<meta content="text/html; charset=utf-8" http-equiv="Content-Type">
<body bgcolor="#FFFFFF" text="#000000">
<p>I'm unclear as to in what sense "semantic" is being used here. In
my Chambers Dictionary semantic is defined as "relating to
meaning, esp. of words". I would have thought that LaTeX macros
like \title, \chapter, \caption, and so on are semantic. Or is the
intent to specify words as nouns, verbs, adjectives, etc.? If
every word has to have a defined semantic then you are faced with
approaching an infinity of specifications. What is the semantic of
the word "taghairm"?</p>
<div class="moz-cite-prefix">On 30/08/20 17:44, Philip Taylor wrote:<br>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<div class="moz-cite-prefix">William F Hammond via texhax wrote:<br>
<blockquote type="cite" cite="mid:firstname.lastname@example.org">
<pre wrap="">Aaron Gray <a moz-do-not-send="true" class="moz-txt-link-rfc2396E" href="mailto:email@example.com"><firstname.lastname@example.org></a> writes:
<pre wrap="">Should be titled "Towards a Semantic TeX" maybe.
<pre wrap="">Isn't LaTeX already partly semantic?
It can be, if the user elects to use semantic markup.
Unfortunately, because of its design, it is quite possible to
write a valid LaTeX document that gives absolutely no clue
whatsoever as to the implied semantics. Far better, in my
experience, to express the document semantics in XML and then
write a TeX processor that can handle the chosen XML subset and
directly typeset the document from the XML source. See, for
example, <i><a moz-do-not-send="true"
the Greek manuscripts of the Lambeth Palace Library: An
exercise in transforming Excel into PDF via XML using (Plain)