What I don't like about monero, and why I think it'll ultimately lose to another anon product, is that the transaction history is written to the blockchain, albeit in obfuscated form. But there is no proof on the bounds of what a sophisticated blockchain analysis can uncover given enough information. Roughly speaking, its conceivable that given enough transaction information downstream from a transaction of interest might reveal a "most likely" pairing of an address with a transaction. Or perhaps a global analysis of the monero blockchain along with traditional blockchain analysis could reveal a most likely pairing of all transactions with addresses. The point is that without such a proof I'm not sure how much people will trust it compared to a protocol with such a proof (say zcash). Your exposure to future analysis is essentially indefinite.