Re: Age of Continuations?

[Dave Long <address@hidden>]

>                  ... looks like the most solid original date for 
> continuations is 9/64, and for them as concepts in use to the 
> 1969-1972 timeframe, which is good enough for my purposes. 

>From the sound of Landin's "Histories of Discoveries
of Continuations: Belles-Lettres with Equivocal Tenses"
<http://www.dcs.qmul.ac.uk/~peterl/danvy/danvy.html>
it seems like mid-60's is about right, but the concepts
had been in use as early as the first closed subroutines.

> So reductions were afoot in opposite directions. Goto's were being
> expressed as calls (without return) [9]. Calls (with return) were
> being expressed as goto's ...
>
> Concerning this there was a certain nostalgia among old programmers
> of stackless architecture who'd lived through the anguished cries
> about early fortran's failure to provide ``re-enterability'' ...
> Shunting arguments, including return links, around had been a
> familiar chore.
> ...
> The control-address-as-data-item became the stack-frame-as-data-item
> became the program-closure-as-data-item. 

-Dave