[Cbor] EDN (fileext) naming and compatibility (Was: Rebooting the naming discussion)

[Vadim Goncharov <[e-mail]>]

On Tue, 5 May 2026 02:11:46 +0200
Carsten Bormann <[e-mail]> wrote:

> On May 5, 2026, at 02:06, Vadim Goncharov <[e-mail]> wrote:
> > 
> > So this just means that we will not use .cdn for it. Given that draft, and
> > thus eventually RFC, is titled "CBOR Extended Diagnostic Notation", this
> > four words result in .cedn - very simple resolution, I think.  
> 
> We could do that.
> Of course, the question would immediately be “what is CBOR (non-extended)
> diagnostic notation then”?
> 
> Let’s get rid of “extended”; there is no need for that.
> We don’t usually slap “extended” on a protocol each time we extend it.
> You can go all the way from RFC 793 to RFC 9293, it’s still not called
> “extended TCP”.
> 
> Grüße, Carsten
> 
> 
> (The draft is called that way because we adopted it when it still was a
> small addition to RFC 8610’s EDN. It no longer is that, it is the modern
> version of CBOR diagnostic notation.)

I don't agree with this bikeshed - the more specific name/extesnsion is, the
better, modulo being "not very long". While here and on compatibility, I also
don't agree with comma-less syntax in maps and "+" for concatenation. This
breaks for me programming language syntaxes to be the superset of *full* EDN
(forcing to omit something instead).

-- 
WBR, @nuclight