README.md in logstash_writer-0.0.1 vs README.md in logstash_writer-0.0.2

@@ -4,15 +4,15 @@
 
 # Installation
 
 It's a gem:
 
-    gem install gemplate
+    gem install logstash_writer
 
 There's also the wonders of [the Gemfile](http://bundler.io):
 
-    gem 'gemplate'
+    gem 'logstash_writer'
 
 If you're the sturdy type that likes to run from git:
 
     rake install
 
@@ -39,10 +39,14 @@
 as that is fixed, we are stuck with the `json_lines` approach.
 
 
 # Usage
 
+Start by including the necessary files:
+
+    require 'logstash_writer'
+
 An instance of `LogstashWriter` needs to be given the location of a server
 (or servers) to send the events to.  This can be any of:
 
     # An IPv4 address and port
     lw = LogstashWriter.new(server_name: "192.0.2.42:5151")
@@ -115,11 +119,11 @@
 * **`logstash_writer_queue_size`** -- the number of events currently in
   the backlog queue awaiting transmission.  In *theory*, this value should
   always be `received - (sent + dropped)`, but this gauge is maintained
   separately as a cross-check in case of bugs.
 
-* **`logstash_writer_last_sent_event_timestamp`** -- the UTC timestamp,
+* **`logstash_writer_last_sent_event_time_seconds`** -- the UTC timestamp,
   represented as the number of (fractional) seconds since the Unix epoch, at
   which the most recent event sent to a logstash server was originally
   submitted via `#send_event`.  This might require some unpacking.
 
   If everything is going along swimmingly, there's no queued events, and
@@ -127,10 +131,10 @@
   be whenever the last event was sent.  No big problem.  However, in the
   event of problems, this timestamp can tell you several things.
 
   Firstly, if there are queued events, you can tell how far behind in real
   time your logstash event history is, by calculating `NOW() -
-  logstash_writer_last_sent_event_timestamp`.  Thus, if you're not finding
+  logstash_writer_last_sent_event_time_seconds`.  Thus, if you're not finding
   events in your Kibana dashboard you were expecting to see, you can tell
   that there's a clog in the pipes by looking at this.
 
   Alternately, if the queue is empty, but this timestamp is perhaps older
   than you'd expect, then you know the problem is "upstream" of