Ask HN: Is there evidence regarding PG's theory about powerful languages?

Re:Ask HN: Is there evidence regarding PG's theory about pow... by hiphipjorge on Monday April 11, 2016 @01:22AM (#11469790)

This is totally subjective, but I think we're at a point where the language ecosystem is much more important than the actual language. This was probably not the case when pg was writing software for Viaweb and there wasn't nearly as much FOSS that you could just use. At that point in time the LANGUAGE was probably much more important because there was not a lot of differentiation beyond that. Now, we're at a point where the language isn't as important and the ECOSYSTEM around it is much more powerful that then language itself.

[ Parent | on HN ]

Re:Ask HN: Is there evidence regarding PG's theory about pow... by hga on Monday April 11, 2016 @04:29AM (#11470304)

That's certainly true. While Clojure isn't exactly my favorite Lisp (it isn't even a LISP, as in a LISt Processing language any more, and is rough around the edges in various places), running on the JVM ecosystem makes it a great choice for many things (well, aside from logging, standard JVM logging is "sacrifice a chicken" territory, but there are native solutions now as well).

Want to us a RDBMS? There are many libraries for doing that, my favorite is Korma, and they all use the JDBC interface and battle tested JDBC implementations, that part of the plumbing is trivial. Etc.

[ Parent | on HN ]

Re:Ask HN: Is there evidence regarding PG's theory about pow... by a_bonobo on Monday April 11, 2016 @01:38AM (#11469827)

I've observed this in bioinformatics -

Python, Perl, R are much more common than (let's say) lisp or Ruby, partially because more people know these first languages (possible exception in R - is that even taught in university?), but IMHO more because libraries such as BioPerl, Biopython, Matplotlib (Seaborn), numpy/scipy, ggplot2, the entire CRAN ecosystem are so much more mature than what you get in lisp/Ruby, you can get amazing stuff done very quickly.

[ Parent | on HN ]

Re:Ask HN: Is there evidence regarding PG's theory about pow... by adenadel on Monday April 11, 2016 @01:39PM (#11473518)

I agree with you completely and just wanted to comment that R is fairly widely taught in universities. The difference with R is that it is often taught in statistics courses rather than CS.

[ Parent | on HN ]

Re:Ask HN: Is there evidence regarding PG's theory about pow... by collyw on Monday April 11, 2016 @07:58AM (#11470919)

I was thinking something similar. I used to use Perl until around 5 years ago, and its a pretty expressive language, great for building all the blocks for your application quickly. Then I discovered Django, and all those blocks I would have been building have been built for me.

[ Parent | on HN ]

Re:Ask HN: Is there evidence regarding PG's theory about pow... by Kinnard on Tuesday April 12, 2016 @06:53PM (#11484286)

Certainly javascript via npm is the best example. Javascript is eating everything because of its massive ecosystem.

[ Parent | on HN ]