Roll v1.0
diff --git a/README.md b/README.md
index 43dc2cf..d32cd52 100644
--- a/README.md
+++ b/README.md
@@ -1,11 +1,11 @@
# HyperLogLog for Erlang
-`hyper` is an implementation of the HyperLogLog algorithm with bias
-correction from HLL++ as the described in the excellent
-[paper by Google][]. Using HyperLogLog you can estimate the
-cardinality of large data sets or streams using very little
+This is an implementation of the HyperLogLog algorithm in Erlang. We
+have implemented the bias correction from HLL++ as the described in
+the excellent [paper by Google][]. Using HyperLogLog you can estimate
+the cardinality of large data sets or streams using very little
memory. The relative error is 1.04 * sqrt(2^P). When creating a new
-instance of `hyper`, you provide must provide P, allowing you to trade
+HyperLogLog filter, you provide must provide P, allowing you to trade
memory for accuracy. The union of two filters is lossless.
In practice this allows you to build efficient analytics systems. For
diff --git a/src/hyper.app.src b/src/hyper.app.src
index c8c15d9..339eb19 100644
--- a/src/hyper.app.src
+++ b/src/hyper.app.src
@@ -1,7 +1,7 @@
{application, hyper,
[
{description, ""},
- {vsn, "1"},
+ {vsn, "1.0"},
{registered, []},
{applications, [
kernel,
