2+3=Output: The Factory Model of Modern Academia

August 24, 2026August 24, 2026 / Michael Xue / Leave a comment

Behold—the “2+3” BS-to-PhD pipeline at XXX…XXX University—or, as it increasingly feels, Education™: now optimized for throughput, compliance, and quiet efficiency.
Welcome to the system.

Input: student, age ~18, mildly uncertain, statistically hopeful.
Output: PhD, age ~23, highly specialized, existentially ambiguous.

Processing time has been reduced. Reflection cycles deprecated. Exploration modules removed for efficiency.

The university no longer presents itself as a place of learning. That language is… outdated. It is now an industry node—a clean, well-lit facility where human curiosity is streamlined into measurable output. The brochures still say “discovery,” but the architecture says “production.”

You enter. You are assigned a track. You proceed.

There is no wandering here. Wandering introduces variance. Variance reduces efficiency. Efficiency is the objective.

The educators—once professors, mentors, inconveniently human guides—have been reassigned in function. They are now closer to system operators. Their role is not to inspire, but to maintain flow rate. Keep the pipeline moving. Prevent blockages. Ensure each unit reaches the next stage on schedule.

Questions are permitted, but only if they align with the system’s direction. Doubt is… inefficient. Changing your mind? That is a system error.

Somewhere along the line, education stopped being about forming a mind and became about shaping a product. Smooth edges. Standardized outputs. Predictable competencies. You are not encouraged to become unpredictable—that would make you difficult to process. Instead, you are optimized.

The pipeline hums.

Students move through it in tight formation, each one a nearly identical unit of ambition and exhaustion. They learn quickly—because they must. They specialize early—because they are told to. They produce—because that is what the system measures.

And yes, from the outside, it works beautifully. Degrees are awarded. Timelines are shortened. Metrics improve. The machine is efficient.

Inside, however, something quieter happens.

Curiosity is trimmed to fit deadlines. Depth is compressed into deliverables. Identity—once something explored slowly—is selected early and rarely revisited. The process does not ask who you might become. It asks only: what function will you serve?

The metaphor is no longer subtle. This is not a classroom. It is a factory line.

And the students—bright, capable, full of potential—begin to resemble something else. Not thinkers in formation, but workers in sequence. Repeating, producing, advancing. Not quite forced, not quite free. Just… moving.

The system does not need to coerce. It only needs to continue.

At XXX…XXX University, the “2+3” model is presented as the future. Faster. Leaner. More efficient.
And perhaps it is.
But in that future, the university is no longer a place where minds are cultivated.
It is a place where they are processed.

Let My People Go

August 17, 2026August 17, 2026 / Michael Xue / Leave a comment

There is a phrase from the Bible that has always struck me for its extraordinary simplicity:

“Let my people go.”

It is a demand for liberation.

I want to borrow those ancient words for a different kind of liberation.

By “my people,” I mean mathematics itself.

Let the mathematics go.

Let it out of the cage we have built around it.

For decades, we have been told that mathematics education must become more engaging, more relevant, more collaborative, more student-centered, more exploratory, more interdisciplinary, and more equitable.

There is nothing inherently wrong with any of these goals.

But somewhere along the way, I wonder whether we began treating mathematics itself as the problem.

Mathematics became something that needed to be reformed before students were allowed to encounter it.

And what if the problem is not mathematics?

What if the problem is that we have forgotten to let students actually do mathematics?

A child encounters a mathematical question.

She tries something.

It doesn’t work.

She tries again.

She notices a pattern.

She asks why.

She makes a conjecture.

She tests it.

And then comes the most important question:

Can you prove it?

That is mathematics.

Not because proof is the only thing mathematicians do.

But because proof transforms a suspicion into knowledge.

A pattern may suggest that something is true.

Examples may persuade us that something is true.

A calculator may confirm it for a thousand cases.

But mathematics asks for something more:

Why must it be true?

This is one of the great intellectual ideas of mathematics, and it should not be reserved for the final years of schooling.

A student should grow up knowing that mathematics is not merely a collection of procedures to execute.

It is a world of definitions, assumptions, conjectures, arguments, counterexamples, and proofs.

Consider a simple statement:

The sum of two odd numbers is even.

A student can calculate:

$3+5=8$ ,

$7+9=16$ ,

$11+13=24$ .

Wonderful.

But none of these calculations proves the statement.

They merely illustrate it.

The mathematician asks:

Why can this never fail?

And now we can write

$2m+1+2n+1=2(m+n+1)$ ,

which is even.

The calculation has become an argument.

The pattern has become a theorem.

This transformation—from seeing that something happens to knowing why it must happen—is one of the most beautiful things mathematics has to offer.

Yet students can spend years in mathematics classrooms without experiencing it.

They can become remarkably competent at manipulating symbols while having little idea what a mathematical proof actually is.

