A mathematical proof should resemble a simple and clear-cut constellation, not a scattered cluster in the Milky Way.
A Mathematician's Apology, Section 18 (p. 113), Cambridge University Press. 1967
