[jlees]

Indeed - giving talks, supervising research students, etc.

Having said that, I worked with a professor who had effectively taken the same route; his work was relevant in the eighties and hardly used since. My job was to try and reimplement his algorithms (ALGOL 68) in Java, which involved talking to him a lot to understand the work -- it felt like he was genuinely happy to have someone to talk to about it, and various departmental shifts and changes over the years had gradually disconnected him from the world, rather than his own decision to start gathering dust.