Then one day they are told:

“Prove that…”

And suddenly everything falls apart.

This is not necessarily because the students cannot reason.

Perhaps we simply never taught them what mathematical reasoning looks like.

Perhaps we gave them answers when we should have given them questions.

Perhaps we trained them to ask:

“What formula do I use?”

when mathematics was asking them to ask:

“Why is this true?”

I am not arguing against reform.

Good teaching matters.

Students matter.

Motivation matters.

Multiple representations matter.

Discovery matters.

Problem solving matters.

But these things should be servants of mathematics, not substitutes for it.

We should not be afraid of the mathematics.

We should not apologize for definitions.

We should not hide theorems behind endless activities.

We should not treat proof as an advanced specialty.

And we certainly should not confuse doing something mathematically flavored with doing mathematics.

Let the students encounter mathematics.

Let them encounter difficult ideas.

Let them struggle.

Let them make mistakes.

Let them discover patterns.

Let them argue.

Let them be wrong.

Let them find counterexamples.

Let them ask why.

And then let them prove.

Perhaps what mathematics education needs is not more educational machinery.

Perhaps we need to get out of the way.

Let them meet mathematics itself.

And let mathematics speak for itself.

What Did We Reform Mathematics Into?

August 14, 2026August 14, 2026 / Michael Xue / Leave a comment

I have a simple question about mathematics reform.

What did we reform mathematics into?

I don’t mean to ask whether every reform is bad.

Some changes have been useful. Teachers have learned things about how students learn. Technology has opened possibilities that didn’t exist before. Some old practices deserved to be reconsidered.

Fine.

That’s not the question.

The question is what happens when we keep adding things to mathematics education until, eventually, we have everything except mathematics.

We want students to collaborate.

Good.

We want them to communicate.

Good.

We want them to explain their thinking.

Excellent.

We want them to explore.

Fine.

We want them to make conjectures.

Wonderful.

We want multiple representations.

Sure.

We want authentic problems.

Absolutely.

We want students to reflect on their learning.

Okay.

We want them to reflect on their reflection.

I’m beginning to worry.

Then we want them to collaborate while reflecting on their representations of authentic problem-solving experiences.

At this point, someone should probably ask:

Where did the mathematics go?

⸻

Imagine walking into a modern mathematics classroom.

Students are sitting in groups.

They are talking.

They are moving sticky notes around.

Someone is drawing a diagram.

Someone else is explaining their thinking.

A third student is asking a probing question.

The teacher is circulating.

There is a rich mathematical discourse taking place.

Everyone is engaged.

Everyone is communicating.

Everyone is collaborating.

Everyone is reflecting.

Nobody has done any mathematics.

Well, perhaps that’s unfair.

Maybe they have.

Maybe one student has written

$(a+b)^2 = a^2+2ab+b^2$ .

But before anyone is allowed to do anything with it, the class must first explore multiple representations of the expression.

So the students draw a square.

Then they color the regions.

Then they discuss the visual model.

Then they compare it with an algebraic representation.

Then they explain which representation they prefer.

Then they write a reflection:

Today I learned that there are many ways to represent an algebraic expression.

Excellent.

But I have a small question.

Do they know that

$(a+b)^2=a^2+2ab+b^2$ ?

Or did we merely have a very successful meeting about it?

⸻

This is where things get interesting.

We criticize traditional mathematics education for being too procedural.

“Don’t just teach students how to do it. Teach them why.”

Fine.

I agree.

But then we sometimes replace the old procedure with a new one.

The old procedure was:

Learn the formula.
Use the formula.
Get the answer.

The new procedure can become:

Explore.
Collaborate.
Represent.
Discuss.
Share.
Explain.
Reflect.
Revisit.
Re-represent.
Reflect on the re-representation.
Complete the exit ticket.

Somewhere around step 9, the quadratic equation has quietly left the building.

Both approaches can become mechanical.

The old procedure produces an answer.

The new procedure produces a classroom activity.

Neither guarantees mathematics.

⸻

And here is the irony.

Mathematics already contains an extraordinary form of discovery.

We don’t have to manufacture it.

Take two circles.

Give the students

$x^2+y^2=r_1^2$

and

$(x-d)^2+y^2=r_2^2$ .

Let them subtract.

Let them simplify.

Let them substitute.

Eventually they arrive at an equation involving $y^2$ .

And suddenly the possibilities are sitting there:

$0, \quad 1, \quad 2$ .

There is the discovery.

No laminated activity cards required.

No sticky notes.

No “turn and talk.”

No colorful poster entitled MY MATHEMATICAL JOURNEY.

Just mathematics.

The equations themselves are doing something interesting.

The mathematics is generating the questions.

The algebra is revealing the structure.

The student is discovering something that has to be true.

That’s not a simulation of mathematical discovery.

That’s mathematical discovery.

