And also, proposal, useful for make new atmos feat...
# atmos
m
And also, proposal, useful for make new atmos features works fine with all existing stable features from the start (except very weird corner cases) https://github.com/cloudposse/atmos/issues/2865
e
I think it's a lot easier said than done. It's a Cartesian product that explodes and we simply just don't have the resources, financial or human, to help with that.
Until more corporate sponsors step up to help us, stability is going to continue to be a big challenge for us as we push the product forward. I hope that overall it begins to stabilize as the surface area we cover matures.
Definitely hate having regressions of stable features
Definitely experimental features should not affect stable features
But we continue to be surprised by the ways that teams are using Atmos in ways that we hadn't thought of, and therefore there is no substitute real-world testing.
So by adding the experimental tag we hope to convey the relative immaturity of some features while not withholding it. At the same time because those features touch at the core of Atmos, preventing instability in stable features is just not entirely possible.
So we commit to fixing them rapidly and we appreciate all the submissions we get. We also promptly review all contributions.
Would love to have something like "pacts" qhere we can know concretely that any feature or change doesn't break something.
Part of the reason we introduced atmos.tools/examples with videos is to prove every feature works, but the ways/combinations that features can be used goes beyond simple matrix tests.
👍 1
We're also shipping config editions. atmos.tools/changelog/config-editions
🐕 1
This will allow us to update defaults for better DX, while allowing teams to pin to date based versions of config.