Re: [Cbor] Unusual map labels, dCBOR and interop

[Wolf McNally <[e-mail]>]

Anders,

> On Mar 27, 2024, at 10:30 PM, Anders Rundgren <[e-mail]> wrote:
> 
> Here we go again 😵

Yes, you mis-characterize and I respond. We can do this as many times as you like.

> Indeed.  Different persons have provided pretty verbose rationales for dDCOR where they freely mix things that are unrelated to a basic deterministic encoding scheme such as the range of integers or the support of "elision" in Gordian Envelopes.

Gordian Envelope has been proffered as an existence-proof of a semantic layer that benefits from a lower layer that supports determinism.

To refute this, I believe you were working on an implementation of Gordian Envelope that doesn’t use a separate dCBOR layer (something I stated was perfectly possible, just a lot more complex to get right)— how is that coming along?

> My comments were always about "Numeric Reduction" which condensed rationale is stated above.  I have excluded cognitive overload or JavaScript considerations as valid motives for creating an IETF standard.

Which were never primary motivations for dCBOR, merely side-benefits. And your adoration of the builder pattern is not an argument for *not* having a standard.

The group has discussed at length the utility of determinism, as well as the challenges in achieving it.

> Regarding consensus see: https://mailarchive.ietf.org/arch/msg/cbor/uI9s9VgYcH6DNFLaXb-f7iNDXDE/

“Sources say…”

~ Wolf