I am continuing my research on a 24-bit encoding system for encoding
graphics, based upon the idea of files containing codes in three-byte
sequences. I am presently naming the system the gallery system. My
approach is that I have now started writing a book which has the title The
Gallery where the gallery system is applied in various scenarios, thereby
explaining how it all works, in a hopefully very readable format.
The system is not a Unicode system, yet passes through to apply Unicode. A
gallery system code is of the format G+HHHHHH where H represents a
hexadecimal character. When HHHHHH is in the range 000000 through to 10FFFF
the gallery code represents a Unicode code. Thus a graphic defined using
the gallery system can contain descriptive text, by default in a description
area, yet, by use of gallery codes, the Unicode text can be appended at any
place in a graphic, so that, if, say, a graphic is of a clay tablet and
someone is trying to analyse what are the various cuneiform characters, then
a sequence of one or more Unicode characters can be tagged to a particular
place on the graphic, that place either being one point or the interior of a
virtual loop drawn onto the surface of the tablet, which loop can be either
visible or not visible as one chooses at any particular time.
I am now looking at two particular application areas for the system, namely
as a graphics tool to assist researchers in analysing writing which has been
written upon clay tablets and as a high performance graphics system for use
in distance education applications within telesoftware services broadcast
using the DVB-MHP (Digital Video Broadcasting - Multimedia Home Platform)
system. The main website for the DVB-MHP system is the http://www.mhp.org
website. There are further links available in one of the DVB-MHP sections
at http://www.users.globalnet.co.uk/~ngo which is our family webspace in
England.
As the DVB-MHP system uses Java, Java class files being broadcast, and Java
uses Unicode, it has become important to my research that the gallery codes
can be accessed from within a Unicode plain text file, so that gallery codes
can be delivered within a Unicode plain text file to a Java program running
in someone's television set.
Accordingly I am now defining the "gallery interface" using some of the
Unicode code points from the Private Use Area. While recognizing that this
usage of these code points from the Private Use Area is a non-exclusive
usage, I am publishing it here in order that people who like to make note of
usages of the Private Use Area may have the information and so that any
developers who are considering using the Private Use Area for making some
other definition may consider whether they wish to avoid overlapping with
this defined usage in order to keep open the possibility of using the
gallery system in conjunction with their own software systems.
A particular aspect of the DVB-MHP system which is of particular interest is
that there is a provision for having plug-ins available. I am not as yet
entirely sure how plug-ins work in the DVB-MHP system but there is a
possibility that a plug-in could be broadcast and that that plug-in could be
a plug-in that would accept a Java String and interpret the codes of the
gallery interface so as to provide a superb graphics facility for all
programs that were broadcast upon a DVB-MHP channel while the plug-in were
in place.
The gallery interface is based upon the method used for surrogates in the
Unicode specification, though with three sets of code points, namely GALLERY
INTERFACE UPPER, GALLERY INTERFACE MIDDLE and GALLERY INTERFACE LOWER each
providing 8 bits of the total 24 bits used for each gallery code point:
there is no offset constant.
The Private Use Area code point ranges that I have chosen to use are as
follows.
U+F400 through to U+F4FF GALLERY INTERFACE UPPER
U+F500 through to U+F5FF GALLERY INTERFACE MIDDLE
U+F600 through to U+F6FF GALLERY INTERFACE LOWER
I am also specifying that all three Unicode characters are to be included in
the correct order for each gallery character that it is required to produce,
just as regular Unicode requires that both surrogates are to be included in
the correct order each time.
As an example of the use of the system, please consider that it is desired
to encode the gallery command G+ABC321 into a Unicode plain text file.
The codes would be U+F4AB, U+F5C3, U+F621 in that order.
Naturally, as a method for encoding 24-bit code points this method has a
high overhead of space and is not intended to be used for stand-alone
graphics files, where the three bytes per 24-bit code point storage method
would be used. However, for including high quality graphics within text
files for special applications, the interface specified above may well be
very useful.
William Overington
16 May 2002
This archive was generated by hypermail 2.1.2 : Thu May 16 2002 - 07:39:05 EDT