We have set an application to edit and publish online educational contents. The editors create the content in XML with Arbortext 6.0 and then through some XSL and XSL-FO it gets online.
It happens that some of the online students use tablets to read the contents, so we have to be very careful as to which characters are used in the documents, since some tablets don't have the fonts to display them. Notwithstandig, some fonts are needed even though they are not presents in all tablets, so we have created some webfonts for, let's say, some mathematical characters like the R for real numbers, or the square root. The editors write the UTF-8 character in the XML document, and then, through the XSL we apply to them a webfont.
The problem is that we cannot create the total arrary of UTF-8 in webfonts. We use just around 40. So whenever an editor uses a new character not part of the 40, it doesn't display in some tablets.
So far if we set the XML to ISO-8859-1 instead of UTF-8, it has no impact, since the chracters get in the XML in haxadecimal form intstead of as two bytes characters, which is of no help at all...
Any idea of how to control the set of characters used?
DeTeDe, XML - Editorial