- From: Malcolm Rowe <malcolm-what@farside.org.uk>
- Date: Fri, 11 Jun 2004 16:10:09 +0100
Dean Edwards writes: > would it be useful to extend the "precision" attribute to cover > dates/times? this might eliminate the need for some of the extra input > types. Someone else also pointed out a requirement to enter just 'month/day', for anniversaries, but I'm not sure there's a good enough reason to include it. Absent that, it sounds like we have cases that involve: * a point in time, either at second, minute, day, (week?,) or month precision. * a time, either at minute or second precision. It sounds like a precision-like attribute is something that would make sense, and also that a 'time' type is probably necessary, too. We should probably not call it 'precision', though, unless it's ok for the domain of allowable values in 'precision' to depend upon the value of 'type'. Regards, Malcolm
Received on Friday, 11 June 2004 08:10:09 UTC