete Math this semester. Discrete Math is our college’s “introduction to proofs” class. We spend a lot of time talking about and practicing proofwriting. In earlier blog posts I shared my “Nuts and Bolts of Writing Mathematics” and an “editing checklist” that I give to them.

Yesterday, I gave them an example of a very, very poorly written proof that I created. (It was fun trying to insert as many problems as I could into the short proof!) I had them work on the assignment in class for about 7 minutes. Then we came together and talked about all the errors that we could find, and finally we rewrote the proof in the correct way. Here’s a pdf of the assignment, and I’ve posted a screenshot below.

Screen Shot 2015-10-10 at 1.06.34 PM

In case you are wondering, here are the errors in the proof. Numbers refer to sentence numbers.

  • The proof should be begin immediately after the word PROOF.
  • The entire proof should be written in paragraph form.
  • 1. Capitalize the first word of the sentence.
  • 1. Split this sentence in two: The first sentence should be “Let m be an odd integer and n be an even integer.” In the second sentence we apply the definition of even and odd.
  • 1. Do not use k for both the even and the odd numbers.
  • 1. Don’t use a symbol (#) in place of a word.
  • 2. Start the sentence with a word, not a mathematical expression.
  • 2. The correct expression is mn2, not m2n.
  • 2. “Equivalent to” is a term we use for logical expressions, not numbers. It should be “equal to”—and in fact, we should just use “=.”
  • 2. Algebra error: the term 8k should be 8k2.
  • 2. In a 200-level math class we do not have to show all of these algebraic details.
  • 3. This is not how we show that a number is even—we must show that mn2 satisfies the definition of even.
  • 3. Do not use the word “obviously.”
  • 3. In a proof like this we’d have to show that the three terms satisfy the definition of even.
  • 3. “There” should be “their.”
  • 3. Missing dollar signs around 2k in the LaTeX code.
  • 4. Do not use the passive voice.
  • 4. Missing period at the end of the sentence.
  • 4. and 5. Write in present tense.
  • 5. Do not put an example in a proof.
  • 5. There are missing dollar signs in the LaTeX : after m=5 and before n=4.
  • 5. It should be first person plural, not first person singular (“we” not “I”).
  • 5. Replace “is = to” with “=.”

Here is our corrected proof:

Screen Shot 2015-10-10 at 1.20.24 PM

Posted by: Dave Richeson | August 27, 2014

Tangent lines to the sine function with rational slope

Today I was wondering the following thing (I won’t bore you with how I ended up with this question):

Are there any rational values of {m} for which the line {y=mx} is tangent to the graph of {y=\sin x?}

Clearly the answer is yes: {m=1.} But my gut feeling was that this was the only such {m.} After some head scratching, I obtained the following proof.

Suppose they are tangent at {x=x_{0}}. Then the point of tangency is {(x_{0},mx_{0})=(x_{0},\sin(x_{0})).} Moreover, the slope of the tangent line is {\cos(x_{0})}. Thus, {x_{0}} must satisfy the two equations:

{\sin(x_{0})=mx_{0},} and


Using a trig identity,




Note that {x_{0}} is an algebraic number—it is the root of a polynomial with integer coefficients.

In 1882 Ferdinand von Lindemann famously proved that {e^{a}} is transcendental (that is, non-algebraic) for every non-zero algebraic number {a,} and a similar proof holds for the sine and cosine functions.

Thus, because {\cos(x_{0})=m,} and {m} is algebraic, it must be the case that {x_{0}=0}. In particular, {m=1.}

This concludes the proof.


Posted by: Dave Richeson | March 25, 2014

Gabriel’s paper horn

I just returned from the eleventh Gathering for Gardner. One of the many special things about this unusual conference is that the attendees are strongly encouraged to participate in the “gift exchange.” We were each asked to bring a physical exchange item (one for each of the 350 conference-goers) or to submit a written contribution.

The bag of gifts that I received at the end of the conference was amazing. It contained exotic dice, clever puzzles, unique playing cards, mathematical earrings, 3d printed objects, cool pencils, puzzle books, artwork, etc. Nearly all of the gifts were handmade. We haven’t received the written gifts yet. What a fantastic idea!


My gift was a set of instructions and templates for making Gabriel’s horn out of paper cones. Gabriel’s horn is the surface obtained by revolving the curve y=1/x (x\ge 1/2) about the x-axis. Mathematics professors wow introductory calculus students by sharing its paradoxical properties: it has finite volume, but infinite surface area. As they say, “you can fill it with paint, but you can’t paint it.”

Photos of the paper horn (in white and in rainbow) and a photo of the cones are shown below. I’m linking to a pdf copy of the article here in case you want to make one.


Note: my inspiration was this blog post by Dan Walsh in which he makes a pseudosphere out of cones.


Posted by: Dave Richeson | February 18, 2014

Undergraduate Math Bloggers

I was interested seeing how undergraduate math students used blogs (and related platforms, like Tumblr). So I posted a call on Google+ and Twitter:

I received quite a few links. I’m looking forward to reading them. It seemed like there was a lot of interest, so I’m posting the unedited list here.

If you know of any others, post them in the comments.

