#microformats 2023-10-20

2023-10-20 UTC
#
j​kingweb
Done.
Renfield, sebbu, gRegorLove_ and [tantek] joined the channel
#
[tantek]
There's also the yearless date, e.g. many birthdays or holidays/anniversaries are entered with just the month and the day, which in ISO8601 is something like --03-17
#
[tantek]
I'd lean towards keeping Z bc it's such a common use case and likely worth the brevity vs slightly more consuming code trade-off, though I’m sympathetic to that too
#
j​kingweb
Yearless date is not something currently accepted by VCP parsing, though, unless I'm grossly misreading.
#
sknebel
vcp supports less formats, yes. but the change should not break any of the otherwise possible formats (e.g. whats in HTML spec as datetime values)
#
sknebel
I think the framing should be something like "if temporary value matches the following pattern, normalize it as follows: <> if this other pattern, as follows: <> otherwise, pass through unchanged"
#
sknebel
which then cleanly leaves everything not specifically handled alone
#
sknebel
I dont think there are cases where there is a risk of confusing formats, luckily
#
[tantek]
Indeed I don't think we have seen any use cases for vcp for yearless dates so therefore we haven't designed any
#
[tantek]
vcp also predates the time and data elements so it's likely we can provide better authoring guidance
[manton], eitilt, [benatwork], [schmarty], ehmry and [jeremycherfas] joined the channel