I have an urge to damn it all and delete tests left and right and leave it to whoever touches the untested code next to write it from scratch.
But there's some knowledge stored in these tests that would be a real loss. And I doubt my teammates would like losing the regression tests, even if they're a bit slow and unwieldy.
Every path is hard :(
Large, slow tests aren't just a burden on your testing and development process.
They're also a trap for spending way too much time and energy reworking them to be smaller and faster and then not being sure what you actually accomplished at the end of the day.
- Me, having fallen into said trap today. 💀
It reminds me of "slashing the load" as @GeePawHill puts it, on a personal level. Hoping to see these kinds of effects.
For my next three week sprint starting tomorrow, I've decided to cash in my overtime and work 6 hour days for its entirety. As an experiment.
Now that it's almost here, I'm really hyped to see what happens! Two extra hours of time and energy every day! What's that going to do for my off-work routine??
How do we feel about pair programming 'round here?
I've been writing a lot about pairing for the last few months.
How to migrate from one Mastodon server to another without losing followers:
1. Sign up on new server
2. On NEW server: Go to Account -> Moving FROM another account
3. Enter old account's handle
4. On OLD server: Go to Account -> Moving TO another account
5. Enter new account's handle and submit
On the birdsite I was fairly picky about following people. I only followed people who posted stuff I like and didn't post much else, and I rarely unfollowed anyone.
Now I'm deciding to go for a wholly different strategy: I'm following anyone who could lead me to people whose posts I want to see, and will unfollow people at the drop of a hat.
I do need to make these stakes clear, since I'm used to following and unfollowing being in the same social space as friending and unfriending people.
The Duality of Xan
Nat Bennett
Eugen Rochko