Patrick Durusau:
> This is where I disagree, not that examples aren't helpful but that they
> are necessary in the normative text.
They wouldn't be normative, they'd be adjacent to the normative text.
> There may be a very limited number
> of cases and the directives so concede where examples may be necessary.
>
> HOWEVER, it is clear from your response that you are taking this as an
> opportunity to insist on having examples for every function. I really
> don't think anyone needs an example of the "+" function, for example. Or
> a large number of others, assuming that we do our job correctly as a
> committee and write really good prose and/or notation.
Ah, so THAT'S the issue, you don't want to see them with EVERY function.
We _could_ expose examples on only selected functions.
The trick would be deciding which ones, then. Sadly, making that
decision on a case-by-case basis would obviously take additional time;
I don't know if we have that kind of time.
I want the spec to be superior to OXML in every way (a very low bar indeed).
The only thing I can see that's better about OXML is that it includes examples.
We HAVE them, but we have removed ours from view, because we believed
that ISO wouldn't accept a spec with lots of examples. Clearly that's not so.
Indeed, why _is_ OXML acceptable? It includes function examples.
> The point being that we could easily see a 1,500 to 2,000 page ODF
> standard that is mostly *non-normative* examples.
It's not too bad with the OpenFormula spec; even displaying all hidden text it's
not THAT many pages.
--- David A. Wheeler
