A recent article by Neal Koblitz in the Notices of the AMS, titled "The Uneasy Relationship Between Mathematics and Cryptography," reminds me of the Leo Breiman paper "Statistical Modeling: The Two Cultures."
Both papers talk about the cultural and practical differences between various academic departments and how they do research. In the Breiman paper, he noted how progress was a bit limited in the field of statistical modeling by "classical" statisticians until computer scientists came along and opened up the field with new approaches, and corresponding new results.
The Koblitz paper discusses the field of cryptography, and describes how mathematicians and computer scientist cryptographers at first worked well together in coming up with various crypto-systems, but how now that relationship is no longer as solid. While the crypto-systems published by the computer scientist cryptographers are often more practical (usable), they also sometimes rush to publish papers with dramatic errors that might have been caught, say, by a more rigorous approach by a mathematician, such as Koblitz.
Technorati Tags: academia, security, technology