Thank for sharing that paper. I am currently implementing a compiler that has very similar ideas to those presented in it. It's very different syntactically, but the mental model is quite similar. It's simultaneously encouraging and discouraging to see your ideas reflected back to you from 30 years in the past :)

Also thanks for sharing; looks interesting. I’ll read this when I have time and re-comment then. Off the bat: Forsp variables are definitely not linear.

what is cvbp? google yields nothing.

It's a typo for CBPV, Call-By-Push-Value.

I think that was meant to be CBPV (Call-By-Push-Value)

Looks complicated to reason about, but really interesting. Freeforth implements a similar mechanism to eliminate swaps via register renaming, obv. the focus is different in that Freeforth tries to still be an optimised stack, rather than a VLIW processor. I guess you could look at this paper as a vector processor, similar to the way APLs handle things with function composition. In that, you'd need the compiler to take care of things as managing it manually would be a nightmare. I guess the best way forward would be a specialized (forth) word lib which invokes the compiler to manage the top N stack cells for the duration of the word.