⸻

Perhaps this is what bothers me most.

In our eagerness to make mathematics engaging, we sometimes seem embarrassed by mathematics itself.

As though an equation cannot possibly hold a student’s attention without assistance.

An equation needs a story.

A picture.

A game.

A real-world context.

A group activity.

A manipulative.

A digital animation.

A culturally responsive launch.

A collaborative investigation.

A reflection prompt.

And, preferably, an exit ticket.

At some point I expect someone to put

$x^2+y^2=r^2$

into a little paper bag and ask the students to guess what it is before revealing the equation.

But sometimes

$x^2+y^2=r^2$

is enough.

Sometimes the question

“What happens if I subtract these two equations?”

is enough.

And sometimes a student sitting alone with a piece of paper, making a mistake and figuring out why it is wrong, is doing something more mathematical than an entire classroom full of carefully designed activities.

⸻

There is another problem.

We have become very good at describing what students should do.

Students should collaborate.

Students should communicate.

Students should explore.

Students should reason.

Students should model.

Students should represent.

Students should reflect.

But what, exactly, should they know?

That question sometimes feels almost impolite.

“Of course they should know things,” we say.

Yes.

But which things?

Definitions?

Theorems?

Algebraic identities?

Geometric relationships?

Methods of calculation?

Proofs?

Counterexamples?

Special cases?

Theorems and their hypotheses?

Or is “knowing how to communicate mathematical thinking” now sufficient evidence that mathematical thinking has occurred?

I hope not.

A student can communicate beautifully and be beautifully wrong.

A group can collaborate efficiently on an incorrect argument.

A diagram can be gorgeous.

A presentation can be excellent.

A reflection can be profound.

And the mathematics can still be false.

Mathematics has this rather inconvenient feature:

It doesn’t care how good the presentation was.

⸻

Imagine a student proving that

$2+2=5$ .

The student has worked collaboratively.

The group has created three representations.

The students have discussed multiple strategies.

They have explained their reasoning.

They have reflected on their learning.

The teacher has asked excellent questions.

Everyone has demonstrated agency.

The classroom is buzzing with mathematical discourse.

And yet:

$2+3 \ne 5$ .

Mathematics remains stubbornly uncooperative.

It refuses to negotiate.

It does not hold a class meeting and reconsider.

It does not say, “I really appreciate your perspective.”

It simply says:

No.

And that “no” is one of the great beauties of mathematics.

⸻

I don’t want to reform mathematics into something more fashionable.

I don’t want to reform it into a social activity.

I don’t want to reform it into a collection of strategies.

I don’t want to reform it into “problem solving” detached from mathematical knowledge.

And I certainly don’t want to reform it into an endless conversation about how we think about mathematics without actually doing much mathematics.

I want mathematics.

The real thing.

The definitions.

The notation.

The calculations.

The patterns.

The mistakes.

The counterexamples.

The special cases.

The proofs.

The beautiful shortcuts.

The ugly calculations.

The moments when an equation suddenly makes sense.

The moment when two apparently unrelated ideas turn out to be the same idea.

And above all, that extraordinary moment when you discover something that has to be true.

Maybe mathematics education doesn’t need to be constantly reinvented.

Maybe we don’t need another acronym.

Maybe we don’t need another framework.

Maybe we don’t need another seven-step model of mathematical engagement.

Maybe we could simply get better at teaching mathematics.

That would be reform enough.

The Emperor Has No Proof

August 12, 2026August 13, 2026 / Michael Xue / Leave a comment

There is a strange feature of modern mathematics education.

We talk constantly about problem solving, discovery, reasoning, multiple representations, and mathematical thinking.

All good words.

But sometimes I wonder whether we have forgotten the mathematics itself.

I am not opposed to reform. Mathematics education should change when we discover better ways to teach mathematics. But there is a difference between improving mathematics education and replacing mathematics with a collection of educational slogans.

A student who asks “Why?” is doing mathematics.

A student who checks a boundary case is doing mathematics.

A student who refuses to accept a statement until the missing step has been justified is doing mathematics.

And none of this requires a special classroom activity.

Consider a very elementary question.

Two circles are given by $x^2 + y^2 = r_1^2$ and $(x-d)^2 + y^2 = r_2^2$ .

How many points can they have in common?

We are accustomed to saying that two distinct circles can intersect in zero, one, or two points.

Fine.

But why?

Let’s actually do the mathematics.

Assume first that $d > 0$ . Subtract the two equations: $(x-d)^2-x^2 = r_2^2 - r_1^2$ .

Hence $-2dx + d^2 = r_2^2 - r_1^2$ ,

so $x= \frac{r_2^2-r_1^2-d^2}{-2d}$ .

There is only one possible value of $x$ .

Now substitute that value into $x^2 + y^2 =r_1^2$ .

We obtain $y^2 = r_1^2 - x^2$ .

