This is an archived snapshot of W3C's public bugzilla bug tracker, decommissioned in April 2019. Please see the home page for more details.
On the xml-schema-dev list today, Daniel Veillard [daniel@veillard.com] raises the point that "finite-length" sometimes seems to mean "zero or more", and sometimes "one or more" (as in "an optional sign followed by a finite-length sequence of digits". I've raised objections to "finite-length" before on the grounds that processors cannot reasonably be expected to distinguish finite from infinite sequences. But Daniel's objection seems less frivolous. I would respectfully suggest that we use the more conventional qualifiers "zero or more" and "one or more" as appropriate.
The working group discussed this issue today and agreed that it's a substantive issue that needs to be clarified. We instructed the editors to prepare a draft making clear, wherever "finite-length" is used, whether it means "zero or more" or "one or more".
A wording proposal intended to resolve this issue is at http://www.w3.org/XML/Group/2004/06/xmlschema-2/datatypes.b5625.html (emmber-only link). The proposal inserts appropriate qualifiers ("empty", "non-empty", "zero or more", "one or more", etc.) in sentences otherwise imprecise on the lower bound of acceptable lengths.
The proposal mentioned in comment #2 was adopted on today's WG call. With that, the WG believes the issue to have been resolved. Michael, you may or may not wish to convey this information to Daniel Veillard, or to take responsibility for closing the bug yourself; your call.