Hacker News Reader

Comment by rgalanakis | Hacker News Reader

I changed it to "relatively few." We actually have several thousand altogether, but we have a lot of code. I don't want to discourage the great progress we've made.

View on HN

The core point is the same: With sufficient test coverage, you can change one variable (Python 3 vs 2) and control for all others. Tests are not an end, they are a means to an end; the courage to change your software in the face of a constantly changing world.

Sounds like you're not there, but maybe you have specific domains of EVE that you feel comfortable changing at your whim. That's the success to focus on.

>With sufficient test coverage, you can change one variable (Python 3 vs 2)

That is a pretty hilarious way of putting it.

Somewhat naive too to be honest. As someone that had just plain old ruby circa 1.8.5ish code (not rails) that I ported to 1.9 a while back, even with tests there were differences in behavior that I guarantee you probably weren't thinking of writing a unit test for.

Unit tests are nice, but they don't make things braindead easy. Just make it a bit easier to narrow down where something might be broken/different.

With sufficient test coverage, you can change one variable (Python 3 vs 2) and control for all others

The safest change to make is the smallest change possible, after all.

I understand your position. I've managed a group that was struggling to increase test coverage. It's hard to make testing work if management wasn't behind it from the get-go. Short term thinking almost ensures this doesn't happen. It is possible to catch up, however. A friend of mine did it with a DoD project. (In C++!)

Reply on news.ycombinator.com