Mainly Tech projects on Python and Electronic Design Automation.

Sunday, September 17, 2017

Pythons Positive Press Pumps Pandas

With the publishing of two Stack Overflow blog posts, this week has seen several articles feeding off them and cherry picking parts, as well as divorcing those quotes from context, but all in praise of Python.

That's got to be all good for Python, right?

Well this is hype. It can be riding the tigers tail were control is limited and we can end up in a local maxima that it is hard to escape from. Ruby had this kind of thing with Ruby on Rails, a successful application that ended up overshadowing the language - my search for Ruby on Rails for example had Google giving, under the "People also ask" banner: "Is Ruby and Ruby on Rails the same thing?".

Several of the articles referencing the SO originals latch on to SO 's belief that the Pandas library and therefore data science is the key to Pythons success. That is all well and good, but these articles, and the original SO ones, don't show that Pandas is one library amongst a broad area of computing were Python has a useful presence. It is often the breadth of Pythons influence that turn users to Python:

  • The statisticians who first learned R, Who still use R, but also use Python - Some switch wholly from a language designed for them to Python because when they want to do things with their stats then Python is more likely to support not just the stats, but their other field too.
  • Teaching. Prestigious Universities changing their computing course-ware language to Python because it is more relevant than Lisp/Scheme. If they want to ultimately control a robot, Python is more likely to have existing libraries for that.
  • Pythons ease of learning reputation "rubs-off" on libraries. Wrapping advanced GPU acceleration libraries, or machine learning libraries in Python; even though they are mainly written in other languages the Python integration gives those libraries an air of approach-ability. There is substance to this, as when well wrapped, these external libraries become Pythonic; aspects of their use just come for free from the users knowledge of existing Python as they conform to its idioms.
I would like to congratulate the Pandas team in their success, (Yay), but with current reporting, we need to remind those new to, and just looking at Python, that it is no one-trick Pony!


1 comment:

  1. Hi
    Well considered comment. As someone who is new to the present computing scene I have been doing some general reading to see what's out there. Python does have a lot going for it. It is true that some babies can sometimes take on a life of their own that looks as if they will overshadow the parents,but the source can in the future still give life to others,even more successful. Do others see libraries such as Ruby an Rails and Pandas have this capability? Is it therefore not wise to ensure one invests in not disrespecting the power of the parent while welcoming its children.



Subscribe Now: google

Add to Google Reader or Homepage

Go deh too!

Blog Archive