13.8.2011 0:18, Leo Broukhis wrote:
Namely, there are two characters that could be considered candidates:
no-vinculum radical symbol and radical extension.
Leo
On Fri, Aug 12, 2011 at 1:56 PM, Leo Broukhis<[email protected]> wrote:
In light of W*dings fonts being reviewed as a source of addenda to
Unicode and the reasons for disunification mentioned in N4115, I'd
like to ask if there is anything in the Symbol font that is not yet
adequately represented by encoded characters (to wit:
http://www.numericana.com/about.htm) and that could be added as
separate Unicode codepoints to put the Symbol font issue to rest.
The “Symbol font issue” is mostly a misunderstanding, by people who
found or heard of a “nice trick” of extending character repertoire in
HTML using <font face="…"> that refers to a font that arbitrary
allocates characters to positions in an 8-bit coding space. Contrary to
what Numericana claims, this was not designed or even mentioned in any
HTML specification. Rather, the Symbol font was designed, with no
connection with HTML, as a counterpart of a replaceable type ball in IBM
electric typerwriters.
The reason why <font face="Symbol"> ever “worked” in HTML documents was
the sloppy implementation of browsers—they applied a font without even
checking whether the font actually contains the characters in content.
So instead of deciding that <font face="Symbol">a</font> cannot possibly
be rendered using the Symbol font as requested, because that font does
not contain the letter “a,” they just used the font, letting it display
whatever happens to be there in the code position for “a” (Greek letter
small alpha “α” in this case). Naturally, people who relied on such
behavior got upset if the bug was fixed or e.g. if the font setting was
overridden in a style sheet.
However, it might be argued that the Symbol font has been used in text
documents (normally not plain text but text that may contain different
fonts) and that the characters so used are existing usage that needs to
be taken into account. There are two big ifs here: if this involves
symbols that do not exist as Unicode characters and if the existing
usage is relevant enough, then there might be something to be consider
for inclusion into Unicode. The burden of proof lies, of course, on both
ifs, with those who propose new characters.
You mention, and Numericana mentions, two glyphs of the Symbol font, the
one at 0xD6 (214 decimal) and the one at 0x60 (96 decimal). The former
is simply the square root sign. Contrary to what Numericana claims. the
Unicode SQUARE ROOT U+221A is not defined as having a vinculum; it may
have a small start of a vinculum, and it may not, and it mostly does not.
The Symbol font 0x60 appears to be a combining overline of prefix kind:
it combines with the following character. Such behavior is not something
that should be introduced to Unicode. Just because some people have
constructed radical expressions that look good when rendered by some
software doesn’t mean that the tricks used deserve to be defined as
Unicode characters.
It would make more sense to suggest that a vinculum character be added,
defined to be a combining character (in the Unicode sense) and being
physically a horizontal line (of the full advance width of the
character) at the same height as the leftmost point of the SQUARE ROOT
symbol. This would let people write simple radical expressions in plain
text. I don’t think this would solve much. Mathematical radicals are
essentially two-dimensional, and using vinculum that joins with the
square root character would solve just the most trivial part of the
issue. Traditional math notations require advanced typesetting and
cannot be implemented in plain text; and when one needs to linearize
such notations to plain text, radicals are among the minor issues—as √(a
+ b) with no vinculum would be quite acceptable, and so would sqrt(a + b).
--
Yucca, http://www.cs.tut.fi/~jkorpela/
