i think david yetter's analysis of the dichotomy "categories as foundations" vs "categories as algebra" was spot on --- with respect to people and the community. indeed, one could split most of our papers into one category or the other. but at the end of the day, i think, we'll all agree that the source of the unreasonable effectiveness of categorical algebra is its foundational content (although there is probably a lot of it that we dont understand yet); and the other way around. eg, if you look at grothendieck's work, he started working in algebra, and ended up developing foundational structures, because he needed them. and a lot on the "algebra" side now is built upon them. ok, then for a while it was thought that he exaggerated with foundations, and that a more direct approach "could have been in better taste" (to cite eilenberg). but maby the fermat theorem would have a more useful proof if it was developed in grothendieck style. and nowadays, there is a lot of foundational content in tannaka duality etc, in TQFT in general, but we only see hints of it at the moment (and i for one just see the reflections of these hints in other people's eyes). i am of course saying things very clear and familiar to many people on this list, but maby they are worth saying nevertheless. it might be good if the links between "categories as algebras" and "categories as foundations" would not boil down just to the greatest of the category theorists, leaving the rest of us in two camps. -- dusko