diff --git a/src/SUMMARY.md b/src/SUMMARY.md
index 84bddcd..adba62b 100644
--- a/src/SUMMARY.md
+++ b/src/SUMMARY.md
@@ -15,5 +15,7 @@
   - [All Errors](./exercises/all_errors.md)
 - [Chapter 3](./chapter_3.md)
   - [Composition](./exercises/composition.md)
+  - [Covariance]()
+  - [Option, Vec and BoxFuture<'a>]()
 
 [Topics (Spoilers)](./topics.md)
diff --git a/src/chapter_1.md b/src/chapter_1.md
index b743fda..dbb4082 100644
--- a/src/chapter_1.md
+++ b/src/chapter_1.md
@@ -1 +1 @@
-# Chapter 1
+# Chapter 1: Strict Tests
diff --git a/src/chapter_3.md b/src/chapter_3.md
index 9b2e616..f2934c4 100644
--- a/src/chapter_3.md
+++ b/src/chapter_3.md
@@ -1 +1 @@
-# Chapter 3
\ No newline at end of file
+# Chapter 3: Overly Functional
diff --git a/src/exercises/composition.md b/src/exercises/composition.md
index d8cf910..af14bf1 100644
--- a/src/exercises/composition.md
+++ b/src/exercises/composition.md
@@ -1,6 +1,5 @@
 ```rust
 # use std::ops::ControlFlow;
-
 trait Monad<'a>: 'a {
     type W<A: 'a>: 'a;
 
@@ -86,7 +85,6 @@ where
 # U::iterate(ComposedIteration(i))
     }
 }
-
 # trait Local<'a>: Monad<'a> { fn wrap<A: 'a, M: Monad<'a>>(wa: Self::W<M::W<A>>) -> M::W<Self::W<A>>; }
 # struct CfInstance<E>(ControlFlow<(), E>);
 # use ControlFlow::{Continue, Break};
diff --git a/src/exercises/rcchars.md b/src/exercises/rcchars.md
index c51ee7a..19ee5de 100644
--- a/src/exercises/rcchars.md
+++ b/src/exercises/rcchars.md
@@ -45,7 +45,8 @@
 ## Solutions
 - [Implementation] used in rattlescript.
 - [Same solution] but with some extra comments.
-- [Another solution].
+- [Another solution]. I prefer this one.
+- There's an even simpler version that even an AI can generate (don't do that for `unsafe` code in production), but I dislike it for several reasons, even though it's mostly sound.
 
 [Implementation]: https://github.com/HavenSelph/rattlescript/blob/f8bafb8b063b9bf056efb1ea14188db0624d981c/src/interpreter/value.rs#L21-L50
 [Same solution]: https://gist.github.com/timotheyca/7e46c9734653b2fcbe826ea4d13b9aa0
diff --git a/src/topics.md b/src/topics.md
index 85d9296..e8f0cc3 100644
--- a/src/topics.md
+++ b/src/topics.md
@@ -2,11 +2,14 @@
 
 Rows are ordered based on estimated difficulty.
 
-| lifetimes\[-adjacent\] | `trait`s           | `struct`s    |
-|------------------------|--------------------|--------------|
-| [explicit lifetimes]   | [`Fn`?]            | [`Duration`] |
-| [extracting lifetimes] | [exclusive traits] | [`RcChars`]  |
-| [composition]          | [merging traits]   | [`AnyStr`]   |
+| lifetimes\[-adjacent\] | `trait`s           | `struct`s     |
+|------------------------|--------------------|---------------|
+| [explicit lifetimes]   | [`Fn`?]            | [`Duration`]  |
+| [extracting lifetimes] | [exclusive traits] | [All Errors?] |
+| [composition]          | [merging traits]   | [`RcChars`]   |
+|                        |                    | [`AnyStr`]    |
+
+p.s. most of the exercises are actually about `trait`s, this categorisation isn't strict.
 
 
 [explicit lifetimes]: ./exercises/refbind.md
@@ -18,3 +21,4 @@ Rows are ordered based on estimated difficulty.
 [`AnyStr`]: ./exercises/anystr.md
 [composition]: ./exercises/composition.md
 [`Duration`]: ./exercises/duration.md
+[All Errors?]: ./exercises/all_errors.md