Note that the paper's title is somewhat misleading: it's not a newbie's guide to how mathematicians write proofs, but a proposal for a new method of writing proofs that he thinks mathematicians and computer scientists should use. An interesting proposal, though.