This font format is an example of a bad business strategy. In order to make it possible to set type in the languages of East Asia without extending the Type 3 or Type 1 fonts to support more than 256 glyphs, Adobe decided to use hierarchical fonts with several levels, in which references to characters are dispatched by a central file to subfonts, which ultimately yield very ordinary Type 3 or Type 1 fonts. These fonts were called "composite fonts", and the entry for the font type in the FontInfo dictionary took the value 0. Apparently Adobe did not immediately release precise and comprehensive specifications; as a result, several Asian companies tried to "guess" the specifications (through reverse engineering) and delivered fonts that were more or less similar but not always compatible. Adobe subsequently developed a more powerful font format, known as CID, and immediately declared the Type 0 fonts null and void. The "composite" fonts were then given a new name: OCF (for "original composite fonts"), where "original" was a synonym for "obsolete". To this day, Adobe has never stopped denigrating OCF and singing the praises of CID. Most Asian foundries issue their fonts in both OCF and CID. (One wonders why, if CID is so much better.)
Now let us examine the description of OCF, based on a bit of information published by Adobe.
We have already said that OCF fonts are those whose FontType entry takes the value 0.
The first problem that arises concerns the ...