brain-dump TLDR: I like clojure's costs-to-benefits.. and it's successful because it's what the industry needs.
I wouldn't say lisp family langs are the 'best' languages, but I would say its tradeoffs maximize the power of the individual, which interests me since I am one :-). Talking about tradeoffs can get very verbose, but I find it helpful to make an analogy for things like mutability/immutability as a number line. Let's think of some tradeoffs and virtues, it's not always black/white or bad/good: regularity/expressiveness type-safety/dynamicity mutability/immutability speed code size (language implementation or generated) 'weight of an abstraction' (ie how hard/big is it to jump a level of expressiveness) orthogonality of feature-set (pay for what you use) popularity blub-ability (do you need to be an expert to program in it?) team-suitability Then we can start to think of it as quantities, a vector. We can subjectively normalize these for orthogonality, eg.. for some languages, some tradeoffs will interact more with others in practice, and we can come up with a cost-model according to our usage. Then programming in this language can be thought of as a projection (dot-product) of our minds along this vector space over time. These tradeoffs affect 'costs', ie time/money/effort. There is a fixed cost to learning clojure well enough (IME the time it takes to grok Joy of Clojure, implementation details like core.clj and Compiler.java, learning enough emacs to get by), and little recurring costs along the way (like how hard it is to refactor something). In my opinion, I love clojure because of the orthogonality of its feature-set (design by decoupling) and because of the expressivity of lisp itself, that outweighs everything else. Its code has a high SNR because of it. I think s-expressions are the closest thing to how brains actually work, ie it's an efficient way to 'project' at any level of abstraction, and the cost of traversing abstractions up and down is low due to the regularity of the syntax. Scheme makes a good teaching language because there are benefits to seeing the structure directly. We know from experience that restricting abstraction is bad (java). So, I'm pretty satisfied with it, I might learn other languages for specific use-cases, but I think time spent in lisp isn't wasted, since it's easy to pull an abstraction out of anything repetitive, and we as users can steal ideas from other languages more effectively without involvement from compiler writers. A new language becomes necessary when we have to revisit core primitives, which we will be compelled to do once we identify certain costs are too high for too many people. People might have different ideals, use-cases or costs based on their experience or situation.' I think the curation approach of 'design by decoupling' turned out to be a great idea simply because of the huge amount of stuff that's been written by many people in isolation over the last decades. If we took it to an extreme, then it would also cease to be a good idea. I think C++/Java-style OO has already been taken as far as it needs to go. We've reached the point where the costs-to-benefits are well-understood and felt to be lacking for many current and future use-cases. On Fri, Dec 27, 2013 at 8:24 AM, Massimiliano Tomassoli <kiuhn...@gmail.com>wrote: > On Friday, December 27, 2013 7:31:25 PM UTC+1, Sean Corfield wrote: >> >> I'm not sure why you are less trusting of their >> real world experiences than what your professors told you about >> (theoretical) OOP... >> > > First, someone said that OOP doesn't alleviate the problems that it's > supposed to solve and this I don't believe. > Then, someone else claimed that OOP is useless in presence of > immutability. I don't agree with that. It would also imply that Scala's > developers are just plain stupid. > So I was under the impression that you were criticizing something you > didn't fully understand. > Also, it's difficult to trust someone who claim that LISP is the best > language there is and everybody else is just stupid for not realizing that. > I only tend to trust people who realize that their language is not perfect > and is just one of the many languages available. > > -- > -- > You received this message because you are subscribed to the Google > Groups "Clojure" group. > To post to this group, send email to clojure@googlegroups.com > Note that posts from new members are moderated - please be patient with > your first post. > To unsubscribe from this group, send email to > clojure+unsubscr...@googlegroups.com > For more options, visit this group at > http://groups.google.com/group/clojure?hl=en > --- > You received this message because you are subscribed to the Google Groups > "Clojure" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to clojure+unsubscr...@googlegroups.com. > For more options, visit https://groups.google.com/groups/opt_out. > -- -- You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com Note that posts from new members are moderated - please be patient with your first post. To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en --- You received this message because you are subscribed to the Google Groups "Clojure" group. To unsubscribe from this group and stop receiving emails from it, send an email to clojure+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/groups/opt_out.
