The existence of various writing systems would mean that there have to be standardized ways of representing geographical names originating in languages with other writing systems. It would be technically feasible to standardize names within a group of languages using the same writing system, e.g. those using Roman script. But for names from languages outside the group there are often different methods of representing them in Roman letters. One and the same non-Roman name might be rendered differently, depending on the target language, tradition and purpose of the text. A single Russian place name may serve as a good example of confusion created by various language-oriented renderings:

What happens with language-oriented conversions?

Although the name contains only five sounds, they have in this table 17 different graphic representations in Roman letters. The phoneme [] is represented by eight letters or letter combinations (ch, s, š, s, sch, sh, sj, sz), [] by three (ch, h, kh) and [] by four (i, i, õ,y).

What happens with method-oriented conversions?

Apart from country-oriented conversion methods, there are also different conversion methods that don't target specific countries, but are based on specific scientific viewpoints or institutional traditions.

For the conversion of Arabic names, for instance, there are systems developed by ISO (233), by UNGEGN, by DIN and others. Wikipedia gives an overview (see Romanization of Arabic).

The existence of different methodological systems also would result in different name versions in the Roman alphabet for the same Arab name:

What happens with method-oriented conversions?* - continued)

So, even if these systems are not language-oriented, there still are sizable differences.