<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body>
<div class="moz-cite-prefix">On 10/29/2022 1:18 PM, Sławomir Osipiuk
wrote:<br>
</div>
<blockquote type="cite"
cite="mid:1667073947992.3769056767.4121947635@gmail.com">
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
<span class="viv-signature"></span>On Saturday, 29 October 2022,
15:43:03 (-04:00), Asmus Freytag via Unicode wrote:<br>
<br>
<blockquote style="padding-left:1ex;border-left-color:rgb(0, 0,
255);border-left-style:solid;border-left-width:2px;margin-left:0px;margin-bottom:0.8ex;margin-right:0px;margin-top:0px;">
<div class="moz-cite-prefix">According to <a
class="moz-txt-link-freetext"
href="https://en.wikipedia.org/wiki/TI_calculator_character_sets"
moz-do-not-send="true">https://en.wikipedia.org/wiki/TI_calculator_character_sets</a>
the "negation" is mapped to U+207B SUPERSCRIPT MINUS in TI
Character sets. Unless that information is definitely
incorrect, this should be the end of discussion.</div>
<div class="moz-cite-prefix"><br>
</div>
<div class="moz-cite-prefix">A./</div>
</blockquote>
<div> </div>
<div>I tried to look through the sources for that page but found
no definitive mapping. The Unicode values seem to have simply
been matched by sight by the editor. The sources contain only
bitmaps of the characters and their TI-internal byte values.
Just another reminder that Wikipedia is not always reliable.</div>
<span class="viv-signature-below"></span>
</blockquote>
<p><font face="Candara">The Wikipedia article does show a mapping.
And, no matter its origin, that mapping appears uncontested. (I
haven't looked through the page history, but that's where you
would find any disagreement on the issue; unless you can point
us to something in there, I'll assume it's uncontested; let me
know what you find).</font></p>
<p><font face="Candara">Because it's a mapping and out there,
there's now a published choice for how to represent that
character in Unicode. That fact alone changes the question from
a completely open one to one where there's a de-facto "proposed
solution". If you (or anyone) disagrees, you would have to
demonstrate why that choice is incorrect or insufficient.</font></p>
<p><font face="Candara">And, "matching by sight" isn't necessarily
an incorrect approach. Unicode distinguishes between the
identity of a character and the thing that it denotes in a
certain context --- with very deliberate exceptions.</font></p>
<p><font face="Candara">For '.', for example, the precedent is very
strong: The identity is the "period" whether used as a full stop
or decimal point, or delimiter in internet addresses or
abbreviation marker. For ':' we don't code a different
character for the use of abbreviation marker in Swedish, and so
on.<br>
</font></p>
<p><font face="Candara">For letters, on the other hand, membership
in a certain script, or having a particular case mapping can
contribute to the defining characteristics of a character's
identity, leading to disunification of otherwise identical
shapes.</font></p>
<p><font face="Candara">For dashes, Unicode considers that
differences length, and position relative to baseline or
centerline are charateristics that make two dashes distinct
symbols. However, that means that when two dashes have identical
appearance, they should not be disunified based simply on how
they are used. (The issue is a bit more complex than that,
because ASCII unifies two of them into 002D, but that's a
historical one-off, not a precedent).</font></p>
<p><font face="Candara">So, if you disagree with this mapping,
you'll have to demonstrate that there's a consistent visual
difference to the "actual" character, such that it would render
SUPERSCRIPT MINUS distinct from the unary negation. Otherwise,
the conclusion stands that there is one known convention (TI
character set) that uses SUPERSCRIPT MINUS to indicate unary
negation.</font></p>
<p><font face="Candara">A./</font></p>
<p><font face="Candara">PS: interestingly enough, one of the sources
cited for the Wikipedia article actually has a mapping to U+203E
(spacing overline). You now have two choices of "de-facto"
mappings; however, I think we can agree that U+203E seems a much
poorer match for the glyph given for negation that U+207B; the
former is at caps height, the latter between centerline and
caps. The dot matrix glyph image has the negation 1 pixel above
center. The resolution severely limits the available positions;
like the position of SUPERSCRIPT MINUS in Cambria math, the TI
negation sits on just between the centerline of superscripted
digits and their (raised) baseline. I think whoever came up with
that mapping did a better job than whoever mapped this to
U+203E.<br>
</font></p>
</body>
</html>