Subject: Re: [office] YEARFRAC, etc.
On Tue, 2008-04-15 at 08:47 -0400, Patrick Durusau wrote: > Andreas, > > Two replies in one. ;-) > > On my interpretation, sure, guesses for grabs as to whether it is > correct or not. I do think it is possible to state the rule without > enumerating what happens in every case. It really depends on what the > rule really is and my suggestion may be far too simple or naive. This exactly is the crux of the matter. Anybody who wants to implement YEARFRAC has to know "what the rule really is". That's all I am really asking for! > > On the power of the formula subcommittee, sure, it has no power to force > the acceptance of its work by others. > > But, realize that standards are not developed in vacuums or at least > they shouldn't be. Particularly when other standards are addressing the > same issues. One of the purposes of standards and document format > standards in particular is to promote interoperability. That goal is not > limited simply to any one standard, since interoperability with > documents that follow it should be a given. Interoperability, to the > extent possible, with other standards on the same subject is also a goal. > > Perhaps it would help to realize that a "my" standard versus "their" > standard attitude is not helpful. OpenDocument does not benefit from > another standard being wrong, having insufficient definitions, or simply > being inconsistent with OpenDocument. That sort of attitude is why we > have suffered for years from a lack of interoperability between various > applications. Please remember that I am a Gnumeric developer. I have no personal interest in ODF. (Gnumeric supports ODF only marginally while its support for MS Excel files in whatever format is likely going to be much bettter.) What I am interested in is that the openformula definitions for the various functions are mathematically correct (where this is applicable), consistent, and wherever possible (under the previous two requirements) compatible with Excel's functions. So I don't think that your expectations differ that greatly from my expectations. Having said that, I have to observe of course that either nobody knows or at least nobody is willing to tell the function definitions used in MS Excel. Since YEARFRAC seems to be a situation where Excel does not appear to follow any obvious rule and OOXML clearly doesn't define the behaviour. SO suggesting we come up with a definition that will be accepted by other standards is useless. > > Hope you are at the start of a great day! And I hope so are you! Andreas -- "Liberty consists less in acting according to one's own pleasure, than in not being subject to the will and pleasure of other people. It consists also in our not subjecting the wills of other people to our own." Rousseau Prof. Dr. Andreas J. Guelzow Dept. of Mathematical & Computing Sciences Concordia University College of Alberta