re-word the go lock-in section to be less gatekeep-y

Signed-off-by: James Humphries <james@james-humphries.co.uk>
This commit is contained in:
James Humphries
2025-07-21 18:00:56 +01:00
parent d0c9697aff
commit af81a9200b

View File

@@ -35,15 +35,13 @@ While the underlying protocol uses gRPC and protobuf, virtually no one interacts
### The Cost of Go Lock-in
Consider who we're excluding from the ecosystem:
- **Security teams** who may work primarily in Python
- **Cost management teams** using data science stacks (Python/R)
- **Platform teams** with existing TypeScript/JavaScript tooling
- **SRE teams** who just need simple Bash script integrations
By limiting extensions to Go, we're excluding teams that work in Python, TypeScript, Rust, Ruby, Java, and many other languages.
Each team has their preferred tooling and expertise - and I don't believe that we should force them to learn Go just to extend OpenTofu.
These teams currently resort to:
- Parsing plan JSON files (missing critical runtime context)
- Writing wrapper scripts around `tofu` commands
- Writing wrapper scripts around `tofu` commands and reading either the human-readable or machine-readable output
- Building complex CI/CD pipelines for what should be simple validations
- Maintaining separate tools that duplicate OpenTofu's resource understanding