When you want to make a text available on the Internet, there are many assumptions that have to be met in order to succesfully do this. A written text makes use of a script, this means that a set of associated characters combine to make up words and numbers. All these characters have to exist in a font when you want to express them digitally and, this font has to be shared by the people who will eventually have the text expressed on their monitor or printer.
Even when the characters are properly expressed, it is not assured that the computer accepts the assertion that the text is in a specific language. The software has to be aware of all kinds of attributes before it accepts the existence of a language. Open Office for instance requires a "locale" as a precondition. A locale defines attributes like what script is used, how to express a date or number and a sorting order.
All the attributes that allow for the digital existence of a language can be realised by collecting the underlying data and express them in applicable standards. The current model for standards is based on for profit industry involvement. Supporting additional languages is expensive and is used as an argument to delay and prevent the evolution of applicable standards.