W3C home > Mailing lists > Public > site-comments@w3.org > April 2002

Re: Proposing an entity

From: Masayasu Ishikawa <mimasa@w3.org>
Date: Tue, 02 Apr 2002 12:39:13 +0900 (JST)
Message-Id: <20020402.123913.41198220.mimasa@w3.org>
To: Julizhang@aol.com
Cc: site-comments@w3.org
Hello,

> Julizhang@aol.com wrote:

> > Could you tell me which organization I must contact to propose an entity?  

It depends.  Do you want to propose an entity for SGML?  XML?
For a particular markup vocabulary such as XHTML or MathML?  Or HTML?

> > What I want to propose is the superscript 4.  Now the entity for this symbol 
> > is &#8308;.  But it does no display on many browsers.  

Assuming that you are talking about HTML user agents, many newer browsers
render it correctly.  At least DocZilla RC1, IE 6.0, Mozilla 0.9.9,
Netscape 6.2.2 and Opera 6.01 worked fine.  Netscape Communicator 4.79
can render it if you use UTF-8 encoding and set an appropriate font.
Even text browsers like Lynx 2.8.4 and Links 0.1 can render it as "^4".

It is a shame that Amaya 5.3 doesn't render it correctly.  Other failed
browsers include iCab 2.71, Opera 5.12 and w3m 0.3.  Multilingual version
of w3m could render it on multilingual terminal emulator, though.

> > I am proposing it because it is frequently used on the phonetic writing 
> > system for Chinese.  Chinese is a  tone language.  It has 4 tones.  Placing a 
> > superscript after letter spellings as tone indicator is convenient.  But 
> > currently there are only sup1, sup2, sup3, but not sup4.

In the case of HTML (at least until HTML 3.2), it was because Latin
alphabet No.1 (Latin-1) had superscript 1, 2, 3, but not superscript 4
(and others).  Of course the document character set for HTML 4.0 and
later is ISO/IEC 10646, so there's no restriction to assign an entity
beyond Latin-1 block, but HTML 4.0 didn't define entities for
characters in "Superscripts and Subscripts" block (U+2070 - U+209F), 
and the W3C HTML Working Group has no plan to add any character
entities into HTML anymore, let alone they have no plan to extend
HTML itself.  They will not add new entities into XHTML, either.
The W3C Internationalization Working Group advised them to not add
new entities.

> > On the web page, 
> > superscripts can be achieved with the <sup></sup> tags.  But the tags do not 
> > work on other situations, such as in emails and BBS.  

I don't really understand your motivation here.  In situations where
markup won't work, I suppose entities won't work, either.  And where
entities work, &#8308; or &#x2074; would work more reliably than
defining a new entity.  In the case of XML, character entity references
defined in the external DTD subset won't be recognized by non-validating
XML processors, while numeric character references like &#8308; or
&#x2074; will work with all XML processors.

> > I propose to make sup4 an entity.  It will help a quarter of the world 
> > population on the Internet, word processing and other applications.

I'm afraid if you care for legacy HTML browsers, defining a new entity
won't solve your problem.  We cannot change old browsers, and even if
we add an entity like "sup4", old browsers won't recognize it anyway.
That's the same situation for the latest version of browsers, if we
add an entity, you have to wait and hope that the next generation of
browsers would support that entity, and all the current browsers won't
recognize that new entity.  While it may be a bit hard to memorize,
&#8308; or &#x2074; would work more widely across browsers, as I
showed above.

In XML (and hence in XHTML too), you don't have to ask other
organization to add your preferred entity.  You may define it
by yourself.  For example, you may define "sup4" entity in
the internal DTD subset like this:

    <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN"
    "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd"[
    <!ENTITY sup4 "&#x2074;">
    ]>
    <html xmlns="http://www.w3.org/1999/xhtml" ...>
    ...

Then you can use &sup4; in your document.  Entities defined in
the internal DTD subset will be recognized by non-validating XML
processors, too.  Of course it won't work on HTML browsers, though.

"Ian B. Jacobs" <ij@w3.org> wrote:

> Unless I'm mistaken (which is possible as I am not an
> internationalization expert), W3C has an entire Recommendation
> to meet your needs. It's called Ruby Annotation [1] and was
> published in May 2001.

It's completely irrelevant.

Regards,
-- 
Masayasu Ishikawa / mimasa@w3.org
W3C - World Wide Web Consortium
HTML Activity Lead
Received on Monday, 1 April 2002 22:39:15 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 24 October 2012 16:21:26 GMT