There’s no way that you need to know 9 out of 10 characters to get a “gist” of what the article is about – a gist by definition is a just the general idea, or the essence if a text.

While writing this post I was reading a lot of Taiwanese news articles. When reading news, especially as a foreign reader, you need to be able to allow yourself to skip over what you don’t know, or you’ll spend all day in the dictionary – and in these cases it’s not possible to say “choose a more suitable text”, since this is the news we’re talking about. With other texts, if you find yourself constantly frustrated, then finding more suitable material might be appropriate.

Even 80% (the most common ones) puts you only close to understanding what the context of the story is.

Only by losing a rare character here, another there (way over 90%) can you get into the gist of what a story is about. Of course, then you should look up those characters after the first skim-through of the article. Reading article in the ‘net with a mouse-over dictionary is handy.

If you pass more than every fifth character, might as well grab a more familiar text.

I do agree with you about that the capital sin of text books is to locate the pinyin so and make it so prominent that the eye of a student is drawn to them automatically.

Dropping pinyin completely after the introductory course is best; cover the pinyin or get books that don’t have it. For self-study, it’s better to look up characters separately and just remember the pronunciation by using them in sentences than to ever write down pinyin. I myself have difficulties remembering tones, and often write just the tone on top of a character. Bad habit, but not the same in my opinion. Any ideas on specifically learning the tones would be appreciated.

To alleviate the problem by introducing another script is just silly. Suppose you learn zhuyin first instead of, say, 100 common characters (and why would you ever), how is your situation different from pinyin? Every time you see a new character, boohoo, take refuge to the zhuyin; and you’ll equally end up ignoring the characters.

You mention rubytext (or furigana), and that’s probably the way pinyin should be written in intro level books as well; not separate from characters but jammed between them.

Chinese has a system of transliteration of foreign names (it’s sort of retarded, sure, but official), e.g. Puerto Rico -> 波多黎各. Additionally, most complex characters contain phonetic radicals. Why not learn them instead of some hypothetical and weird never-used-in-practice script?