[tescocles]

This wouldn't work in the same way in Japanese because characters can have different readings depending on the context. Could that be overcome with font ligatures? Still, even if we could overcome that massively complex task, some groups of characters (i.e. the aforementioned context) can still have multiple ways of reading them.

[dry_soup]

If you scroll down on the article in the OP, you'll see that Cantonese readings of Chinese characters are also context-specific, and they appear to have solved that problem with ligatures.

[soundnote]

The proper solution for Japanese is a time machine to the 1940s and instituting a proper script reform instead of the touyou kanji list and Shinjitai simplifications. :)