XBOX 360 2560

But there are hundreds of thousands of companies around the world, and each company may have several different logos. There's not enough space for all of them unless you resign yourself to a much larger code space. If you only allow the biggest ones to have codepoints in the shared area, you need some political process for deciding which ones need it, and then you need a private use area for all the littler ones that don't make the cut.

Yes.

Storing letter plus accent as one codepoint seems seductive if you only look at major western European languages and that's what ISO did and what most font makers did. But it starts breaking down when you get to non-European languages written in Roman script. Vietnamese is the clbuttic example, with up to three diacritical marks on each letter and almost every letter having some. Storing each possible combination of letter + diacritical marks would get into the thousands. I think even Unicode decided to store them as combining characters in the case of Vietnamese -- so the rendering software has to be able to cope with combining characters. They might have saved themselves a lot of codepoints and had a cleaner design by using them consistently for all languages written with the Roman alphabet.

That's what libraries did with their character set, Extended Latin Alphabet Coded Character Set for Bibliographic Use (ANSEL) (ANSI Z39.47). It uses ASCII in the lower 128 positions and combining diacritical marks and a few special characters in the upper 128 positions. It's sufficient to represent all languages that use the Roman alphabet and the Romanized forms of all other languages. Most of what became the ANSI standard was already in widespread use in the library community by the 1960s. It was used for catalog card production before there were online catalog displays.

It boggles the mind that ISO came up with their nightmare of a different extended ASCII for each little group of European countries or that Unicode copied that approach.

-- Patrick