In the PDFs, paragraphs are accompanied by numbers, presumably so learners can quickly find the matching paragraph if they scroll from kanji to kana or romaji or English. If that is the case, then why don't they match sometimes? For example, from intermediate lesson 29:
This lesson doesn't even have the same number of paragraphs in each transcript. The kanji transcript has 25, the kana has 17, the romaji has 16 and the English has 10. This is the case in some other PDFs too.
How can anybody possibly find the corresponding paragraph using these numbers if they're all wrong?