I’m slightly dyslexic, so putting things in proper alphabetical order has always been an unduly arduous task for me. I’d never thought of it as something that had to be discovered though. Apparently so.
From Ann Blair’s Too Much to Know: Managing Scholarly Information before the Modern Age
Indexing notes raised the same kinds of questions as choosing headings, since each passage would be entered into the index under a keyword. In calling for an index to be drawn up for each notebook, Drexel recommended alphabetizing a passage under its principal thing (caput rei), for example, “the incredible growth of divine grace” under G for grace. Drexel explained that the index should be drawn up on sheets corresponding to each letter of the alphabet (with one sheet devoted to little-used letters like K, Q, X, and Z to save paper), with references to the notebooks that contained the excerpts (though Drexel does not specify the form of these references, possibly by page number). Since the sheets would receive new headings over time, the index or list of headings would be alphabetized only by first letter. To find a given heading would therefore require browsing the headings beginning with the same letter to find a particular one.
What a system. I wonder what Drexel would think of Google, or even the more primitive early search engines. Or hell, even the Find function in a text document.