Comments for PR-speech-grammar-20031218

Hello,

Congratulations on your Speech Recognition Grammar Proposed
Recommendation [1].

In 2.1, a blockquote containing the ABNF and XML follows an
unordered list. At least in my browser, that looks like the
blockquote follows logically from the last list item. (More
reasoning: blockquotes are for quotations, not for layout, and
the h4 follows logically from the h3, "2.1 Tokens".) A similar
thing happens in 2.6, where the ABNF and XML look like part of
"Special Cases." Would it be possible to bring the ABNF and XML
out to the normal left margin and indicate those sections with a
border on a div and a numbered heading? See what you think.

In the table of contents, s/4.12Tag/4.12 Tag/

In prose normative and informative can be lowercase.

In place of "--" there could be em dashes or single hyphens.

In 4.9.1, [RFC2396] without a link should read RFC 2396.

This sentence is too long:

    A Conforming Grammar Processor may be able to activate the root,
    any single public rule, or any set of public rules or roots of one
    or many grammars where each rule or root and all directly or
    indirectly referenced sub-rules contain a single language but not
    all the same language, or where at least one rule or root or at
    least on of the directly or indirectly referenced sub-rules
    contain more than one language.

And what does "contain a single language but not all the same
language" mean?

Could read (you can probably do much better):

    A Conforming Grammar Processor may activate a part (i.e., the
    root, any single public rule, or any set of public rules) of one
    or many grammars where the parts contain multiple languages, with
    one or more languages in each part.

In W3C references, usually the title comes first, then the editors and
authors. In the W3C Manual of Style there is an example citation, and a
link to the new W3C Bibliography Extractor which can automate this. In
any case, URIs shouldn't be links, and titles should be. Please see:

http://www.w3.org/2001/06/manual/#References

"Intraspect" is an unknown to me (and a product). I don't have my
big dictionary handy and don't see the word in RFC 2616. Would
introspect and introspection work?

Minor typos:
s/capabilties/capabilities/
s/identifer/identifier/
s/refererence/reference/
s/resouce/resource/
s/World-Wide Web/World Wide Web/
s/Schema/schema/
s/meta data/metadata/
s/meta-data/metadata/
s/whitespace/white space/
s/language ,/language,/
s/at least on of the/at least one of the/
s/matches the grammar,/matches the grammar./
s/uris/URIs/
s/array of 1 token/array of 1 token./
s/1 tag/1 tag./
s/tokens, tags and $NULLs is matching by input that/tokens,
  tags and $NULLs is matched by input that/ (I think)

[1] http://www.w3.org/TR/2003/PR-speech-grammar-20031218/

Best wishes for your project,
-- 
Susan Lesch           http://www.w3.org/People/Lesch/
mailto:lesch@w3.org               tel:+1.858.483.4819
World Wide Web Consortium (W3C)    http://www.w3.org/

Received on Sunday, 1 February 2004 23:35:41 UTC