diff --git a/Cargo.toml b/Cargo.toml
index 61addb8..06ca0cb 100644
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -5,9 +5,9 @@ license = "MIT OR Apache-2.0"
 edition = "2021"
 exclude = ["benchmarks", "doc"]
 repository  = "https://github.com/tower120/chute"
-description = "Lockfree mpmc/spmc multicast queue."
+description = "Lockfree mpmc/spmc broadcast queue."
 categories = ["concurrency", "data-structures"]
-keywords   = ["lockfree", "mpmc", "spmc", "multicast", "queue"]
+keywords   = ["lockfree", "mpmc", "spmc", "broadcast", "queue"]
 
 [dependencies]
 branch_hints = "0.4"
diff --git a/Readme.md b/Readme.md
index c9dc896..b9b7136 100644
--- a/Readme.md
+++ b/Readme.md
@@ -8,13 +8,13 @@
 
 ![Queue illustration](doc/img/mpmc_white.png)
 
-An mpmc[^mpmc]/spmc[^spmc] lock-free multicast[^broadcast] queue.
+An mpmc[^mpmc]/spmc[^spmc] lock-free broadcast[^broadcast] queue.
 
 [^mpmc]: Multi-producer multi-consumer.
 
 [^spmc]: Single-producer multi-consumer.
 
-[^broadcast]: Also known as a broadcast queue. Each consumer gets 
+[^broadcast]: Also known as a multicast queue. Each consumer gets 
 every message sent to queue, from the moment of subscription.
 
 * Lock-free consumers without overhead[^lockfree_overhead].
diff --git a/src/lib.rs b/src/lib.rs
index 88d9922..18fa76b 100644
--- a/src/lib.rs
+++ b/src/lib.rs
@@ -1,9 +1,10 @@
-//! Multi-producer multi-consumer lock-free multicast[^multicast] queue. 
+//! Multi-producer multi-consumer lock-free multicast queue[^multicast]. 
 //! 
-//! [^multicast]: Each consumer gets every message sent to queue, from the moment of subscription.
+//! [^multicast]: Or broadcast queue. Each consumer gets every message sent 
+//! to queue, from the moment of subscription.
 //!
-//! Memory-wise all consumers works with the same shared data blocks, so there 
-//! is no duplication.
+//! Memory-wise all consumers work with the same shared data blocks, so there 
+//! is no duplication. Queue is unbounded - it grows and shrinks dynamically as needed.
 //!
 //! Read performance is **stellar** - there is only one atomic read per block.
 //! Most of the time - this is just plain continuous data reading.