Coding is traditionally used at the physical layer in communication networks to recover from errors and erasures incurred in transmission or storage. It is easy to account for the cost of coding (e.g., in bandwidth and energy) and argue for its use. These days, novel schemes, that transcend the traditional role and place of coding, are being proposed and considered for implementation in commercial products and systems. Such schemes (e.g., rateless and network coding) impact not only energy and bandwidth, but also network traffic and protocols. They make invalid various independence assumptions and complicate addressing and security schemes in content networking. We discuss coding in several transmission/storage network scenarios and question/explain its potential benefits