Re: comments on HTML+ draft RFC

"William M. Perry" <wmperry@mango.ucs.indiana.edu>
To: www-talk@nxoc01.cern.ch
Subject: Re: comments on HTML+ draft RFC 
In-reply-to: Your message of "Tue, 03 Aug 1993 09:29:38 -0400."
             <199308031329.AA19040@willow.tc.cornell.edu> 
Date: Tue, 03 Aug 1993 09:48:31 -0500
Message-id: <24339.744389311@mango.ucs.indiana.edu>
From: "William M. Perry" <wmperry@mango.ucs.indiana.edu>
Status: RO

Thus wrote: Jim Davis
[... good discussion about some HTML+ aspects deleted ...]
>Re: Tables:  Why not spell it as <table> instead of <tbl>

   Because we love TLAs. :)

>Can one use id to refer to a table?  Can table elements include an A tag?

   It doesn't say so, but I think this would be a good idea.

>The first "guideline" says: "There is no need to declare empty cells
>at the end of a row".  But this is certainly not a guideline, e.g. it
>is not something that a browser is free to ignore.  A browser which
>does so is, I think, in error.

   Well, since I just started adding HTML+ capabilities to my emacs
browser, I think I should hop in here.  :)

   If I am reading your sentence correctly, you think this means a
table could look like this:

    | HEADER  	    | HEADER	    |	HEADER	|
    | data  	    | data  	    |	    	    	<-- no 3rd column
    | data  	    | data  	    | Data  	|
    [...]

    If so, then I think you read the HTML+ spec wrong - there is no
need for the _author_ to specify every column of data.  I think it
would be silly for browser writers not to fill it in.  Although I hope
WYSIWYG editors won't leave too much of this out.

>Can a table have more than one <tt>?  Must all <tt> be before all
><th>?

   Good question.  I've got table formatting done, in a very basic
way, and I assumed there could be only one <TT> - didn't matter where
it was.

>It seems that the designer of this feature must have intended some
>semantics which the example hints at but the spec does not mention.
>Note that the first two lines of the table have only three entries
>(line 1 has two entries, the first spanning two columns), yet they
>begin in column 2, not column 1.  I am just guessing, but perhaps the
>designer intended or assumed that one column would be reserved on
>these two "header" lines, to be used below by the row headers.  As a
>further guess, the table's "data" begins with the first line that
>contains a <td> instead of just a <th>?  And if there's a <th> on
>such a line, it's a row header, and goes into this reserved first
>column?

    If you check the tables source, you'll notice that there are empty
<TH> tags at the start of those lines.  So there is really a ' ' in
the empty header. :)

    Right now, when a table is formatted, I threw in support for a
"s=##" tag for <TD> elements also.  (Whole 3 lines of code :)  Anyone
think this should be in the spec?  Not sure when it would be useful,
but you never know.

>Maybe instead of using <th> for both the column headers (in this
>case, height, weight, category) and row headers (males, females) we
>should have separate tags?

    The "male" and "female" labels are just data elements, and look
like row headings because of the blank '<TH>' above them.  I think
this is the way to go.  Although right now the way I ave it written
you could do something like:

<TH> Some header <TD> blah <TD> blah <TR>

and it would turn out ok.  Is this going to be legal HTML+?

    I had a few questions about the HTML+ spec too, specifically with
tables.

Are we going to let people put tags in a table?  I would vote against
it, just so my life could be easier, since I don't use the common code
library.  Imagine a table that looks like this:

    +---------------+--------------+------------+--------------+
    | 	    	    | 	    	Average	    	| Other	       |
    +---------------+--------------+------------+--------------+
    | 	    	    | height	   | weight    	| Category     |
    +---------------+--------------+------------+--------------+
    | males 	    | 1.9   	   | 0.003      | <A HREF=     |
    | 	    	    |	    	   |	    	| "some.html"  |
    |	    	    |	    	   |	    	| >Text</A>    |
    +---------------+--------------+------------+--------------+
    | females       | 1.7   	   | 0.002      | yyy          |
    +---------------+--------------+------------+--------------+

    In my case, this is what it would look like after formatting the
table, but nothing else.  Now, say I go to parse the links - no way.
The link text has to be consecutive - it can be across 500 lines of
text if need be, but it should be 
<A HREF=
"some.html"
>
Text</A>, whith nothing between the = and " but whitespace and/or
newlines.  (Same with " and >, > and Text, etc)  Even if I could parse
that out, the depth of the table would change from 3 lines for males
to 1 line for males, and then you get into lots of nasty stuff like
checking 2 or 3 times to see if the table has changed size (after
making headings, after making links, after making lists, etc)  Makes
me wish emacs had a 'narrow-to-rectangle' function, then it wouldn't
be a problem. :)

    I'm not sure if anyone else out there is not able to use the
common code parser, but if so, please chime in.  I think it would be
more appropriate to have the link above replaced with something after
the table like: More information on <A HREF="some.html">Males data</A>
is available.  (Worded better, but similar :)

    Any comments?
    	Bill P.