And now the answer is sitting right there.

If the right-hand side is negative, there are no real points.

If it is zero, there is one point.

If it is positive, there are two points.

So: $0, 1$ , or $2$ .

Nothing fancy happened.

We did not need a colorful activity. We did not need to “explore multiple representations.” We did not need a real-world context.

We needed algebra.

And notice something else.

What happens when $d=0$ ?

We don’t simply divide by $d$ and carry on. We go back to the original equations: $x^2 +y^2 = r_1^2$ ,

Subtracting gives $r_1^2 = r_2^2$ .

Thus, if $r_1 \ne r_2$ , there are no common points. If $r_1 = r_2$ , the two equations are identical, and there are infinitely many common points.

So the careless statement

“Two circles intersect in at most two points”

isn’t even true without qualification.

The mathematics tells us exactly where the qualification belongs.

This is what bothers me about some of the rhetoric surrounding math reform.

We have become so concerned with how students experience mathematics that sometimes we seem less concerned with whether students actually know mathematics.

There is a difference.

A student can be encouraged to “discover” something without ever being required to establish it.

A student can produce multiple representations without understanding why the representations are equivalent.

A student can explain a strategy without knowing whether the strategy is valid.

And a student can be praised for mathematical thinking while never being asked to prove the statement in front of them.

At some point, somebody has to ask:

Where is the proof?

That is not an old-fashioned question.

It is the question.

Mathematics has a peculiar feature that makes it different from many other subjects: we don’t get to vote on whether a statement is true.

We can conjecture.

We can experiment.

We can draw pictures.

We can look for patterns.

We can use technology.

All of those things can be useful.

But eventually mathematics asks us to cross a line—from “I think this is true” to “I know why this must be true.”

That transition is not an educational accessory.

It is mathematics.

And perhaps that is the reform I would like to see.

Not a return to mindless memorization.

Not endless lectures.

Not a rejection of calculators, computers, experimentation, or discovery.

Just this:

Put the mathematics back at the center.

Let students ask questions.

Let them struggle.

Let them discover.

But then make them finish the job.

Because when the curtain is pulled back, when the educational vocabulary falls away, when the activity sheet is put aside, there ought to be something underneath it.

There ought to be mathematics.

And if there isn’t—

the emperor has no proof.

Exercise 1. Assumed $d>0$ . Starting from $y^2 = r_1^2 - x^2$ and $x = \frac{d^2+ r_1^2-r_2^2}{2d}$ , show algebraically that the two circles intersect at exactly two points if and only if $|r_1-r_2| < d < r_1+r_2$ .

The Theorem Hidden Inside the Problem

August 12, 2026August 13, 2026 / Michael Xue / Leave a comment

In my previous post, Engineer vs. Mathematician, we started with a concrete question:

Given two circles, do they intersect at exactly two points?

The engineer solves this particular problem.

The mathematician looks deeper and searches for a general condition that answers every problem of this type.

The result is the theorem: $|r_1-r_2| < d < r_1 + r_2$ ,

where $r_1$ and $r_2$ are the radii of the circles, and $d$ is the distance between their centers.

Once we know this condition, any specific problem becomes a simple verification.

There is one exceptional case worth setting aside. If $d=0$ and $r_1=r_2$ , the two circles coincide and therefore have infinitely many common points. Our question concerns circles with distinct centers, so from here on we assume $d > 0$ .

But this raises a deeper question:

How could we discover this condition if we did not already know the theorem?

The answer begins with a simple principle:

Find every way the desired situation can fail. Then what remains must be the answer.

Instead of asking:

When do the circles intersect at exactly two points?

ask:

What conditions make it impossible for the circles to intersect at exactly two points?

We can identify these conditions directly from the geometry.

1. One circle is enclosed inside the other

Suppose first that $r_2 > r_1$

Fig. 1

From Fig. 1, we have

$d+r_1 \le r_2 \implies r_1-r_2 \le -d$ .

For $r_1 > r_2$ (see Fig. 2),

Fig. 2

$d+r_2 \le r_1 \implies r_1-r_2 \ge d$ .

What about $r_1 = r_2$ ?

Since $d>0$ ,

$d + r_1 > r_2$ ,

so one circle cannot enclose the other.

Thus, the only ways one circle can be enclosed inside the other are

$r_1 - r_2 \le -d$

$r_1-r_2 \ge d$ .

Therefore, for one circle not to be enclosed inside the other, we must have

$r_1 - r_2 > -d$

and

$r_1-r_2 <d$ .

Together, these give

$|r_1 - r_2| < d$ .

2. The circles are separated

Fig. 3

Fig. 3 shows that

$d \ge r_1 + r_2$ .

Therefore, for the circles to be able to intersect, we must have

$d< r_1 + r_2$ .

