This is still in progress, but my deprecation poli...
# development
f
This is still in progress, but my deprecation policy work has scope-crept in a general API stability statement that tries to analyze Pants' API stability from the POV of multiple user personas. This is still WIP but feel free to look. I'll also share a bit tomorrow in the team meeting. https://docs.google.com/document/d/1PQsPRh43g4_pcqnTPhDWGSZha9x_0IPzEkHCw_MAE60/edit?usp=sharing
The main thing I've come up with is that...well I can't find a user type that breaking changes don't hurt. Especially if we can come up with a way to steer people towards preferred, newer APIs while retaining older ones as aliases and opt-in backends.
b
Another thing to consider here, and I was thinking about bringing it up but I'll be at Dr, if we could decouple engine releases from plugin releases, that allows for longer/shorter cycles and more stomachable code duplications (to support "old" paths)
Although this could be part of the "solution" 🤔
f
That sounds like it could be fruitful, but I'm probably not the best to judeg. I'm trying to stay focused on a problem analysis here. I don't have the depth of experience with the Pants codebase to suggest implementation details. Right now I'm doing my best to wear a combination of product manager and UX hats rather than a dev hat 🤠 😆
b
Oh yeah. I'm a shark-jumper 😛