<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body>
<div class="moz-cite-prefix">On 1/9/2021 3:30 PM, Richard Wordingham
via Unicode wrote:<br>
</div>
<blockquote type="cite" cite="mid:20210109233026.72168354@JRWUBU2">
<pre class="moz-quote-pre" wrap="">On Sat, 9 Jan 2021 14:14:42 -0800
Asmus Freytag via Unicode <a class="moz-txt-link-rfc2396E" href="mailto:unicode@unicode.org"><unicode@unicode.org></a> wrote:
</pre>
<blockquote type="cite">
<blockquote type="cite">
<pre class="moz-quote-pre" wrap="">Finally, the letter should be encoded as precomposed only, to avoid
the issues we've had for other characters where the "nominal"
diacritic indicated in the decomposition would force a shape that's
not compatible with the range of allographs.
</pre>
</blockquote>
<pre class="moz-quote-pre" wrap="">PS: some of the allographs for the diacritic strike me as something
other than
an "over-strike". The attached two sections or the partial vertical
over-strike
of the top of the bowl only would normally not be decomposed,
because they are more like an "attached" diacritic than an overlay.
This further argues for not decomposing this character.
</pre>
</blockquote>
<pre class="moz-quote-pre" wrap="">
This is very like the slash in the dollar sign, which may likewise
be reduced to projections above and below. I don't think there is any
doubt that this Polish character should be encoded as an indecomposable
character; the question is whether it has already been encoded.
</pre>
</blockquote>
<p>It is very much unlike the case for $. For that letter, as
originally encoded<br>
everywhere, there are a number of accepted allographs. The
encoding<br>
covers all of them. (And none are decomposable).<br>
<br>
The Polish letter has a range of allographs that include shapes
used<br>
for other letters. Those letters do not have any such allographs.
(And<br>
some of the allographs shown in the proposal match different
letters<br>
that are well distinct from each other as encoded today).<br>
<br>
In my reading, the range for a letter's allographs can (and
should) be<br>
one of the factors that determines its identity. Part of that
evaluation<br>
needs to also focus on what is the most common (modern) allograph.<br>
<br>
In the current instance I see nothing that would support a
unification<br>
to an existing character, unless Unicode were to resurrect the
discredited<br>
concept of "arms-length unification", or unification based on
superficial <br>
similarities.</p>
<p>A./<br>
</p>
</body>
</html>