We have now eliminated every configuration in which two intersections are impossible. The boundary cases—where $d=r_1+r_2$ or $d=|r_1-r_2|$ —give tangency and therefore only one intersection. The remaining configurations are precisely those satisfying

$|r_1 - r_2| < d <r_1 +r_2$ .

Thus the condition for two circles to intersect at exactly two points emerges not from applying a theorem, but from eliminating every configuration in which two intersections are impossible.

The familiar approach begins with a theorem.

If the circles intersect at a point P, then the centers and the intersection point form a triangle with side lengths $r_1, r_2, d$ .

The Triangle Inequality tells us that a non-degenerate triangle exists only when $|r_1 - r_2| < d < r_1+r_2$ .

The result follows immediately.

But the theorem itself is not the starting point. It is the destination.

Without knowing the Triangle Inequality, we can still uncover the same truth by examining the structure of the problem:

Too far apart: impossible.
One inside the other: impossible.
The remaining configurations: two intersections.

A theorem is often presented as a finished product: $|r_1 - r_2| < d < r_1 + r_2$ .

But behind every theorem is a path of discovery.

Sometimes we build the object we want and identify the conditions that make it possible.

Sometimes we examine every way it can fail and eliminate those possibilities.

Both paths reveal the same mathematical structure.

The theorem was not merely applied to the problem.

The theorem was hidden inside the problem, waiting to be discovered.

A note for the next post: The argument above assumes that two distinct circles can have at most two intersection points. Why is that true? What prevents two circles from intersecting at three or more points? We will take up that question in the next post.

The Inequality Beneath the Inequality

August 5, 2026August 5, 2026 / Michael Xue / Leave a comment

This post is a revisit of the triangle inequality from a different perspective. In The Triangle Inequality Before the Triangle, we explored the triangle inequality before a triangle exists: given three lengths $a, b, c$ , what conditions allow them to form a triangle?

Here, we start with an actual triangle and uncover the algebra beneath the familiar inequality.

Consider an arbitrary triangle $\Delta ABC$ .

Without loss of generality, place

$A=(0,0),\qquad B=(c,0)$ ,

where

$c=AB>0$ .

Let the third vertex be

$C=(x,y)$ ,

where

$y>0$ .

Then the side lengths are

$a=BC=\sqrt{(x-c)^2+y^2}$ ,

and

$b=AC=\sqrt{x^2+y^2}$ .

We prove that

$a+b>c$ .

Since

$y>0$ ,

we have

$x^2+y^2>x^2$ ,

and therefore

$b=\sqrt{x^2+y^2}>|x|$ .

Similarly,

$(x-c)^2+y^2>(x-c)^2$ ,

$a=\sqrt{(x-c)^2+y^2}>|x-c|$ .

Therefore,

$a+b>|x|+|x-c|$ .

Now,

$|x-c|=|c-x|$ ,

$|x|+|x-c|=|x|+|c-x|$ .

At this point, the geometry disappears. The remaining argument is purely algebraic.

We now use the algebraic inequality for absolute values:

$|u|+|v|\geq |u+v|$ .

This is an inequality about real numbers. It does not involve triangles, distances, or geometry.

Applying it with

$u=x,\qquad v=c-x$ ,

we obtain

$|x|+|c-x| \geq |x+(c-x)|$ .

Simplifying,

$x+(c-x)=c$ ,

$|x+(c-x)|=|c|$ .

Because

$c>0$ ,

we have

$|c|=c$ .

Hence,

$|x|+|c-x|\geq c$ .

Therefore,

$a+b>|x|+|c-x|\geq c$ ,

and thus

$a+b>c$ .

$\blacksquare$

The familiar triangle inequality $a+b>c$ is not an isolated geometric fact.

Beneath it lies a simpler algebraic inequality:

$|u| + |v| \ge |u+v|$ .

The algebraic inequality lives on the number line, while the triangle inequality lives in the plane.

The geometry supplies the distances, but the algebra supplies the comparison.

A statement about triangles in the plane is revealed to be an application of an inequality about numbers on the real line.

Exercise-1 The proof relies on the hidden inequality

$|u| + |v| \ge |u+v|$ .

Prove this hidden inequality.

The Triangle After the Triangle Inequality

August 4, 2026August 5, 2026 / Michael Xue / Leave a comment

Theorem (Converse of the Triangle Inequality)

Given $a, b, c > 0$ , if

$a+b > c, \quad a+c >b, \quad b+c >a$

then there exists a triangle whose side lengths are $a, b$ and $c$ .

Proof

Place

$A = (0, 0), \quad \quad B = (c, 0)$ .

We seek a point

$P = (x, y)$

such that

$AP = b, \quad\quad BP = a$ .

These conditions are equivalent to

$x^2+y^2 = b^2$ ,

and

$(x-c)^2 + y^2 = a^2$ .

