I assigned minute and second zero through 60, not for any "leap" considerations, but because 60 minutes and 60 seconds are supported by xs:dateTime. And the reason for that apparently is that it is supported by ISO 8601. We had discussion about this a while ago; it was suggested that ISO made a bad choice in allowing these values and that we should profile them out. But while profiling out features in 8601 is appropriate, I don't think profiling out features in W3C schema is a good idea.
> -----Original Message-----
> From: Discussion of the Developing Date/Time Standards
> [mailto:[log in to unmask]] On Behalf Of Saašha Metsärantala
> Sent: Thursday, March 31, 2011 10:24 AM
> To: [log in to unmask]
> Subject: Re: [DATETIME] BNF complexities
> > BNF for the dateTime spec
> Thanks a lot for that! It is really appreciated! I plan to read it
> tomorrow. Right now, I'll only write a little comment about:
> > minute = zeroThru60
> > second = zeroThru60
> Leap seconds occur but (as of today) not leap minutes. Therefore, I
> wonder why both are assigned zeroThru60.
> I know that some people mean that they would prefer one leap minute
> (about once in a century) instead of one leap second several times each
> decade, but as of today, leap minutes are not used (yet). I would
> therefore suggest not to use zeroThru60 for BOTH minutes and seconds.