Re: XHTML : "attribute minimization", terseness of.

thanks for the link, Arjun :)

----------------------------------

as far as I know, "boolean" has been around a lot longer than "HTML",

 and :

<posit
   id="language::EN::semantics"
>



<Whereas>

   "Boolean" is more of a child-object of "english", the latter of which
seems to have spawned the child-object "SGML"

</Whereas>

<point>

 I don't see why "boolean" shouldn't be a possible attribute-type
    in the (container | object ) "SGML", as well.

</point>

</posit>


----------------------------------

<posit
   id="language::EN::definition.boolean"
>


<Whereas>

<quote>

    <excerpt>

Boolean

adjcective.

     1)  Of or relating to a logical combinatorial system treating
variables, such as propositions and computer logic elements, through the
operators AND, OR, NOT, and XOR: a browser that supports Boolean searches.

   2)  Of or relating to a data type or variable in a programming language
that can have one of two values, true or false.


    </excerpt>

  <source>
     <name>
The American Heritage &reg; Dictionary of the English Language,
      Third Edition,
&copy; 1996, 1992 ,  Houghton Mifflin Company.
     </name>

     <uri>
http://www.dictionary.com/cgi-bin/dict.pl?term=boolean
    </uri>
</quote>
</Whereas>

 <point>

   <by

"definition.boolean( 
               Dictionary.AmericanHeritage::word.boolean::Definition.number1 )"
    >

            <value.returned>

setting a value to "1" or "0", or any other set identifier, does seem fairly "boolean", if it's meant as part of an "if...then" sequence for a parser (human, or otherwise) to interpret.

           </value.returned>

    </by>

    <by
      id="definition::boolean( 
               Dictionary.AmericanHeritage::word.boolean::Def.number2 )"
      >
           
                <value.returned>
  
                    the quality "boolean" 
                        of
                            an attribute with possible val
ues "0" or "1"
                         if
                             "0"="false"
                                and
                             "1"="true"
                         is
                             implicit.

                </value.returned>

        </by>

   </point>

</posit>

----------------------------------

<posit
   id="XHTML::Semantics"
>

     <question>
would it be more acceptible, to whomever (people) or whatever (software or
language-specifications)  [ who | that ] have a problem with it, if we use
the word "binary" instead of "boolean" ?
     </question>

<point>

 instead of redundant declarations like

           <dl compact="compact">

    <!--
            which,
                if "compact" has to be written twice,
             does have
                  <flag redundancy="yes" />
                about it
     -->

   let's use:

        <dl compact="1">

  instead.

</point>

<optionally>
...and throw a link into the TR explaining what "binary" is about.
</optionally>

<qualifier>

though, as far as I can tell from the currently-known information,

 there should be no problem with using a "boolean" system for specifying the
attribute-values here,

      unless
         someone is simply trying to highlight a point about
           saying
               "boolean attribute"
           instead of
               "boolean attribute.value"

          ...or otherwise.

</qualifier>

</posit>



----------------------------------


<posit
   id="XHTML::extensibility"
>
    <Whereas>
       the XML specification is freely available for the public.
   </Whereas>

     <point>

if someone wants to say...

      <dl compact="compact">

... or
    if they want to say...

        <dl compact="somethingOtherwiseUnsupported">


  ...let them modify the attlist for "DL", to include "compact" as a
possible attribute-value, making their new DTD seperate and distinct from
the XHTML DTD.


     </point>
</posit>

----------------------------------

<posit
       id="Important:::work.towards::internationalization"
>

 <Whereas>
   HTML is still english-centric
 </Whereas>

    <point>

using
          <nmtoken compact="1">

      to specify the same as:

         <nmtoken compact>

 ...should help with the internationalization of XHTML.

ie:
           numbers are relatively language-neutral.


                      compact="compact"

            ...is relatively _not_ language-neutral.


       </point>
</posit>

----------------------------------




I hope that we can come to an agreement about this.



--Sean Champ






<postscript>

has the IETF done anything about, or towards, devising a form-mail for
sending to administrators of web-sites that could use a URI-tidy?

it might seem "snippish" or otherwise, but perhapps it would carry a point,
none the less.

</postscript>





----- Original Message -----
From: "Arjun Ray" <aray@q2.net>
To: <www-html@w3.org>
Sent: Thursday, March 02, 2000 8:31 AM
Subject: Re: XHTML : "attribute minimization", terseness of.


>
>
> On Thu, 2 Mar 2000, Sean Champ wrote:
>
> >   in re. section 4.5 at http://www.w3.org/TR/xhtml1/
> >
> > can we use boolean values like <dl compact="1">
> >
> > instead of redundant ones like <dl compact="compact">
>
>   1.  It's not really redundant.  The *name* didn't have to be
>       'compact'.  That is, in the "old style" <dl compact>, the
>       'compact' is the attribute *value*.
>
>   2.  There is no such thing as "boolean" in SGML/XML.  This
>       bogotic terminology came to life in the HTML 4.x specs
>       and has caused nothing but confusion ever since.  See
>
>       http://www.deja.com/=dnc/getdoc.xp?AN=580535642
>
>       and the entire thread (I tried copying the URI for that,
>       but it turned out to be hideously long.)
>
>
> Arjun
>
>

Received on Thursday, 2 March 2000 15:39:56 UTC