It should be noted that Burroughs (now Unisys) WFL is a high-level, block structured language that is not interpreted, but compiled to machine object code. The object code is a one-shot for that job run, however, and cannot be reused, although the compiler is one-pass and quite fast. It can declare, open, and close files, and modify their filesystem attributes, but not read or write their data. In addition to step-by-step sequential execution, it has facilities for initiating tasks in parallel, monitoring their state, and recognizing when they complete.
Not trying to take any William S. Hawes glory away, and not doubting your timeline, but from the above it does not follow that âRexx on Amiga came out when OS/2 2.0 came outâ.
All that is claimed is:
(a) IBM gave license to Commodore
(b) IBM got GUI tech from Commodore
Given your timeline, ARexx happened before (a). And the OS/2 which included Commodore GUI tech must have happened after (b). But thatâs it, no other dependencies.
I feel we shouldnât argue too much over this one - thereâs little information to go on.
But one possible picture, given what Iâve read, is that Hawes could have written his implementation quite independently, and well before any deal with IBM, and that IBM could have taken the position that a license was needed and a deal was needed. IBM could surely out-spend Commodore on lawyers, so Commodore could have done a deal regardless of their beliefs about the need for a license.
Exactly my thinking actually. IBM lawyers flit in âsooo we heard you are selling something called Rexx here?â
(Apologies if I sounded like arguing, that was not my intention, just pointing out that from the data points there is a quite logical ordering of events.)
1 Like
Precisely my point. There is no source â apart from this one article about OS/2 â that provides proof of an IBM licence for Rexx on Amiga. So we probably shouldnât consider it a thing.