Subtracting the two equations gives

$x = \frac{b^2+c^2-a^2}{2c}$ .

Substituting this into the first equation yields

$y^2 = b^2 -\left(\frac{b^2+c^2-a^2}{2c}\right)^2$ .

Thus the existence of the desired point $P$ is equivalent to the existence of a real solution of the quadratic equation

$Y^2 - \left[b^2 -(\frac{b^2+c^2-a^2}{2c})^2\right] = 0$ .

Its discriminant is

$\Delta = 4\cdot\left[b^2-(\frac{b^2+c^2-a^2}{2c})^2\right]$ .

Multiplying the expression for $\Delta$ by $c^2$ ,

$c^2\Delta = 4b^2c^2 - (b^2+c^2-a^2)^2$ .

A remarkable factorization gives

$c^2\Delta = (a+b+c)(a+b-c)(a+c-b)(b+c-a)$ .

Since $a,b,c>0$ , we have $a + b + c >0$ . By hypothesis,

$a+b-c >0, \quad a+c-b > 0, \quad b+c-a >0$ .

Hence every factor on the right-hand side is positive, so

$\Delta > 0$ .

Therefore the quadratic equation has two distinct real solutions for $Y$ . Hence there exists a point $P=(x, y)$ whose distances from $A$ and $B$ are $b$ and $a$ , respectively.

$AP = b, \quad BP = a$ .

Consequently,

$AB = c$ ,

so there exists a triangle with side lengths $a,b$ and $c$ .

$\blacksquare$

The proof actually gives the triangle explicitly.

Since the discriminant is positive, the two solutions are

$y = \pm\sqrt{b^2-(\frac{b^2+c^2-a^2}{2c})^2}$ ,

and therefore the two possible vertices are

$P = \left(\frac{b^2+c^2-a^2}{2c}, \pm\sqrt{b^2-(\frac{b^2+c^2-a^2}{2c})^2}\right)$ .

These are reflections of each other across the $x$ -axis, producing the two possible triangles with side lengths $a, b$ and $c$ .

Exercise-1 Verify by hand that $4b^2c^2-(b^2+c^2-a^2)^2$ factors into $(a+b+c)(a+b-c)(a+c-b)(b+c-a)$ using the difference of square twice.

The Triangle Inequality Before the Triangle

August 4, 2026August 4, 2026 / Michael Xue / 1 Comment

One of the most familiar statements in geometry is the triangle inequality:

The sum of any two sides of a triangle must be greater than the third side.

Most of us first encounter this as a simple test.

Can sides of lengths (3,4,5) form a triangle? Yes.

Can sides of lengths (2,3,6) form a triangle? No.

The triangle inequality becomes a rule to memorize before moving on to the “real” mathematics.

But this misses something deeper.

The triangle inequality is not merely a property of triangles.

It is the condition that determines whether a triangle can exist at all.

Before we discuss angles, areas, similarity, or congruence, there must first be a triangle.

The triangle inequality is the mathematics before the triangle.

Suppose a triangle has side lengths $a, b, c$ .

Let

$s=\frac{a+b+c}{2}$

be the semiperimeter.

Heron’s formula gives the area:

$A = \sqrt{s(s-a)(s-b)(s-c)}$ .

Squaring both sides,

$A^2 = s(s-a)(s-b)(s-c)$ .

Because a genuine triangle has positive area,

$A^2 > 0$ .

Now substitute

$s = \frac{a+b+c}{2}$ .

The formula becomes

$A^2 = \frac{a+b+c}{2}\cdot\frac{b+c-a}{2}\cdot\frac{a+c-b}{2}\cdot\frac{a+b-c}{2}$ .

Therefore,

$16A^2 = (a+b+c)(a+b-c)(a+c-b)(b+c-a)$ .

Since

$a+b+c > 0$ ,

we must have

$(a+b-c)(a+c-b)(b+c-a)>0.\quad\quad\quad(*)$

The triangle inequalities are hidden inside this product.

Group the first two factors together. Since the product $[(a+b-c)(a+c-b)](b+c-a)$ is positive, $(a+b-c)(a+c-b)$ and $(b+c-a)$ must have the same sign. Thus there are two possibilities:

Case 1.

$\begin{cases} (a+c-b)(a+b-c)>0 \\ b+c -a>0 \end {cases}$

Case 2.

$\begin{cases} (a+b-c)(a+c-b) < 0 \\b+c-a<0 \end {cases}$

Examining these cases, we have

$\begin{cases} a+b-c > 0\qquad(1-1)\\ a+c-b>0\qquad(1-2) \\ b+c-a >0\qquad(1-3) \end{cases} \implies \begin{cases} a+b>c \\ a + c >b, \\ b+c>a \end{cases}$ precisely the triangle inequalities.

