Re: [xml-dev] Why is terseness of minimal importance?

["Liam R. E. Quin" <liam@fromoldbooks.org>]

On Thu, 2022-01-13 at 17:53 +0100, Marcus Reichardt wrote:
> [...]

> Though I find the feature Michael mentioned - requiring end-element
> tags to contain the end-element rather than just allowing "</>" as in
> SGML - to be particularly redundant in a markup language not having
> CONCUR or other overlapping-markup feature.

Those of us supporting users of SGML systems were finding that
minimization, including </>, had a high cost, because users would think
the document's interpretation was different than the parser's.  Error
checking is not primarily for the people with a high awareness and
care, but for the other 99% of us :) :)

At one point i said i wished we could write, instead of, e.g.
   </div></div></div></div></div>

the less terse
   <div class="wrapper"><div id="main"></div class="page"></div>

i.e. we could allow repeated attributes with the proviso they must
match the corresponding attribute in the start tag. This would be like
the practice in C of

#ifdef SOME_FEATURE
    .. . . code for implementing SOME_FEATURE, possibly
           several pages long...
#endif SOME_FEATURE
where the text after #endif was originally simply ignored by the
compiler (preprocessor) and later, compilers started to have options to
check it.

This idea was ridiculed at the time, however.

Now, seeing 50 or even 100 </div> tags in a row in an HTML page isn't
that wildly unusual.


-- 
Liam Quin, https://www.delightfulcomputing.com/
Available for XML/Document/Information Architecture/XSLT/
XSL/XQuery/Web/Text Processing/A11Y training, work & consulting.
Barefoot Web-slave, antique illustrations:  http://www.fromoldbooks.org