[Cbor] Re: Abandon cbor-packed (Re: Early allocation for packed CBOR)

[Carsten Bormann <[e-mail]>]

On 16. Dec 2024, at 04:51, Vadim Goncharov <[e-mail]> wrote:
> 
> angry

Vadim,

I very much see that you are angry, but I’m having a hard time finding content in your message.

First you dismiss my observation that Packed CBOR is not a generic (byte sequence) data compression scheme.
Thank you for the pointers to various such schemes that may be applicable for constrained devices in certain configurations; I’m sure it would be worth to come up with a paper that compares such schemes for the use with CBOR and maybe makes some recommendations.
But this is not what Packed CBOR is about, and Packed CBOR is not “ineffective” — it works fine, thank you.

Then you argue for abandoning Packed CBOR in favor of a rewrite of CBOR that is incompatible with CBOR as defined in RFC 8949 and claim that “it is very simple to retain compatibility - “modified" CBOR is just wrapped into bstring/array under tag”.
Certainly, we also can transport JSON or ASN.1 BER inertly in CBOR text or byte strings, I’m not that interested in adding a few more “mostly CBOR” or “modified CBOR” to that list.
What you wrote in your message about what this CBAR (?) could be is incomprehensible, unfortunately.

Then you deny that CBOR (itself?) is extensible, when clearly it is using the tag mechanism — the two table setup tags in cbor-packed draft are useful as “batteries included”, and other table setup mechanisms can be added using additional setup tags.

Grüße, Carsten