W3C home > Mailing lists > Public > www-html-editor@w3.org > October to December 1997

WD HTML 4.0 (970917):typos, character entities in comments

From: Olle Jarnefors <ojarnef@nada.kth.se>
Date: Fri, 3 Oct 1997 23:50:30 +0200 (MET DST)
Message-Id: <199710032150.XAA21112@faun.nada.kth.se>
Cc: ojarnef@nada.kth.se
To: www-html-editor@w3.org
Editors,

Unfortunately I sent some of these comments to the wrong
email addresses yesterday, so you might already have seen
some of them.


Comment OJ-1
------------

I happened to notice a remaining typo in

   http://www.w3.org/TR/WD-html40/html40.txt

in section

> B.1 Notes on helping search engines index your Web site

namely

>       <META name="description" content="Idylic European vacations">
                                          ^^^^^
Should be:                                Idyllic


Comment OJ-2
------------

Here's small suggestion: The NAME values of META
elements and the names of link types now have an
unambiguous interpretation when the HEAD attribute
PROFILE indicates a URL to a suitable profile for metadata.
Why not make a similar disambiguation of CLASS names by
means of a CLASSPROFILE element in the HEAD?

Why a HEAD element rather than a HEAD ot HTML attribute?
Because one may very likely want to use CLASS names
from more than one publicly defined profile
in the same document.

How avoid possible name collisions between these
profiles? By providing two attributes in the
CLASSPROFILE element:
+  HREF, which points at a description of the class
   profile;
+  PREFIX, which defines a prefix used locally in
   this document in all CLASS names taken from this
   class profile.


Comment OJ-3
------------

For LINK elements, both the TYPE and
the CHARSET of the target resource can be specified,
for A elements only the CHARSET. Why this difference?
Shouldn't TYPE be added to A?


Comment OJ-4
------------

Typos: Section "13.2.1 Syntax of anchor names" includes
six occurrences of '>/A>'.


Comment OJ-5
------------

Another thought: Does SGML exclude that all substrings of
a comment having the form of character entity references,
are interpreted as such? This would be very useful, I think,
 making it trivial to include such things as sequences
of hyphens or indeed whole comments in the content
of a comment. This is also how e.g. Netscape and GNU-Emacs
already treat entity reference-like ingredients of comments.

Irrespective of whether this is a good suggestion or not,
I think that the HTML 4.0 spec should say _something_ about
the interpretation of entity references in comments.


Best regards,

/Olle
Received on Friday, 3 October 1997 17:50:50 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 March 2012 18:16:43 GMT