Re: help: ci type='fn', plus

A number of examples in the current draft make use of type="fn".
All these should be read as  type="function".  But note that
type="fn" remains perfectly legal since any CDATA string
can be a type value.   Apart from this there is no need to
change these examples.

As for the n-ary operators, the current internal draft - soon
to be made public -- treats n-ary operators more uniformly.
Chapter 4  shows plus is n-ary and can have 0 or more terms. 
Appendix C talks about sums with 0 and 1 term and more
terms.

Stan Devitt
Math Working Group

Romeo Anghelache wrote:

>
> Hello,
> I am a bit confused on the following issues and need some help:
>
> 1. Section 4.4.2.1 (apply) paragraph #4 states "if the object being 
> applied is not already one of the elements known...it is treated as if 
> it were the content of an fn element"; but fn is deprecated in favor 
> of csymbol; then
> writing f(x) as
> <apply>
> <ci type="fn">f</ci>...
> is deprecated too, in my understanding (this example, <ci type="fn"> 
> appears in a lot of places in the current spec)
>
> then what should one write, instead,
> <apply>
> <csymbol>f</csymbol> ?
>
> (If yes, then also the quoted phrase should be modified accordingly.)
>
> 2.Paragraph 3, same section, says 'plus' has 0 or more arguments.
> But <plus/> is an n-ary operator, and section 4.2.3.1 defines n-ary 
> operators as having at least 2 arguments. Which one is right?
>
>
> Thanks,
> romeo
>

Received on Wednesday, 23 July 2003 15:34:41 UTC