Martin Atkins

Hmm, this is an interesting design question... Elsewhere in the language it's typical for the JSON variant of the language to use simple strings containing just direct traversal syntax in...

I'm afraid I'm not familiar enough with the state encryption config structure to have a strong opinion about this yet, and today isn't a working day for me so I'm...

Hi again! Today, as promised, I took some more time to understand how the encryption-related codepaths currently work. It seems like the root of the inconsistency here is that the...

Thanks for sharing this use-case, @geekofalltrades. As you suggested, getting something like this working across all resource types in all providers would be a pretty heavy lift today because only...

I believe that the Go team's policy is that compatibility-related GODEBUG settings are maintained for at least two years but might be removed after that. In this particular case I'm...

Hi @stevehipwell, OpenTofu is following what we understood to be the modern recommendations for non-container-image artifacts: a custom `arrifactType` value associated with standard `mediaType` values. For modules in particular the...

The reference I have most immediately to hand is [a section of an ORAS guide](https://oras.land/docs/1.3.0-beta/concepts/artifact#determining-the-artifact-type), which of course is not a recommendation from the specification, so I understand that's not...

I would like to understand more about what you find surprising about using `archive/zip` here. Is it just that container images traditionally use tar, or is there something _OpenTofu-specific_ about...

Today I retraced my steps on the earlier research we did when deciding to use `artifactType` instead of custom media types, and in particular the comparison with Helm's approach that...

Hi again! We previously discussed that using the zip archive format might make it hard to construct a package which preserves the Unix-style "execute" permission that a module could rely...