Fwd: Michael Brauer, regarding "Proposal: Identifying nonstandardfunctions"

["David A. Wheeler" <dwheeler@dwheeler.com>]

Michael Brauer said the following in the "office" list.
I'm forwarding it here to the "office-formula" list,
so that those who are only on the "office-formula" list will see it.
It's about how to identify application-specific functions.

----- Forwarded Message -----

I'm replying to the office list rather than the office-formula list, because 
I'm not a member of the Formula SC.

robert_weir@us.ibm.com wrote:
> 
> I don't think we want to get into the registrar business.  A vendor can 
> figure out how to create a unique function name if they want to create 
> such an extension, whether 123.DAYS, or 123_DAYS, or whatever.  If we 
> want to recommend a mechanism, then that's fine.  But creating a 
> registry is overkill.  As you said, the number of applications is 
> relatively small.

I agree to Rob. I think a registration business is outside the scope of the 
OpenDocument TC, and I believe also outside the scope of OASIS.

I further think we have to differ between using XML namespaces for the 
identification of extensions/implementations, and the use of 
[prefix]:[local-name] syntax of XML namespaces. If it is not possible or 
reasonable to use a colon within formulas (and I have no doubts that this is 
the case), then I think it is better to use a dot instead and to keep the 
other semantics of XML namespaces, then it would be to define a new 
extension/identification method.

Michael



----- End Forwarded Message -----