Posted by: Dave Richeson | September 9, 2013

2013: the year of pi

A couple days ago I saw this tweet.

Pretty cool! Let’s see why


Two terms are easy to deal with:

\arctan(0)=0 and


But why is \arctan(2)+\arctan(3)=3\pi/4?

One way to prove this is via the inverse trigonometric identity

\displaystyle \arctan(a)+\arctan(b)=\arctan\left(\frac{a+b}{1-ab}\right)\, (\text{mod }\pi).

In this case

\arctan(2)+\arctan(3)=\arctan(-1)\, (\text{mod }\pi).

Because \arctan(2) and \arctan(3) are between 0 and \pi/2,  \arctan(a)+\arctan(b)=3\pi/4.

I didn’t use this approach on my first attempt to solve the problem.  (To be honest, I didn’t know this identity existed before finding it online.) I used geometry and trigonometry. We are interested in the angle \theta in the diagram below.


The law of cosines tells us that


This implies that \cos\theta=-\frac{1}{\sqrt{2}}, and hence \theta=3\pi/4.

(I wonder if there is a year of tau coming up sooner than the year 112233.)

Update: I just saw that Cut-the-Knot has a page devoted to this topic too.

Posted by: Dave Richeson | July 4, 2013

Using a kayak to measure the perimeter of a lake

I’m on vacation this week on a lake in northern Michigan (hold up your right hand, palm toward you, point at the first knuckle of your middle finger—that’s where I am). Yesterday I paddled around the perimeter of the lake in a kayak. On a whim I brought my GPS-enabled phone. My route is shown below. Note that the cartographer didn’t do a perfect job—despite what you see, I can assure you that I paddled only on water.

When I got back, the GPS said that I had paddled for 2.74 miles (14470 feet). Let’s say that I stayed exactly 50 feet off the shore for the entire trip. What is the perimeter of the lake?

We should probably make some assumptions about the shoreline and about my path. For simplicity, let’s say that my path is differentiable, that it bounds a region X, and that the lake is the set {N_{50}(X)}, where N_{r}(X)
is the radius {r} neighborhood of {X}. Intuitively, the boundary of N_r(X) is always r units off the port side of the boat. (See Ravi Vakil’s article “The Mathematics of Doodling” for more about this set.)

You may not think you have enough information to solve the problem, but you do! The clue is found in a well-known puzzler. Suppose we have an electrical wire running round the equator of the earth. We want to get it up off the ground and put it at the top of 50-foot electrical poles. How much longer must the wire be? [If you haven’t seen this puzzler before, stop now and think about it.] If the radius of the Earth is {R} feet, then the original wire must be {2\pi R} feet long. The new wire is a circle with a radius 50 feet longer than before, so its length must be {2\pi (R+50)=2\pi R+100\pi} feet. That is, it must be {100\pi\approx 314} feet longer.

It turns out that this argument can be generalized in various ways (again, see Vakil’s article for more information). First of all, if {X} is any convex set with a polygonal boundary, then, just like in the puzzle, the perimeter of {N_{r}(X)} is {2\pi r} units longer than the perimeter of {X}. Vakil illustrates this with the picture below.

It turns out that the exact same result holds if {X} is nonconvex—such as the region enclosed by my kayak trip. As the kayak goes in coves and around peninsulas, the “extra” regions of convexity and nonconvexity cancel each other out. In the end it is the same as going around a circle one time. Thus the perimeter of the lake is {2\pi\cdot 50\approx 314} feet longer than my kayak path: {14470+314=14784} feet.

Posted by: Dave Richeson | April 16, 2013

Countability of the rationals drawn using TikZ

I’m continuing my exploration of TikZ (here is my first post about TikZ). I will be showing my Discrete Math class how to “count” the positive rational numbers. (See this old blog post for more information about countable sets.) I used TikZ to create the picture below.

Screen Shot 2013-04-16 at 9.23.14 PM

Here is the source code for this figure. If you click on the link you can get an editable copy of the document in WriteLatex.

Posted by: Dave Richeson | April 12, 2013

Greatest living mathematician and expositor?

On Twitter I posed the following question:

I got a great repsonse. Here is the complete—unedited—list of names (in alphabetical order).

If you have other suggestions, add them in the comments.

Posted by: Dave Richeson | April 9, 2013

Bubble diagrams for functions in LaTeX using TikZ

I am teaching Discrete Math this semester (our intro-to-proof course). One of the topics is functions. Not surprisingly my students and I have to draw “bubble diagrams” for functions between finite sets—and we have to include them in LaTeX documents. Rather than simply sketching them in Adobe Illustrator and importing them as graphics, I decided to try creating them in TikZ. After a lot of tinkering I came up with something pretty nice (see below). Moreover, I set it up so that all the complicated TikZ and LaTeX commands are in the header of the document. So my students and I can easily generate new bubble diagrams.

Screen shot 2013-04-09 at 3.53.26 PM

I’ve put a sample document on WriteLatex. If you click on this link you can get an editable copy of the document. You can edit it and it won’t change my copy—so go wild with it! (This cool feature of WriteLatex is described on their WriteLatex for Education page.)

« Newer Posts - Older Posts »