>The problem (if there is one!) is only for font technology.
>
>> Ethiopian writing: [...] "The capability to the same electronically
>> would be well received. /Daniel."
>
>Same for this one: Unicode's task was to provide a code point for the
>Ethiopic full stop, and they did. Whether the corresponding glyph is
colored
>or not is problem for fonts and word processors.
Well, may I please suggest that the issue is one for Unicode as well as for
font technology?
Firstly, for the avoidance of doubt in the matter, whereas I am an advocate
for adding codes into Unicode for effects for organizing and controlling
data in ways which some people consider should be done only by markup
methods, I am hoping that, without that aspect of my research prejudicing
the matter, readers might consider the possibility of adding into regular
Unicode some operators for use in ZWJ sequences for requesting that a
chromatically decorated glyph of the 'operated upon' regular Unicode
character be produced if the font can provide it, yet otherwise that a
monochrome ordinary glyph be provided.
May I please refer to the following document.
http://www.users.globalnet.co.uk/~ngo/courtfor.htm
In that document I wrote as follows.
quote
Here are some codes for use in ZWJ sequences of the form U+HHHH ZWJ U+F3DC
and U+HHHH ZWJ U+F3DD so as to provide facilities to indicate to a chromatic
font that a colour decorated version of U+HHHH is requested, where U+HHHH
represents any Unicode character where such usage would be meaningful. This
facility is provided in anticipation of the possibility of chromatic fonts
being introduced at some time in the future.
U+F3DC ZERO WIDTH DECORATION OPERATOR OF THE FIRST KIND
U+F3DD ZERO WIDTH DECORATION OPERATOR OF THE SECOND KIND
end quote
May I please refer to the following document.
http://www.users.globalnet.co.uk/~ngo/courtcol.htm
In that document I wrote as follows.
quote
U+F3E0 BLACK
U+F3E1 BROWN
U+F3E2 RED
U+F3E3 ORANGE
U+F3E4 YELLOW
end quote
So, it would be the case that in order to set some text in black one would
use U+F3E0 then the text and in order to set some text in red one would use
U+F3E2 then the text.
In order to set some text in black including a character U+1362 in black
with red flourishes one would use U+F3E0 then the text which precedes the
U+1362 character and then U+1362 ZWJ U+F3DC which should do the job
perfectly well as the chromatic font would be set up so that the decoration
of the first kind operator worked in black and red.
More generally, for other chromatic characters from other applications where
the colours are not specific, then the chromatic colours can be changed
before using the ZWJ sequence.
Quoting from the same document.
quote
Colour changing is by a specially devised method which will hopefully be
efficient in practice.
Upon receiving one of the 18 codes to change colour, the system presumes on
a temporary basis that the new colour is to become the foreground colour,
which is what it will usually be. However, the previous foreground colour is
stored. If a command to set one of the other four colours is received, then
the foreground colour is used for that purpose, with the foreground colour
being replaced by the previous foreground colour.
This means that only one code point is needed to change the foreground
colour and two code points are needed to change the contents of any of the
other colour registers.
The decoration colours are intended to be ready for the possible
introduction of chromatic fonts at some future date.
U+F3AC SET NEW BACKGROUND COLOUR
U+F3AD SET NEW FIRST DECORATION COLOUR
U+F3AE SET NEW SECOND DECORATION COLOUR
U+F3AF SET NEW THIRD DECORATION COLOUR
end quote
In using chromatic font technology I suggest that the specific colours could
either be built into the glyph or could just be foreground colour,
background colour, first decoration colour, second decoration colour, third
decoration colour, in abstract terms with the specific colours being
supplied by the rendering software. This would enable some glyphs, such as
those for the Ethiopic manuscripts, to be specified as black and red within
the font, and for some glyphs, such as those for an ornament, to be
specified by the rendering software.
As regards the possibility of including such code points as I have mentioned
above in regular Unicode, well, there are various levels.
There is the issue of whether codes such as U+F3E2 RED above should be
promoted as some people feel that they are markup and should not be included
in regular Unicode. Well, that is an issue and I quite accept that I am
currently in the minority over that issue, though I would ask that readers
might look into the matter of the use of such codes in DVB-MHP (Digital
Video Broadcasting - Multimedia Home Platform) broadcasting
of multimedia where text in Unicode files will be processed by Java programs
which have been broadcast to television sets, before taking a definitive
view on the matter, and indeed at the possibility of their use in general
desktop publishing packages.
However, I wonder if, even if they are against characters such as U+F3E2
being promoted to regular Unicode, readers might have a look at whether
characters such as U+F3DC and U+F3DD should be promoted to regular Unicode
as they would be used, as operators, in a ZWJ sequence for the character
which is to be decorated.
An index to the documents, and some other documents and to background notes
is at the following address.
http://www.users.globalnet.co.uk/~ngo/court000.htm
The main index page of the webspace is as follows. There are various items
in that webspace which will hopefully be of interest. For example, some
documents about the DVB-MHP broadcasting system, including some links, some
artistic fonts produced using The Alphabet Synthesis Machine and some
articles about using the Microsoft Paint program for such purposes as
drawing Bézier Curves, for precisely locating text in diagrams and for using
Paint in conjunction with WordPad to produce graphics of large letters,
filled with gold texture, together with some support files of graphics for
use in Paint, such as some stars and some polygons: and then there is 1456
object code as well.
http://www.users.globalnet.co.uk/~ngo
William Overington
6 July 2002
This archive was generated by hypermail 2.1.2 : Sat Jul 06 2002 - 05:27:31 EDT