I think software is unique in that there isn't a neat "research" vs. "engineering" side, at least in so far as a lot of boundary pushing in the field has happen

I think software is unique in that there isn't a neat "research" vs. "engineering" side, at least in so far as a lot of boundary pushing in the field has happened by engineers actually building software. I think there will always be a frontier where that is the case in software—there will always be hackers building open source projects that push some boundary, fostering their own community, and potentially changing broader paradigms—but that more and more, we're seeing a neater boundary between research (Universities, labs, projects that are essentially R&D teams at big companies) and engineering.