Re:Alan Kay has agreed to do an AMA today
by ducklord on Monday June 20, 2016 @03:01PM
(#11940402)
Hey Alan, you once said that lisp is the greatest single programming language ever designed. Recently, with all the emergence of statically typed languages like Haskell and Scala, has that changed? Why do you think after being around for so long, lisp isn't as popular as mainstream languages like Java, C or Python? And lastly, what are your thoughts on MIT's switch to use Python instead of Scheme to teach their undergraduate CS program?
[ Parent
| on HN ]
I should clarify this. I didn't exactly mean as a language to program in, but as (a) a "building material" and (b) especially as an "artifact to think with". Once you grok it, most issues in programming languages (including today) are much more thinkable (and criticizable).
The second question requires too long an answer for this forum.
Would it be better on a ML ? I remember long topics on FONC.
The one thing that makes LISP great is the "functions are data" formulation.
Many of the "current best paradigms" of Computer Science are actually fads. LISP was a fad of the 1980s. Java was a fad of the 1990s. NoSQL databases are a current fad. It doesn't mean that one is better than another. It's human nature to think that new technology must be better than old technology. In fact, the programming environment on the LISP machines of the 1980s was far better than anything we had till the early 2000s, despite being "old".
I think you are missing what was/is special about Lisp. It's "the idea of Lisp" (and see other places in this AMA where I've said a few more words).
The "idea of Java" is not special (and it's not clear that you could even say there is an "idea of Java").