$\begin{cases} a+b-c<0 \qquad(2-1)\\ a+c-b <0\qquad(2-2)\\ b+c-a>0\qquad(2-3)\end{cases} \overset{(2-1) + (2-2)}{\implies} a < 0$ , contradicting $a>0$ .

$\begin{cases} a+b-c <0 \qquad (3-1)\\ a+c-b >0\qquad(3-2)\\ b+c-a<0\qquad(3-3)\end{cases} \overset{(3-1)+(3-3)}{\implies} b<0$ , contradicting $b>0$ .

$\begin{cases} a+b-c >0 \qquad (4-1)\\ a+c-b<0\qquad(4-2) \\ b+c-a <0\qquad(4-3) \end{cases}\overset{(4-2) + (4-3)}{\implies} c<0$ , contradicting $c>0$ .

Therefore, the only possible case is

$a+b-c>0, \quad a+c-b > 0, \quad b+c-a>0$ .

Hence,

$a+b >c, \quad a+c>b, \quad b+c>a$ .

For an existing triangle, the expression in Heron’s formula must be real and positive.

It is often presented as a statement about triangles.

But it is really a statement about existence.

Heron’s formula does more than calculate the area of a triangle. Its algebraic structure contains the conditions that determine whether three lengths can form a triangle.

The factors

$a+b-c, \quad a+c-b, \quad b+c-a$

are the triangle inequalities waiting to be discovered.

Before Heron’s formula measures a triangle, it first asks whether there is a triangle to measure.

The same idea appears throughout mathematics.

Before solving an equation, we ask whether solutions exist.

Before applying a formula, we ask whether its assumptions are satisfied.

Before studying an object, we ask whether the object is possible.

The triangle inequality is a simple example of a profound mathematical habit:

Before we calculate, we must first know what is allowed to exist.

Exercise-1 Prove:

Theorem (Triangle Inequality Converse).

Given $a,b,c > 0$ , if

$a+b>c, a+c > b, b+c> a$

then there exists a triangle whose side lengths are $a, b, c$ .

Engineer vs. Mathematician

August 1, 2026August 3, 2026 / Michael Xue / 1 Comment

Show that the circles with centers at $O_1(-1, 1)$ and $O_2(3,2)$ , and respective radii $r_1 = 3$ and $r_2 = 2$ , intersect at exactly two points.

Engineer:

Mathematician:

Theorem 1. Two circles $C_1$ and $C_2$ intersect at exactly two points if and only if

$|r_1 - r_2| < d < r_1 + r_2$ ,

where $d$ is the distance between their centers.

Proof

Suppose $C_1$ and $C_2$ intersect at exactly two points, $P_1$ and $P_2$ (that is, $\#(C_1\cap C_2)=2$ ). We have two triangles $O_1P_1O_2$ and $O_2P_2O_1$ .

Applying the triangle inequalities to triangle $O_1P_1O_2$ gives

$r_1 + r_2 > d\quad\quad(1)$

$r_1+ d > r_2 \implies r_1-r_2 >-d\quad\quad (2)$

$r_2 + d > r_1 \implies r_1 - r_2 < d.\quad\quad (3)$

Combining $(2)$ and $(3)$ gives

$|r_1 - r_2| <d.$

Therefore,

$\#(C_1 \cap C_2)=2 \implies |r_1-r_2| < d < r_1 + r_2$ .

Assume

$|r_1-r_2| < d < r_1 + r_2$ .

Then we have

$d < r_1 + r_2$ ,

$r_1 - r_2 <d \implies r_1 < d +r_2$ ,

$r_1-r_2 > -d \implies r_2 < r_1 + d$ .

Thus the three lengths

$r_1, r_2, d$

satisfy the triangle inequalities, so a triangle with these side lengths exists.

Construct a triangle $O_1PO_2$ having

$|O_1P| = r_1, \quad |O_2P| = r_2$ .

Reflect $P$ across the line $O_1O_2$ to obtain another point $P'$ .

Since $d < r_1 + r_2$ and $|r_1 - r_2| <d$ , the triangle is nondegenerate, so $P$ does not lie on the line $O_1O_2$ . Hence its reflection $P'$ is distinct from $P$ .

Both $P$ and $P'$ satisfy

$|O_1P| = |O_1P'| = r_1, \quad |O_2P| = |O_2P'| =r_2$ ,

so both lie on both circles.

Hence

$C_1 \cap C_2 = \{P, P'\} \implies \#(C_1 \cap C_2) = 2$ .

We have

$|r_1-r_2| = |3-2| =1, d = \sqrt{(-1-3)^2+(1-2)^2} = \sqrt{17}$ and $r_1+r_2 = 3+2 = 5 = \sqrt{25}$ .

Since

$|r_1 - r_2| < d <r_1+r_2$ ,

By Theorem 1,

The circles intersect at exactly two points.

The engineer solved one problem.

The mathematician solved an entire class of problems.

