SURVEY! Help us design the Rust Identity Library

[ronniegeraghty]

📝 We'd love your feedback on Azure Identity for Rust!

We're currently exploring different implementation options for the Azure Identity library in the Azure SDK for Rust. One of the key decisions we're evaluating is whether to:

• Build a fully Rust-native implementation, or
• Leverage an AOT-compiled .NET binary to handle certain identity functionality.

We'd really appreciate it if you could take a minute to share your thoughts and preferences in this short survey:

👉 Take the survey here

Your feedback will directly shape how we build this experience going forward. Thank you!

[jlaundry]

@ronniegeraghty could you please share more background on why Microsoft are considering replacing the current azure_identity library with a new one, let alone one cross-compiled from .NET?

And what that would mean for previous decisions; specifically making DefaultAzureCredential a dev time credential, and not support ChainedTokenCredentials?

[ronniegeraghty]

@ronniegeraghty could you please share more background on why Microsoft are considering replacing the current azure_identity library with a new one, let alone one cross-compiled from .NET?

And what that would mean for previous decisions; specifically making DefaultAzureCredential a dev time credential, and not support ChainedTokenCredentials?

Thanks for the thoughtful question — we really appreciate you raising these concerns.

To clarify upfront: we on the Azure SDK for Rust team believe that a Rust-native identity library is the right long-term direction for the Rust ecosystem, and we understand why a solution based on a cross-compiled .NET binary could feel frustrating or out of place.

That said, we don’t plan to replace the existing azure_identity crate. While the crate is still in beta and its API surface may evolve before GA, the outcome of this survey won’t drive any changes to that public API. We're specifically looking at how best to enhance the internal implementation details, and the survey is meant to help assess whether a .NET AOT base would be acceptable to Rust developers.

Teams across Microsoft who support authentication — beyond just the SDK team — want to better understand the range of customer expectations before committing engineering resources to a fully Rust-native implementation.

In other words, this isn’t a decision that’s been made. We’re trying to ensure that any path forward is informed by clear customer feedback, so that we can advocate for the right investment internally.

Thanks again for engaging — feedback like yours is exactly what helps us make better long-term decisions.