Re: [Sedate] WGLC on draft-ietf-sedate-datetime-extended-06

[Carsten Bormann <[e-mail]>]

(Late reply:)

> On 2023-01-20, at 02:28, Martin J. Dürst <[e-mail]> wrote:
> 
> Hello everybody,
> 
> I'm not a regular participant, and so I might get something wrong, but I'm rather concerned that with or without the "[sic!]", this will lead to confusion if not actually explained in more detail in the document.

Well, this is a note about an idiosyncrasy in the tzdb.
So there is no need to understand it to understand the rest of the document — just make sure you don’t accidentally use these weird tz names.
(The occurrence of “GMT” in the name should be a hint that this is legacy, anyway.)

> In particular, I tried to go through the document and figure out on which occasions which convention is to be used. I wasn't particularly successful. As an example, the document defines the term "IANA Time Zone", but searching through the document, this term isn't used anywhere, although examples such as Europe/Paris turn up in various places.


> And the example used (Etc/GMT+10) isn't totally helpful, as it could be the same as current Hawaii or Brisbane. (The Pacific ocean is very wide.)

The Etc/GMT* time zones are indeed not very helpful.

> Also, there's the example of "2020-01-01T00:00+01:00[+01:00]". Would that be the same as "2020-01-01T00:00+01:00[Etc/GMT+1]" or the same as "2020-01-01T00:00+01:00[Etc/GMT-1]", assuming we needed such a thing, e.g. for North of Swalbard or South of Cape Town?

Don’t use Etc/GMT*, I’d say.

Grüße, Carsten