Re: SVG12: microsyntaces

* Andrew Emmons wrote:
>We have reviewed the use of various value types and have added
>definitions where appropriate. Thank you very much for your review,
>please let us know within two weeks if this does not satisfy your
>concern.

In http://lists.w3.org/Archives/Public/www-svg/2005Apr/0156 I proposed
steps to ensure that the draft is both complete and accessible in this
regard. The Working Group originally agreed, and then rejected many of
the steps I've proposed, claiming "it's an unnecessary burden on the
specification editors". I checked this again based on your claim and
already the first example I checked has serious issues in this regard.

Section 5.10.1 defines the class attribute as

  class = <list-of-names>
    This attribute indicates membership in one or more sets. Multiple
    set names must be separated by white space characters.

"<list-of-names>" is not a link, but there is "<list of xxx>" in 4.1
that might be relevant to thise case even though "<list of ...>" and
"<list-of-...>" are quite different constructs. It can then be argued
that this is supposed to be a list of "names" or "name". However, I
could not find a definition of either symbol or type in 4.1. I then
checked the RelaxNG schema and found that it just specifies that any
arbitrary string is accepted. The class attribute in other formats is
constrained e.g. to NMTOKENS as defined in XML 1.0. It is unclear to
me whether this is also the case here, or, if not, what the actual re-
quirements are, and why there are differences between W3C formats.

If you read one line further, you find

  id = "name"

We note that "name" is not <name>, and "name" does not seem to be de-
fined anywhere either, certainly not in 4.1 or 5.10.1. I have pointed
out other problems elsewhere. These problems continue to exist all
over the draft; it is unclear to me why anyone would consider the issue
to be resolved. It is not. If you do as I proposed, starting with making
all references to types such as <list-of-names> above a link to the
actual definition of the type, you can slowly get there. SVG Tiny 1.2
requires implementations to ignore, among other things, incorrectly
specified attribute values, without actually knowing what correctly
specified attribute values would be, that is obviously impossible!
-- 
Björn Höhrmann · mailto:bjoern@hoehrmann.de · http://bjoern.hoehrmann.de
Weinh. Str. 22 · Telefon: +49(0)621/4309674 · http://www.bjoernsworld.de
68309 Mannheim · PGP Pub. KeyID: 0xA4357E78 · http://www.websitedev.de/

Received on Sunday, 23 July 2006 02:36:46 UTC