<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body text="#000000" bgcolor="#FFFFFF">
<div class="moz-cite-prefix">On 06/11/2021 06:20, Don Hosek wrote:</div>
<div class="moz-cite-prefix"><br>
</div>
<blockquote type="cite"
cite="mid:E2550E4A-1C60-4578-856C-8F5848C8F3F8@gmail.com">
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
<div class="">For those who’ve missed the introduction to my finl
project (which is slowly coming along but has little practical
application yet), the half-baked original manifesto is here: <a
href="https://www.finl.xyz/2020/10/20/why-finl-a-manifesto/"
class="" moz-do-not-send="true">https://www.finl.xyz/2020/10/20/why-finl-a-manifesto/</a></div>
<div class=""><br class="">
</div>
<div class="">I’m in the midst of writing the scanner/tokenizer
right now and thinking about command arguments. Since I’m
eschewing the TeX engine on this, I can treat LaTeX-style
commands and all their various arguments as first-class
entities. This is a second-draft set of thoughts on command
parameters. Each parameter has a format and a type. Formats
indicate things like whether it’s a required argument or
optional argument, types indicate what’s in the argument (e.g.,
verbatim text, parsed tokens, math, special types like key-value
lists, etc.).</div>
<div class=""><br class="">
</div>
<div class="">I’d love to get feedback from anyone who cares to
delve into my ramblings.</div>
</blockquote>
<p>What follows is a "spur of the moment" response, Don, and is made
with the deepest respect for you as someone who has been involved
with TeX virtually since its inception (I joined the fold only in
the mid 80's).</p>
<p>My gut feeling is that any language layered on top of
TeX-the-typesetting-engine should use a syntax that is clearly
distinct from that of TeX <i>qua</i> TeX. In particular, finl
commands should <i>not</i> "[consist of a] <code>\</code>
followed by either a named command or a command symbol". Keep "\"
as the command introducer for TeX; use anything else for finl. By
so doing, not only would you allow "finl" programmers access to
any TeX primitive that they need, knowing in advance that it had
not been stolen by finl. but even more importantly you would allow
those seeking to <i>modify</i> finl code to identify, without
ambiguity, which elements are finl and which are TeX My personal
preference would be to implement finl as a dialect of XML, but
myriad other possibilities of course exist.</p>
<p>I may discover, after drinking my third coffee of the day, that
the above is complete and utter nonsense, but it was what came to
mind when I followed your hyperlink to the manifesto and some of
the further hyperlinks embedded therein.<br>
-- <br>
<i>Philip Taylor</i><br>
</p>
<p><br>
</p>
</body>
</html>