For example, they apparently have 23 U-Matic tapes of the HOPL (History of Programming Languages) conference in 1978 [1].

The proceedings [2] describe one of the tapes that don't make it into the transcribed proceedings:

> The after-dinner speeches at the conference banquet were devoted to humorous reminiscences and anecdotes about the languages and events during their development. The banquet anecdotes are not included in this volume because, although they are humorous to hear, the voice inflections make a big difference, and they are not necessarily amusing to read.

[1] https://www.computerhistory.org/collections/catalog/10269512... [2] https://dl.acm.org/doi/book/10.1145/800025

The intention is to specialize away interpretive and meta-programming-like overhead, and then translate the result into usable source code in whichever language is desired. If the code, after specialization, has a suitable form, then you should get conventional imperative code out. If there are still functional-idioms present that haven't been specialized away, the generated code will have a more functional-style. The specialator currently targets JS and C.

The language has an expressive type-system based on self-dependent sub-typing. The type-system is too expressive for everyday use, it is intended to be used with a decidability checker.

This is an experimental work-in-progress.

  λ-Calculus, Combinatory Logic and Cartesian Closed Categories
  https://thma.github.io/posts/2021-04-04-Lambda-Calculus-Combinatory-Logic-and-Cartesian-Closed-Categories.html

  Categorical combinators, P.-L. Curien
  https://www.sciencedirect.com/science/article/pii/S001999588680047X