4 ms·
The predecessor to GOAL was GOOL and was 100% Andy's work. We used it to code the logic for the objects in the Crash games (critters, platforms, etc.) and I act
by dmbaggett 5y ago
The predecessor to GOAL was GOOL and was 100% Andy's work. We used it to code the logic for the objects in the Crash games (critters, platforms, etc.) and I actually wrote the entire load/save UI in it as well! I assume GOAL was also all Andy but I don't know that for sure because I only worked on the Crash games.
ITA's use of Lisp was driven by our co-founder and Chief Scientist, Carl de Marcken. He wrote the prototype in LispWorks, then the production code in Allegro, and then we finally ported everything to CMUCL when Franz essentially tried to extort us for a tax on our entire business. (We couldn't imagine this would become the norm for consumer "apps" five years later, but that's another topic entirely.)
Ultimately the MIT AI Lab circa 1993 was the source of the decisions to use lisp for both these commercial products. At the time, the lab's culture was still very much influenced by all the work done there and in spin-off companies in the late 80s commercializing lisp, e.g., at Symbolics (first registrant of a .com domain name!)
At INKY (my current startup) we use Python. By 2010, when I started the incubator that evolved into INKY, it was pretty clear that scaling a development team on a lisp stack was going to be really difficult.
- bmitc 5y agoThank you for the history and background! It's great to get it from a first-party source. > Ultimately the MIT AI Lab circa 1993 was the source of the decisions to use lisp for both these commercial products. I suppose another example is iRobot, where there's a custom Lisp-like language used, at least in the earlier robots. My understanding is that it's not used as much anymore if at all. > ITA's use of Lisp was driven by our co-founder and Chief Scientist, Carl de Marcken. I chuckled at his website that states "I retired long ago and almost certainly do not find whatever commercial project you want help with interesting". > At INKY (my current startup) we use Python. By 2010, when I started the incubator that evolved into INKY, it was pretty clear that scaling a development team on a lisp stack was going to be really difficult. If you were to start it today, with the developments of languages like F#, Elixir, and Clojure, do you think you'd have a different decision or outcome? I'm just curious, as a hypothetical.