diff --git a/osmarks.net_web_search_plan_(secret).myco b/osmarks.net_web_search_plan_(secret).myco
index d825df4..3eedca0 100644
--- a/osmarks.net_web_search_plan_(secret).myco
+++ b/osmarks.net_web_search_plan_(secret).myco
@@ -48,6 +48,7 @@ The job of a search engine is to retrieve useful information for users. This is
 * "Show me things related to this document" (including e.g. user's own draft code, papers, essays).
 * Refine query with extra detail.
 * There are several ways things can be "related" - semantically, stylistically, keywords, similar authorship (also location for images). Is there a clean way to cover all these or do we just need to have separate indices for each?
+* https://eugeneyan.com/writing/recsys-llm/
 
 = Cost