Put the nodes 1, 2, 3 in rows and the powers x⁰, x¹, x² along the columns: (1 1 1; 1 2 4; 1 3 9). The first link's panel returns determinant 2, and the second link is the formula that explains it, (2 − 1)(3 − 1)(3 − 2) = 2. The third link is the four-node version on 1, 2, 3, 4 with determinant 12, and the fourth is the six differences that multiply to it.
The product form is why the determinant vanishes exactly when two nodes coincide, and therefore why a polynomial of degree at most n − 1 through n distinct nodes exists and is unique: the coefficient matrix is Vandermonde and its determinant is a product of non-zero differences. It also shows why naive interpolation is badly conditioned — at nodes 1, 1.001 and 1.002 the product is (0.001)(0.002)(0.001) = 2⋅10⁻⁹, so the matrix is very nearly singular, which is the reason practical interpolation uses Chebyshev nodes or orthogonal polynomials rather than solving the linear system. Vandermonde's 1771 memoir was about eliminating unknowns, not about matrices, which were not yet objects; the determinant and the product formula are the nineteenth century's reading of what he had written down.
Answer
2
How to read it
The 3×3 determinant at the nodes 1, 2, 3. The second link is the product of differences; the third and fourth are the four-node matrix and its six differences, giving 12.
Source
Alexandre-Théophile Vandermonde, 'Mémoire sur la résolution des équations', Histoire de l'Académie royale des sciences (1771); the product formula ∏(xⱼ − xᵢ) is the standard statement, and it is the proof that interpolation at distinct nodes is unique
Work it in the calculator
These links open the right calculator with this problem's numbers already typed in. The line above is what the screen should show — check it yourself.