Exercise-1 Explain the Triangle Inequalities.

The Mathematics Before the Mathematics

July 30, 2026July 30, 2026 / Michael Xue / Leave a comment

Mathematics is often presented through its visible actions.

We apply formulas.

We manipulate symbols.

We solve equations.

We prove statements.

These are the outward expressions of mathematics—the steps written on the page, the procedures that produce answers.

Yet the most important mathematics often begins before any of these actions.

Before we apply a formula, we must know when that formula is valid.

Before we manipulate symbols, we must understand what those symbols represent.

Before we solve, we must know what we are solving.

Before we prove, we must know what the statement means.

Mathematics has a front door.

That door is built from definitions, assumptions, and conditions. They determine what mathematical objects we are working with and what operations are valid.

Many mathematical mistakes happen because we skip the entrance and begin working inside a mathematical world that we never properly entered.

Here are three examples.

Consider two equations:

$C_1 = 0, \quad C_2 = 0$ .

When these equations represent circles, we can form the linear combination

$C_1 + k \cdot C_2 = 0, \quad k \neq -1$ .

Because both circles contain the same quadratic terms, this often produces another member of the same family of circles.

But there is a front door.

Before using this technique, we must first confirm that $C_1$ and $C_2$ actually represent real circles.

Consider

$C_1 = x^2 + y^2 + 4x +4y +20 =0$ .

Completing the square:

$(x+2)^2 +(y+2)^2 = -12$ .

This equation has no real solutions. It does not represent a real circle.

Now consider

$C_2 = x^2+y^2+4x+4y+9=0$ .

Completing the square:

$(x+2)^2+(y+2)^2=-1$ .

This is also not a real circle.

Now form the combination:

$C_1 + (-2) \cdot C_2 = 0$ .

That is, $k = -2$ , which satisfies $k \neq -1$ .

Compute:

$(x^2+y^2+4x+4y+20)-2(x^2+y^2+4x+4y+9)=0$

Simplifying:

$-x^2-y^2 - 4x -4y+2=0$ .

Multiply by -1:

$x^2+y^2+4x+4y-2= 0$ .

Completing the square:

$(x+2)^2+(y+2)^2=6$ .

A genuine real circle has appeared.

The algebra was completely correct.

But the geometric interpretation would have been wrong if we had assumed that $C_1$ and $C_2$ were circles without checking.

The lesson is not that linear combinations are dangerous. It is that algebraic manipulation and mathematical meaning are not the same thing.

Symbols do not automatically carry the interpretation we want to give them.

We often write:

$\frac{x^2-1}{x-1} = x+1$ .

because

$x^2-1 = (x-1)(x+1)$ .

But the cancellation step requires:

$x - 1 \neq 0$ .

So the statement is not true for all $x$ .

The correct statement is:

$\frac{x^2-1}{x-1} = x+1, \quad x \neq 1$ .

At $x = 1$ , the original expression is undefined, even though the simplified expression gives $2$ .

Mathematical induction provides perhaps the clearest example of this idea.

To prove a statement $P(n)$ for all positive integers, we need two ingredients.

First, we prove the base case:

$P(1)$ .

Second, we prove the induction step:

$P(n) \implies P(n+1)$ .

The induction step tells us that truth can move forward.

It does not tell us that truth exists anywhere to begin with.

Imagine a row of dominoes. The induction step says:

If one domino falls, the next domino will fall.

But someone still has to push the first domino.

Consider the statement:

$P(n):\quad n \ge 5$ .

The induction step is true:

$n \ge 5 \implies n+1 \ge 5$ .

The implication works perfectly.

But without checking the base case, we might incorrectly conclude that the statement is true for all positive integers.

That would give the absurd result:

$1 \ge 5$ .

The induction process did not fail.

We failed to start it.

The base case is the entrance point that allows the entire chain of reasoning to begin.

Once we start looking, front doors appear everywhere in mathematics.

Before dividing by an expression, check that it is not zero.

Before taking a square root over the real numbers, check that the quantity is nonnegative.

Before applying differentiation rules, verify that the required derivatives exist.

Before applying the Intermediate Value Theorem, verify continuity.

Before using a formula, understand the assumptions behind it.

These checks are not obstacles placed in front of mathematics.

They are the mathematics.

Every front door leads somewhere.

Before we step through it, we should know what lies on the other side.

The first question in mathematics is not:

What should I do next?

It is:

What am I actually working with?

Are these really circles?

Does this element really exist?

Has the induction process really begun?

Can this theorem actually be applied?

These questions come before every calculation and every proof.

Mathematics is not merely the manipulation of symbols. It is the study of objects with meaning, and that meaning comes from definitions and assumptions.

The pattern is universal: every part of mathematics has a front door.

A good mathematician does not rush through the nearest opening.

He identifies the correct entrance, checks the conditions, and only then proceeds.