Alignment with EAT

[thomas-fossati]

As both EAT and CoRIM are RATS products, we should aim to align them.

Specifically, I believe we should take a look at RFC9711 and incorporate all the relevant claims into the measurement and/or environment maps.

[yogeshbdeshpande]

EAT is one format, perhaps there are multiple.

The suitable evidence transformation document should state how Evidence can be transformed into:

1. Equivalent ECTs in CoRIM for Verifier Processing
2. CoRIM Profile can define, how equivalent Measurement Value Maps can be constructed.

[steven-bellock]

I agree with @yogeshbdeshpande. There are lots of evidence formats and RATS, while ensuring internal interoperability between its specifications, should not be playing favorites just because a specification comes from RATS.

[thomas-fossati]

EAT is one format, perhaps there are multiple.

Yes, but EAT is the only attestation format defined by RATS, which makes it "special" in a sense.

The suitable evidence transformation document should state how Evidence can be transformed into:

1. Equivalent ECTs in CoRIM for Verifier Processing
2. CoRIM Profile can define, how equivalent Measurement Value Maps can be constructed.

Sure, but... This means that any profile wanting to convert from an EAT-based evidence format must register new codepoints with the measurement/environment maps for the EAT claims it uses.

Wouldn't it be more convenient to prepopulate those maps with EAT claims?

[thomas-fossati]

I agree with @yogeshbdeshpande. There are lots of evidence formats and RATS, while ensuring internal interoperability between its specifications, should not be playing favorites just because a specification comes from RATS.

It’s playing favourites in the sense that EAT claims would be pre-registered as measurement & environment map types. Nothing more than that.

[steven-bellock]

EAT claims would be pre-registered

What about the rest of https://www.iana.org/assignments/cwt/cwt.xhtml?

[thomas-fossati]

EAT claims would be pre-registered

What about the rest of https://www.iana.org/assignments/cwt/cwt.xhtml?

We should only capture the portion of the registry that relates to attestation. Starting with what is registered by RFC 9711 seems like a sensible approach to me.

[thomas-fossati]

I agree with @yogeshbdeshpande. There are lots of evidence formats and RATS, while ensuring internal interoperability between its specifications, should not be playing favorites just because a specification comes from RATS.

It’s playing favourites in the sense that EAT claims would be pre-registered as measurement & environment map types. Nothing more than that.

Also note that CoRIM already treats TCG DICE specially. It'd seem only fair that the IETF attestation format receives similar treatment, at least :-)

[yogeshbdeshpande]

This is out of scope given the feauture freeze we agreed in July 2025