antonagestam

> I don't understand the mentality of posts like this.

The purpose is simply to put my opinions and experience into writing. I chose to share in case it is useful to others, which it turned out to be. The intention is also to inspire teams to formalize their own list like this, and to establish a culture around how to write tests.

> I don't think it's true, but if it's true, make an argument.

That wasn't my purpose of this post. Every listed item in the article is a choice, for each of those there are alternatives with trade-offs.

> I don't think you know what a good test is.

Likewise, from what you describe as a "powerful" test sounds to me like nightmare to maintain and it saddens me to think about what you have left behind for posterity over the course of 30 years. I don't think tests should be powerful, I think they should be dumb and dead-simple to understand, and that they rather build usefulness in numbers.

Yes, there is still a take home assignment involved, I think it has been tinkered with to reduce the time required slightly.

Isn't this just debuggex.com? That's hacker but not news.

From the docs: extras_require A dictionary mapping names of “extras” (optional features of your project) to strings or lists of strings specifying what other distributions must be installed to support those features. See the section below on Declaring Dependencies for details and examples of the format of this argument.

A matter of critical mass ;)