If I put "seaborn does not claim to be as good as ggplot" in the README, will you all shut up about it?

Less snarky: I had been planning on working this afternoon to implement a new feature I am excited about. Then a data science influencer tweeted about how seaborn sucks, so instead I felt bad for a while and then made some bread.

This is how your free software gets made, people.

Also I appreciate the nice responses, but I'm not really interested in hearing about how it's actually better than package x or whatever.

I promise you, writing software is not a competitive sport.

Thanks, everyone, for coming to my pity party.

One other thing I wanted to say is that, despite what you may have read on Medium, seaborn was never intended to be "ggplot for Python."

That's at the root of why I find everyone arguing over that specific comparison so annoying.

(In a very real sense, seaborn was never "intended" to be anything in particular — it was a public github repo of assorted functions for my personal use that people found and got excited about. It has since grown in fits and starts as I've diverted attention from my actual job).

seaborn is what it claims to be: a higher-level interface for matplotlib. It makes some tasks easier than using matplotlib directly, while delegating precise customization to the matplotlib API. In my opinion, this is a good balance, as matplotlib excels at precise customization.

I get why people used to the totalizing abstraction of ggplot would find this confusing or frustrating.

But too much abstraction can also be frustrating when you want to get specific details of a figure exactly right.

There can be more than one approach to data visualization.

I'm also just not personally invested — at all — in whether other people use R or Python. R actually isn't used much in my field (make of that what you will). It has very little relevance to my life, aside from when people are trying to rabble-rouse for Twitter engagement.

Which is all to say: To the R fans, I really don't care about your opinion of my library. And to the Python fans, I'm not interested in helping your team "win data science".

Please just leave me out of it.

The only actually rewarding aspect of writing free software is in knowing that you helped people do interesting work, so to those who responded to these tweets in that vein, thank you.