RE: [office] YEARFRAC, etc.

["Mason, James David (MXM)" <masonjd@y12.doe.gov>]

>I really don't see any alternative to following the "Excel
non-definition" however it gets defined. 
>In some ways that is more typical of standards, that is they codify
existing behavior or definitions. 
>And that is what users are going to be expecting. Not to mention it
will be necessary for interoperability. 


I think Patrick has the right idea here. It's very rare to have the sort
of opportunity that SC34 had 20+ years ago when we designed SGML on a
blank sheet of paper and got it approved. ISO generally expects
standards to do what Patrick describes: "codify existing behavior or
definitions." 

(And to be a bit more precise: SGML wasn't really a blank sheet of paper
to those of us who were working on it, though it looked like that to
ISO. Most of us on the committee were already using some sort of generic
coding, and Charles Goldfarb particularly wanted to build on his
existing base of GML as it was used at IBM. Indeed, you might say that
IBM played the role that MS is being accused of here: there were more
people with an investment in GML than there were in, for example troff
[my generic coding environment], so SGML came out doing things in
GML-like ways, even if the syntax was something new.)

Jim Mason