Incorrect use of the word "valid' in the XSLT 2.0 WD

Section 12.1 states:

The lang attribute indicates that a collation suitable for a 
particular natural language is required. The effective value of the 
attribute must be a value that would be valid for the xml:lang 
attribute (see [XML])

In fact, the 2nd edition of the XML spec is quite clear that no 
particular value for xml:lang is either malformed or invalid, in and 
of itself. And of course in general, validity depends on what the DTD 
says.

I suggest that this section be reworded to state something like "The 
effective value of the attribute must be a language code as defined 
by he values of the attribute are language identifiers as defined by 
[IETF RFC 3066], Tags for the Identification of Languages, or its 
successor on the IETF Standards Track


Similarly

ERR77 states:

[ERR077] It is a dynamic error if a resource contains characters that 
are not valid XML characters. The processor must either signal the 
error, or must recover in an implementation-defined way; one possible 
outcome is that the processor will produce an output file that is not 
well-formed XML.

I think this should be "characters that are not valid XML 
characters". Otherwise it implies that validity is a prerequisite for 
XSLT.  The same problem reoccurrs in ERR078.

Section 14.3.2 states "if the value is not a valid QName" which 
should really be "if the value is not a well-formed QName" 14.6.4, 
ERR093, ERR098, ERR099, ERR101, and ERR106 have the identical problem.

ERR086 also overloads the definition of valid in the description of 
picture strings.

16.1.2 uses the word "validation" when what it probably means is some 
for of well-formedness checking.

Section 18 uses the word valid in the context of checking URI syntax.

In general I'd try to be very careful about using the word "valid" 
when it does not specifically reference the concept of DTD or schema 
validity. Several places in the spec use it to refer to things like 
"expressions that would have been valid under XPath 1.0". I would 
prefer "expressions that would have been syntactically correct under 
XPath" or "expressions that would have been legal under XPath 1.0" 
This definition of the word valid doesn't appear to actually be 
defined until very late in the spec J1.
-- 

+-----------------------+------------------------+-------------------+
| Elliotte Rusty Harold | elharo@metalab.unc.edu | Writer/Programmer |
+-----------------------+------------------------+-------------------+
|          The XML Bible, 2nd Edition (Hungry Minds, 2001)           |
|             http://www.cafeconleche.org/books/bible2/              |
|   http://www.amazon.com/exec/obidos/ISBN=0764547607/cafeaulaitA/   |
+----------------------------------+---------------------------------+
|  Read Cafe au Lait for Java News:  http://www.cafeaulait.org/      |
|  Read Cafe con Leche for XML News: http://www.cafeconleche.org/    |
+----------------------------------+---------------------------------+

Received on Sunday, 3 March 2002 15:49:40 UTC