Some Unicode text files for customizing an interactive multimedia DVB-MHP application which customization includes a picture. This file is not a Unicode text file, it is an ordinary 8-bit text file, provided as documentation of the experiment. The Unicode text files of the experiment are the unitest6.txt file and the unitest6.uof file. The experiment also uses the pngtest2.png file. The experiment is a supplement to the experiment described in the file text7005.txt. Suppose that a generic DVB-MHP Java application package becomes available which acts upon the Private Use Area code points introduced in this series of documents to produce interactive multimedia displays. Suppose that the program is named, say, genericm, for example. Then, if customized with the Unicode text file unitest4.txt the program might be included in the menu items available in a DVB-MHP channel by a broadcaster using a line such as follows, or something similar, I am unaware of the details at present. genericm unitest4.txt In a similar manner if customized with the file Unicode text file unitest5.txt the program might be included in the menu items available in a DVB-MHP channel by a broadcaster using a line such as follows. genericm unitest5.txt The genericm program could be used to produce a large number of seemingly different applications as viewed by the end user, though in fact it would be one application, genericm, being run using a large number of different Unicode text files to customize it. Suppose however that use of the technique suggested in the following web page were built into the genericm Java application. http://www.users.globalnet.co.uk/~ngo/ast03200.htm With a .txt Unicode text file used as the customizing file the action would be as mentioned above, yet if a .uof file were used as the customizing file, then a different effect could be produced, by the generic program acting somewhat differently upon detecting the use of a .uof file. The line used by the broadcaster, using the .uof file from this experiment, might be as follows. genericm unitest6.uof In this experiment the unitest6.uof file, a Unicode text file, contains two lines of text, which are transcribed here. unitest6.txt pngtest2.png The idea is that in this experiment, the file unitest6.txt contains one OBJECT REPLACEMENT CHARACTER SYNONYM character and that object is the image in the pngtest2.png file. In fact, in this experiment I have used U+EBCD, decimal 60365, OBJECT REPLACEMENT CHARACTER SYNONYM CENTRE ALIGNMENT as the OBJECT REPLACEMENT CHARACTER SYNONYM character so that the picture would be horizontally centred. The picture would not be vertically centred but would have its top at the place where the OBJECT REPLACEMENT CHARACTER SYNONYM character would display had it been an ordinary text character and not an OBJECT REPLACEMENT CHARACTER SYNONYM character. If there were more than one picture file referenced by the unitest6.txt file then they would be referenced in order in the unitest6.uof file after the file name unitest6.txt, each on its own line in the file. However, this experiment is to try to establish the technique, so only one picture file is used. I am hoping that someone who is able to produce DVB-MHP Java application programs might like to write the program and try it out. If anyone does do that it would be great if the results were made public. Although the web page referenced above refers to the regular Unicode item U+FFFC OBJECT REPLACEMENT CHARACTER code, when using a program such as WordPad to prepare a file of multimedia content a problem arises over including a U+FFFC character in the document. This is because WordPad has its own facilities for including pictures and direct including of U+FFFC into a WordPad document using Alt 65532 is not possible. This is entirely reasonable of the WordPad program, as WordPad is a general computing program and using it for authoring content for the DVB-MHP platform is but one of many uses for which people may wish to use the WordPad program, yet the situation does prevent use of the WordPad for placing a U+FFFC character into a file for this present multimedia authoring purpose. In view of this, I decided to define U+EBCF to be OBJECT REPLACEMENT CHARACTER SYNONYM and ask that anyone programming picture handling facilities into a multimedia system based on the contents of this present document please consider making U+EBCF and U+FFFC have identical meanings. That will mean that a content author using WordPad on a PC can produce usable content files directly, whereas someone using SC UniPad or some package designed specifically for multimedia authoring using this system, can use U+FFFC directly. Upon considering the matter further I decided to add three more Object Replacement Characters, the three new characters having information about alignment of the picture included in their definitions. The list is thus as follows. U+EBCC, decimal 60364, OBJECT REPLACEMENT CHARACTER SYNONYM LEFT ALIGNMENT U+EBCD, decimal 60365, OBJECT REPLACEMENT CHARACTER SYNONYM CENTRE ALIGNMENT U+EBCE, decimal 60366, OBJECT REPLACEMENT CHARACTER SYNONYM RIGHT ALIGNMENT U+EBCF, decimal 60367, OBJECT REPLACEMENT CHARACTER SYNONYM I have designed a set of glyphs for these characters based around a thick circle. The first three have a vertical line upon the left, the centre and the right of the thick circle respectively. The fourth has no vertical line. The files unitest6.txt and unitest6.uof can conveniently be displayed on a PC using the Microsoft WordPad program provided that the version of WordPad can support a Unicode text file. The file unitest6.txt was prepared on a Windows 98 PC system. Using the Quest text font, (at least version 1.08 is desirable as it covers all of the codes introduced in this series of documents), allows the authoring-time symbols for these Private Use Area codes to be displayed as symbols. I find that using 12 point Quest text in WordPad is a suitable size as the Quest text font is designed to be clear at 12 point particularly. The font is available for download from the following web page. http://www.users.globalnet.co.uk/~ngo/font7007.htm Please note that these particular assignments of these Private Use Area codes are produced by the present author. Other researchers in other contexts may choose to assign them differently. However, if people do choose to use the codes which I have specified in relation to DVB-MHP broadcasting then that is fine and will help interoperability. Please note also that these Private Use Area code meanings have no status within the DVB-MHP specification. They only have validity in those application programs which recognize their use within that program. In preparing the .uof file using WordPad there have been some minor problems over a Unicode text file having a .uof designation rather than a .txt designation. However, this has not caused too much difficulty. In order to use pngtest2.png in a broadcast application it may be necessary for specialist modification of the file so as to make it broadcast standard. Hopefully this experiment will be of interest to readers and will be a catalyst towards progress. William Overington 30 May 2003