diff --git a/CHANGELOG.md b/CHANGELOG.md
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -1,3 +1,7 @@
+# v1.0.0.1
+
+Doc corrections.
+
 # v1.0.0.0
 
 ## Non-breaking
diff --git a/README.md b/README.md
--- a/README.md
+++ b/README.md
@@ -16,7 +16,7 @@
 > for why this project exists and what tradeoffs that implies.
 
 A pure-Haskell [`pqi`](https://github.com/nikita-volkov/pqi) adapter
-that speaks the PostgreSQL frontend/backend wire protocol directly — no
+that speaks the PostgreSQL frontend/backend wire protocol directly - no
 dependency on the C `libpq` library.
 
 `pqi-native` reimplements the wire protocol handled by the PostgreSQL C
@@ -25,8 +25,8 @@
 
 ## Fidelity goal
 
-The goal is **byte-identical output to `libpq`** for every protocol-derived
-value — error message strings, notice text, result status, field metadata,
+The goal is **identical output to `libpq`** for every protocol-derived
+value - error message strings, notice text, result status, field metadata,
 cell data, and all structured error fields. Fidelity is continuously enforced
 by [`pqi-conformance`](https://github.com/nikita-volkov/pqi-conformance), which
 runs every operation on both this adapter and a direct
@@ -42,7 +42,7 @@
 
 Because it implements the same `pqi` interface as
 [`pqi-ffi`](https://github.com/nikita-volkov/pqi-ffi), switching between the
-two is a one-line change — pass a different `Adapter` value, nothing else in
+two is a one-line change - pass a different `Adapter` value, nothing else in
 your code moves. That makes `pqi-native` low-risk to evaluate now and easy to
 fall back from: adopt it where you want to shed the `libpq` dependency, and
 drop back to `pqi-ffi` at any time without touching the rest of your
diff --git a/pqi-native.cabal b/pqi-native.cabal
--- a/pqi-native.cabal
+++ b/pqi-native.cabal
@@ -1,6 +1,6 @@
 cabal-version: 3.0
 name: pqi-native
-version: 1.0.0.0
+version: 1.0.0.1
 category: Database, PostgreSQL
 synopsis: Native (pure-Haskell) adapter for pqi
 description:
@@ -15,7 +15,7 @@
   differential conformance suite, but not yet proven in production. It
   implements the same @pqi@ interface as
   [@pqi-ffi@](https://hackage.haskell.org/package/pqi-ffi), the stable
-  C-backed adapter, so the two are fully interchangeable — swapping
+  C-backed adapter, so the two are fully interchangeable - swapping
   between them is a one-line @Adapter@ change.
 
 homepage: https://github.com/nikita-volkov/pqi-native
diff --git a/src/library/Pqi/Native/LargeObject.hs b/src/library/Pqi/Native/LargeObject.hs
--- a/src/library/Pqi/Native/LargeObject.hs
+++ b/src/library/Pqi/Native/LargeObject.hs
@@ -1,5 +1,5 @@
 -- | The large-object interface, implemented over the server's @lo_*@ SQL
--- functions (rather than libpq's fast-path protocol) — identical results, far
+-- functions (rather than libpq's fast-path protocol) - identical results, far
 -- less machinery. As with libpq, the open\/read\/write\/close operations must be
 -- run inside a transaction managed by the caller.
 module Pqi.Native.LargeObject
diff --git a/src/library/Pqi/Native/Types.hs b/src/library/Pqi/Native/Types.hs
--- a/src/library/Pqi/Native/Types.hs
+++ b/src/library/Pqi/Native/Types.hs
@@ -52,7 +52,7 @@
 -- Carries references to connection state so the cancel implementation can
 -- decide whether a network round-trip is necessary:
 --
--- * @asyncPendingRef@: False when nothing is in flight at all — skip the
+-- * @asyncPendingRef@: False when nothing is in flight at all - skip the
 --   round-trip entirely.
 -- * @pipelineStatusRef@ + @pendingCommandsRef@: in pipeline mode,
 --   @asyncPending@ stays True even after all @CommandComplete@ messages