Re: [Cbor] 🔔 WGLC on draft-ietf-cbor-array-tags-03

[Jim Schaad <[e-mail]>]

I would say that this is analogous to the question of how date and time are encoded.  There are several different options and we don’t say which of them is deterministic for encoding.  We just say how to encode it once the value and tag are selected.

 

Jim

 

 

From: CBOR <[e-mail]> On Behalf Of Jeffrey Yasskin
Sent: Wednesday, March 13, 2019 11:29 AM
To: Laurence Lundblade <[e-mail]>
Cc: [e-mail]; [e-mail]; Francesca Palombini <[e-mail]>
Subject: Re: [Cbor] 🔔 WGLC on draft-ietf-cbor-array-tags-03

 

On Fri, Mar 8, 2019 at 7:43 AM Laurence Lundblade <[e-mail] <mailto:[e-mail]> > wrote:

In addition to what I said before, a deterministic encoding should be specified (formerly known as canonical) in the specification and that should probably be network / big endian.

 

I think it's the job of the protocol to describe how it's encoded deterministically, rather than this tag. For example, the choice of whether or not to tag an array at all is a potential point for non-determinism to enter, so the protocol needs to specify whether there's a tag, and if so, which one. At that point, it's made the choice of little-vs-big endian, so there's no more question for this document to answer.

 

Jeffrey