Re: [Sedate] Can offsets like [+02:00] be used instead of IANA names in brackets?

[Carsten Bormann <[e-mail]>]

On 2022-03-17, at 18:01, Justin Grant <[e-mail]> wrote:
> 
> I'm hearing consensus on this thread that it's OK to add offset strings in brackets, e.g. "2021-12-09T00:00-08:00[-08:00]" in order to be compatible with prior art in java.time.ZonedDateTime. Caveat: this format is allowed but is generally not recommended because of the issues noted by @Edward Welbourne and others. Documentation should discourage but not prohibit use of this format.

Declaring consensus is the remit of the WG chairs, but as an editor of the WG draft I’m hearing clear direction from the WG to include this in the next revision of the draft.

> Does anyone object?
> 
> If not, can we merge https://github.com/ietf-wg-sedate/draft-ietf-sedate-datetime-extended/issues/12 ?

Can’t merge issues, but I’m looking at https://github.com/ietf-wg-sedate/draft-ietf-sedate-datetime-extended/pull/15

The current PR defines a few terms but then doesn’t use them; the editors can supply the necessary glue.
The discussion here so far can probably provide some guidance on what the semantics of this form are.

Why is it necessary to deviate from the definition of Zulu time in RFC 3339 (adding a plus)?

I would like to avoid discussing the weird legacy Etc/GMT timezones (exhibit 1: PR#15 gets this wrong).

I’d prefer to have references to the specifications (RFCs?) relevant for the IANA timezone database; the PR text leaves that as an exercise to the reader.  So if the WG can point us to the documents that you want to have referenced, that might help reducing ambiguity.

Grüße, Carsten