From 8d6d633c66c8ed385f2263cdc5dcad14d1ef5c16 Mon Sep 17 00:00:00 2001
From: Josh Chen
Date: Thu, 4 Jun 2020 14:27:41 +0200
Subject: some documentation

---
 README.md |  3 ++-
 TODO.md   | 14 --------------
 2 files changed, 2 insertions(+), 15 deletions(-)
 delete mode 100644 TODO.md

diff --git a/README.md b/README.md
index 14dd5d9..53be95a 100644
--- a/README.md
+++ b/README.md
@@ -18,13 +18,14 @@ $ echo path/to/Isabelle/HoTT >> ~/.isabelle/Isabelle2020/ROOTS
 
 ### To-do list
 
-In no particular order. Some of the following might require changes to the Isabelle prover itself.
+In no particular order. Some of the following might(?) require changes to the Isabelle prover itself.
 
 - [ ] Implement dedicated tables for type information.
 - [ ] Tactic-based term elaboration has (at least) two problems:
         1. `assume(s)` clauses don't accept schematic vars, and
         2. it often results in overly-flexible subgoals that the typechecker doesn't solve.
       Will need an elaborator integrated into Isabelle's syntax checking.
+- [ ] Proper handling of universes.
 - [ ] Inductive type definitions.
 - [ ] Recursive function definitions.
 - [ ] Higher inductive type definitions.
diff --git a/TODO.md b/TODO.md
deleted file mode 100644
index 84f1160..0000000
--- a/TODO.md
+++ /dev/null
@@ -1,14 +0,0 @@
-# To-do list
-
-In no particular order. Some of the following might need changes to the
-Isabelle prover itself?
-
-[ ] Typing information is implicit in context facts, and the collection must be
-    searched every time we need a type for a term. For performance, should
-    probably implement dedicated tables.
-
-[ ] Tactic-based term elaboration has (at least) two problems:
-    1. `assume(s)` clauses don't accept schematic vars, and
-    2. it often results in overly-flexible subgoals that the typechecker
-    doesn't solve.
-    Will likely need an elaborator integrated into Isabelle's syntax checking.
-- 
cgit v1.2.3