[microformats-discuss] International date formats

[Tantek Ç elik]

On 7/26/05 7:27 AM, "Carl Beeth" <carl.beeth at gmail.com> wrote:

> On 7/26/05, Ian Hickson <ian at hixie.ch> wrote:
>> Human consumption is not an issue here anyway.
> 
> I disagree, as said earlier there is an instant killer app for this
> MF: making dates less ambiguous without need for any processing beyond
> what the browser already has.

As I said in the reply to Ian's message, I want to avoid making this an
issue of absolutes.

Human consumption is not *as much of* an issue here.

We do want to make the date time title reasonably possible for a human to
verify, but don't have to make it "ideally human readable".  That's what the
element contents of the <abbr> are for.

Think of the whole element (<abbr>) as something we want to make overall
good for human friendly (consumption and authoring).  We solve that by
putting the truly human friendly version inside the element contents which
works for the 99% case.  The 'title' is the alternate representation whose
primary purpose is to be machine parsable (and compatible with an existing
date time standard), and whose secondary purpose is to be reasonably human
verifiable that it is the same as the element contents.

BTW, in general we want to avoid duplicating information like this (i.e. the
DRY principle), and thus such a use of <abbr> to separate human vs. machine
readable versions of a piece of data MUST be the exception, not the rule,
and only something we resort to when necessary (i.e. when a greater
principle, like humans first, supercedes).


> Having said that this MF seems like a perfect candidate for a
> greasemonkey extension. takes the iso date in the title of the abbr
> tag and re-parses it in a appropriate local format and re-injects it
> between the abbr tags.

Yes.  That would be an excellent use of GM.

Thanks,

Tantek