Re: Unicode 6.2 to Support the Turkish Lira Sign

From: Jukka K. Korpela <>
Date: Wed, 23 May 2012 17:51:14 +0300

2012-05-23 17:05, William_J_G Overington wrote:

> Recently, Jukka K. Korpela<> wrote:
>> It takes ten years or more, optimistically speaking, before a character added to UTC is generally available and in use. But admittedly, UTC status makes it possible to use the symbol in encoded plain text. I wonder how many databases or systems will actually be updated to reflect the adoption of a currency symbol in UTC, within a decade or two.

I now realize that I had written “UTC” when I meant “UCS”, sorry. I so
easily get messed up with TLAs. No need for any Freudian explanation. ☺

> For me, an important issue is how someone wishing to make a font with a new currency symbol is to proceed when the matter of mapping the glyph into the font arises.

Yes, that’s an important point. My remark was more like a practical
note, warning about excessive expectations. I’m afraid that if
organizations regard it as very urgent to get a currency symbol into
Unicode, they have grossly underestimated the practical problems.

A font can be designed so that the new character is first allocated a
PUA code point, later copied to a normal code point, once it has been
decided on. So accepting a character into Unicode is relevant to people
who produce data, rather than to font designers. And recoding data, by
mapping a PUA code point to an assigned code point, is as such trivial,
though it becomes a problem when there is a lot of data in different
formats using the PUA code point.

> Well, if a new currency symbol is needed quickly, wherever, and printed price labels with the new symbol printed upon them and bank account statements with the new symbol printed upon them are needed, how exactly is that to be done? I suggest that it is either by prompt action by the Unicode Technical Committee and the ISO Committee, or else some other encoding will become used, simply because there will be printed results that are needed.

Price labels, ads, and bank account statements may need new currency
symbols, but this typically involves specialized software that uses
whatever techniques are available, possibly images, rather than printing
encoded text using a font that contains the symbol. It’s unrealistic to
generally expect that fonts used in various applications will be
enhanced anytime soon.

In fixing a particular application to produce a particular currency
symbol, in place of one that has been used previously, one of the
_slowest_ ways is to have the new character has been added to Unicode,
glyphs for it have been added to the fonts used by the application, and
data has been modified to convert the old character by the new one or
code has been modified to generate the new character where it now
generates the old one.

Received on Wed May 23 2012 - 09:54:55 CDT

This archive was generated by hypermail 2.2.0 : Wed May 23 2012 - 09:54:56 CDT