Look at all the "moved from $PL1 to $PL2" blog posts, build a matrix, and predict future PL popularity! https://erikbern.com/2017/03/15/the-eigenvector-of-why-we-moved-from-language-x-to-language-y.html
miniblog.
Related Posts
I'm surprised by how many different weather forecasts I can get out different apps. Surely there aren't many API providers for weather predictions?
Alternative conspiracy theory: do weather apps that predict nicer weather get more downloads?
I've worked at companies where buying more computers was more affordable than developers, and I've also been at places where it's the opposite.
It's really weird having the different mindsets. Sometimes it even varies between teams and I rarely predict it in advance.
A fun project for generating random Stack Overflow questions with a neural net: https://stackroboflow.com/about/index.html
It's also fascinating to read that the author was unable to predict question popularity, even by manual examination.