[the_watcher]

I remember reading one of PG's essays about Lisp being the most powerful language, and he made a list of characteristics defending his position. He implied that the new languages that are becoming popular simply move closer to being a Lisp. My guess is that Clojure is the Lisp that emerges most directly from someone who learned about languages from Ruby. I've only ever hacked with Ruby, if I can find a good intro series for non-coders to Clojure, I'm going to dive into it.

[hibikir]

Lisp has a lot of strengths, don't get me wrong. But it is an ancient language, and despite its age, it never managed to catch on in a big way. It is true that many more popular languages keep stealing Lisp concepts, but that doesn't tell me that Lisp is amazing, but that despite all of its strengths, it must have some terrible weakness that makes it relatively uncompetitive in the general programming population.

My opinion is that it crosses a syntactic threshold of abstraction, after which a language just loses. s-expressions are extremely powerful, but also the language's Achilles heel.

This resembles how there is a 'happy medium' for writing human languages. We could write all of our text in morse code, but we don't. We know that too many glyphs, like in Chinese, slow down learning too, as the effort to remember them all takes lots of practice.

So just like Scalaz's operators are so arcane that us mere mortals are better served by using words instead of <=o=>, a world of s-expressions makes it harder to find your way precisely because of excessive syntax homogeneity.

Still, I think Lisps are worth learning, it's just that I think the ideal language steals much from Lisp, including most of what s expressions are used for in practice, but it doesn't go all the way.