C is Manly, Python is for “n00bs”

The article takes the stance that all stereotypes are groundless, but a more realistic position is that they are often generalizations with some valid foundation. I'll just illustrate with my experiences of Python software. I have used many, many data manipulation, simulation, visualization, etc. tools over the years, most of them coming out of either academia or government labs. I long ago developed the habit, when faced with using a tool written in Python, to look for an alternative written in C/C++, because it has been my experience that the C/C++ alternative will usually be more stable and better thought out. I have not worked deeply with Python or Python programmers, so I can't offer an informed opinion of whether the main problem lies in the people or the tools. It has been my tentative hypothesis that this trend is the result of C/C++ placing a higher bar to use, requiring developers to both be more capable and spend more time planning.

First I was reminded of Uncle Bob's talk on programming language hubris:

https://www.youtube.com/watch?v=YX3iRjKj7C0

but as I went deeper thought about the effects of programming language tribalism, I was reminded of

http://web.mit.edu/humor/Computers/real.programmers

it's reference to

https://en.wikipedia.org/wiki/Real_Men_Don't_Eat_Quiche

and how this played out as a sound bite in US popular culture at the same time as the decline in women computer science enrollment began to decline. Being in the high school graduating cohort where the decline began, the decline of women in computing a subject I'm interested in. It's one of those subtle cultural/generational differences between me and my boy.

The title should have (2015) in it.

This is a weird one, I feel like I am stumbling into a bit elephant trap, the authors seem to know what they are talking about but have missed some fundamental points.

Firstly there is a statement asserting that it only takes a few months to learn a language if you are a competent developer, which I agree is true to a point, but the assertion continues with the view that therefore we shouldn't hire given what languages a person knows.

Now, in a big company where there is a lot of runway for a project and a good training budget - and where you are building capability, spotting talent and hoping to build your own cheap well integrated team rather than a bunch of half crazed guns for hire, then yup, agree, but a bit of nuance is needed. A lot of places cannot function with people learning things for two or three months (or weeks, or days even). There's work to be done and it needs doing.

Then there are statements about Perl, Python and Javascript being fairly similar - which they are in a narrow technical look how the compiler works way. But I don't know about you (obviously) but although I loath Python (irrationally) I would feel relaxed and happy picking up 30k loc to debug and maintain if it were written in Python, I would be somewhat nervous if it was Javascript and I would be having a breakdown if it was Perl.

But as the authors note, horses for courses - Python is not a great fit (ok it's pretty good) for processing lots of strings and so on, Perl is good for that. Javascript - guess what - I think Javascript is a boon for creating web front ends. Perl, not a great choice.

In a similar vein I use R sometimes for interactive data mining. It is very handy in term of setting up analysis pipelines and creating insight. I like it for that. But I have folks working for me who write substantial projects in R and I constantly worry about maintenance and sustainability in the field. It isn't great for creating a structured and well understood code base.

With respect to the gender politics and the "exclusion" angle, this was a very real problem 25 and more years ago - one that I (somewhat shamefacedly) must confess that I innocently profited from. I got the chance to learn Smalltalk, Objective C, C, plsql and C++ using expensive compilers on expensive computers. This gave me a substantial leg up, people in India or Africa had no chance (effectively) of doing this. Now almost everyone can get the opportunity (I'm not saying with ease or that the opportunity is just there, but it is possible) to learn Java or Python or C++, MySql, Javascript and many more valuable useful tools. The question in my mind is why is it that some people end up learning ones that then aren't too useful to them? I learned php some years ago for a particular project, it was useful, I will never use it again.. but why is it that someone learns php and then doesn't learn Python? Or Ruby? Or Julia?

If you want a job in a company that uses Python why wouldn't you learn it? Why wouldn't you write a few projects and put them on Github?