Re: [office-formula] Proposal: Drop "huge" group; any critically missingfunctions for "large"?

A super set of Excel and OpenOffice
2.0x is fine for me.  A clarification though -- are the Excel functions
included in "large" jut the core Excel ones, or do they also
include the ones only available if you load the data analysis add-in?

Also, has anyone looked at Excel 2007
beta to see if the function set has been extended?

-Rob

"David A. Wheeler" <>
wrote on 07/26/2006 11:12:03 PM:

> We have our hands full just defining the set of

> functions up through the "large" group, which is essentially

> a superset of Excel and OpenOffice.org.  Thus, I propose that

> we give up trying to define the "huge" group of functions
at

> this time; if it's deemed worthwhile, that would be a reasonable

> "next version" effort. The "large" group is enough
to capture

> most spreadsheet documents, no matter where they come from,

> and I think that's really what people want.

> 

> Are there any "gotta have it" functions _not_ currently

> planned for the "large" group, particularly those that are

> painful to do without and have REAL users?  I'm basically

> asking those who use applications other than Excel and

> OpenOffice.org; we have pretty complete coverage of them.

> I do NOT want a laundry list of 200 functions to add,

> but adding a FEW functions to help people transition

> their data from other apps would be reasonable.

> 

> Here are some candidates that I see:

> XOR(Logical+)

> GETENV

> ISLEAPYEAR

> ISBETWEEN

> EASTERSUNDAY

> ISPRIME

> BITAND, BITOR, BITXOR, BITLSHIFT, BITRSHIFT; (BITNEGATE w/size param?)

>    (issue: how many bits are enough?)

> 

> Please speak up if these or other functions are

> "gotta haves as interoperable functions, and not in the spec".

> Any application can store the function using the local namespace

> (see the separate discussion about that), so any application

> can continue to use the other functions it has, and define new ones.

> The issue here is, is there another formula function so

> important for interoperability that we need to spec

> it NOW as a standard function?  Otherwise, it

> won't be in the first release of this specification.

> 

> Thanks.

> 

> --- David A. Wheeler

>