How does DT tokenize text?

DEVONthink
1

Abbreviations that use forward slash and dot are appearing as separate words in the concordance.

  • “U.S.A.” → is broken down into 3 separate letters “U”, “S”, and “A”.
  • “b/n” (between) → is broken down into “b” & “n”.

is there any way to customize DT’s understanding of tokens?

2

No, indexing isn’t customizable.