On: Computer scientists prove that a 40-year-old algorithm is optimal

[anacleto]

I love 'proofs' which rely on an unproven component.