As a way of getting past the journal article block I mentioned a few posts ago, I'm revisiting some of the papers that I've cited in most of my own papers, sort of as a way of reminding myself what differentiates my current work from others' work in the same space. (Along with finding new references for things that probably should have references, etc....Suffice it to say that the past few days I've been scouring the web and various library databases and doing lots of reading. Fun stuff when you have the time to do it!) This careful rereading (OK, reskimming) of these papers has led me to pick up on stuff I missed the first (or second) time reading (OK, skimming), which is leading me to new references that I missed before.
In the course of looking up some of these new-to-me references, I've made a startling discovery: Some of the most interesting related work has been done in an entirely different subfield, one that I hadn't thought of looking at before for this particular work. The *very* interesting part is that this particular subfield is one that I've had a growing interest in, and have started taking baby steps to follow, over the past year or so!
This whole experience shows how quickly we can develop "tunnel vision", especially in how we scope and classify our work as computer scientists. I've realized for a while that my work straddles several areas, but I've sort of narrowly defined what those areas are, based on my own background and training, not even dreaming that there might be other, and perhaps better-suited, connections too. And the work I'm finding has been around for a while. Part of me wonders: if I had known about the connection of my work to this other subfield before, would that have changed the scope or flavor of my research at all?