Re: [xsl] Does the count() function require access to the whole subtree?

[Michael Kay <mike@xxxxxxxxxxxx>]

> If it is, presumptively there's some XSLT 3.0 equivalent language
> already?
>

This is turning into a bit of a bicycle shed...

I think that in this discussion people may have overlooked that XSLT 3.0
doesn't use the word "overlap" as a defined technical term. It uses it only in
informal English explanations that are trying to help people understand the
more formal terminology.

The word appears in the spec 9 times, most of them in phrases such as "For
example, an implementation might be able to treat the expression .//title as
striding rather than crawling if it can establish from knowledge of the schema
that two title elements will never overlap.".

We're not talking about whether to replace a technical term in the spec. We're
talking about whether a sentence such as the above is clear to readers, and if
not, how it can be improved. I'm perfectly comfortable with a replacement such
as "... that no title element will ever contain another". But since the word
is usually used in passages that are trying to explain the technical stuff in
more accessible terms, I would prefer to avoid writing "that there will never
be two title elements T1 and T2 such that T1 is an ancestor of T2". And the
one thing we certainly don't need here is a new unfamiliar technical term:
"... that two title elements will never be co-progenitive".

Michael Kay
Saxonica