Years ago, I was mostly interesting in Ruby/Rails because my friends in SF were interested in it and so I wanted to learn from them. There were also a ton of articles on HN that popped up.
But as soon as I became interested in Data Science/ML, I wanted to immerse myself in as many tutorials/projects as possible. As it so happens, a lot of books/tutorials I've read are in Python, so I figured why not. I really appreciate the simplicity of the language and the fact that so many interesting libraries are available.
I find it quite a beautiful language to try to learn.
IMO python and its popularity is somewhat split between a scripting language, maybe a R alternative, with a data science focus on one hand. On the other hand is python as a web dev language with Django and flask as the popular frameworks.
We're looking to hire on the web Dev side [0]. We do get a number of people who have more of a scripting focus, but limited web Dev experience. I think some of that comes because of the online MOOC phenomenon and the proliferation of python courses that are really python scripting (not python web dev) courses.
I'm curious what the HN community thinks of my hypothesis of that split and if the "incredible growth" is one-sided as a scripting language.
It is not just because it attracts beginners, it is also because it is a good tool for a lot of tasks where you can trade some performances for ease of use.
I think a lot of the recent growth is due to theano/keras, the most popular deep learning platform out there. It also attracts more and more data scientists, now that distributions like Anaconda make it 'easy' to install on windows.
Also, everything had to be redocumented and reanswered for python 3. I wouldn't be surprised if people had to look up their issue repeatedly on stack overflow until they finally find help.
Every top answer in python seems to be dating back to 2009 (python 2.1), quite a few are completely broken code nowadays (blame the breaking change in the standard API).
It would be interesting to collect data to see how many continued using python from prototype to production. Python would definetly help to build MVP, when in production where actual concurrency matters; very few like Instagram have built robust python based architectures.
Btw, achieving decent concurrency in python is quite straightforward. Understanding of Greenlets, Gevent, uvloop Cython etc. could prevent headache in future.
