Clase 10 - “Revenge of the Nerds”
“Revenge of the Nerds”
The first thing I find curious about this article is that it does not consider Lisps as a programming language if not as an idea or even more as a theory. The author compares Lisps to the mathematical discovery that as we have found out over history become more relevant instead of obsolete. at least this was the idea of McCarthy who never intended to implement the eval function.
After reading the 9 new ideas of Lisp the one that most resonated and made my mind run was number 8. Especially when the author mentions that this aspect enables programs to create programs, this idea now we know as AI. So it is quite impressive that without any idea that machine learning and IA technology were possible Lisps is made to execute them. Today we are pushing every day for programs to be more independent and as of now we already have programs that write programs that no human can understand.
Still considering all the good and cool things that Lisps has to offer I still have to disagree with the author. I think that in today's world and after all the man-hours that have been placed into more common languages it would be a bad idea to use Lisps. The world is much different today than what it was 70 years ago. So there are many factors that are built around new languages as security, portability, maintenance, and compliance. At the end of the day, the world is not made of only Hackers, developers, and nerds. So if the rest of the world does not agree that the language you are using is viable you won't get very far with your products, even if they are for your own use.
Also, I strongly disagree that the length of the program is the most important factor in calculating the time it takes to write a program. It is way simpler to write 2,000 lines of code of a simple code than 100 of a complex algorithm. The most important part is the bottom line understanding of what the program needs to do and how this will be achieved in code. So I would not conclude that going back to Lisps would be a good idea. The path going forward is to implement the valuable things of lisp to already proven languages.
Comentarios
Publicar un comentario