diff --git a/notes.md b/notes.md
index 1b8943ee9cafc7b5a1e2f53265b90f99a92190f7..8a948798cd9d2c33d3b60794077ecfbea4da5fee 100644
--- a/notes.md
+++ b/notes.md
@@ -54,21 +54,33 @@ need followup.
   - uses it's self-rolled _"Delayed Open Source Attribution License"_ 
   - [license file on GH](https://github.com/ChildMindInstitute/mindlogger-applet-builder/blob/master/LICENSE.md)
 
-# Categorization question
+# Taxonomy
 
-Is there a distinction to be made between these cases?
+I (Seth) think there's a distinction to be made between these three cases:
 
-* licenses that are not open source licenses but that state (in the license
-  text somewhere) that they automatically convert / automatically permit use
-  and redistribution subject to a specified open source license after a period
-  of time
-* publicly announced practices of manually relicensing old codebase snapshots
-  on a particular schedule, which depend on a person at the company explicitly
-  making a delayed open source release
+* "automatic": licenses that are not open source licenses but that state
+  (in the license text somewhere) that they automatically convert /
+  automatically permit use and redistribution subject to a specified
+  open source license after a period of time
+
+* "manual": publicly announced practices of manually relicensing old codebase
+  snapshots on a particular schedule, which depend on a person at the company
+  explicitly making a delayed open source release
 
 (In some sense, this is a potential distinction between a "delayed open
 source licensing business practice" and a "delayed open source license".)
 
+* "post-hoc" / "unscheduled": proprietary software that eventually was
+  relicensed under an open source license, without a public statement of
+  intent to do so at the time of its original publication, or without a
+  published schedule for the conversion
+
+I would include the former two in the scope of the report but not the
+latter one (except to explain how it's different).  Some people have been
+suggesting some of these cases (which can be fairly famous, like Netscape
+Navigator!), but I think these should be thought of as more of a one-time
+"change" than a "delay".
+
 # Threads where we have posted
 
 Look in the follow-ups in these threads (and subthreads thereof) for