All, i recently ran across a use of fold (in the functional programming sense) where i wanted to solve for the accumulator type. i realized that something in the neighborhood of a combinatory logic would be a minimal solution. This got me thinking... there are various versions of combinators for the \lambda-calculus and there's Haghverdi's combinators for linear lambda and there are Yoshida's combinators for Milner's \pi-calculus. The first two will work in the setting i encountered, the latter will not because -- in addition to (parallel) composition -- Yoshida's combinators require a new and replication operator. So, i'm wondering if anyone has given an abstract characterization of a combinatory logic/algebra? If so, does anyone have a reference? Best wishes, --greg -- L.G. Meredith Managing Partner Biosimilarity LLC 806 55th St NE Seattle, WA 98105 +1 206.650.3740 http://biosimilarity.blogspot.com