diff --git a/lib/twitter_ebooks/model.rb b/lib/twitter_ebooks/model.rb
index b3bbb13..8849118 100644
--- a/lib/twitter_ebooks/model.rb
+++ b/lib/twitter_ebooks/model.rb
@@ -80,7 +80,13 @@ module Ebooks
     # @param token [String]
     # @return [Integer]
     def tikify(token)
-      @tikis[token] or (@tokens << token and @tikis[token] = @tokens.length-1)
+      if @tikis.has_key?(token) then
+        return @tikis[token]
+      else
+        (@tokens.length+1)%1000 == 0 and puts "#{@tokens.length+1} tokens"
+        @tokens << token 
+        return @tikis[token] = @tokens.length-1
+      end
     end
 
     # Convert a body of text into arrays of tikis
@@ -143,8 +149,8 @@ module Ebooks
         end
       end
 
-      text = statements.join("\n")
-      mention_text = mentions.join("\n")
+      text = statements.join("\n").encode('UTF-8', :invalid => :replace)
+      mention_text = mentions.join("\n").encode('UTF-8', :invalid => :replace)
 
       lines = nil; statements = nil; mentions = nil # Allow garbage collection
 
@@ -155,6 +161,7 @@ module Ebooks
 
       log "Ranking keywords"
       @keywords = NLP.keywords(text).top(200).map(&:to_s)
+      log "Top keywords: #{@keywords[0]} #{@keywords[1]} #{@keywords[2]}"
 
       self
     end
@@ -218,6 +225,7 @@ module Ebooks
       tweet = ""
 
       while (tikis = generator.generate(3, :bigrams)) do
+        log "Attempting to produce tweet try #{retries+1}/#{retry_limit}"
         next if tikis.length <= 3 && !responding
         break if valid_tweet?(tikis, limit)
 
@@ -226,6 +234,7 @@ module Ebooks
       end
 
       if verbatim?(tikis) && tikis.length > 3 # We made a verbatim tweet by accident
+        log "Attempting to produce unigram tweet try #{retries+1}/#{retry_limit}"
         while (tikis = generator.generate(3, :unigrams)) do
           break if valid_tweet?(tikis, limit) && !verbatim?(tikis)
 
diff --git a/lib/twitter_ebooks/nlp.rb b/lib/twitter_ebooks/nlp.rb
index 541720b..b65e162 100644
--- a/lib/twitter_ebooks/nlp.rb
+++ b/lib/twitter_ebooks/nlp.rb
@@ -1,6 +1,7 @@
 # encoding: utf-8
 require 'fast-stemmer'
 require 'highscore'
+require 'htmlentities'
 
 module Ebooks
   module NLP
@@ -42,7 +43,6 @@ module Ebooks
     # Lazily load HTML entity decoder
     # @return [HTMLEntities]
     def self.htmlentities
-      require 'htmlentities'
       @htmlentities ||= HTMLEntities.new
     end
 
diff --git a/lib/twitter_ebooks/suffix.rb b/lib/twitter_ebooks/suffix.rb
index ff57f97..aa2b791 100644
--- a/lib/twitter_ebooks/suffix.rb
+++ b/lib/twitter_ebooks/suffix.rb
@@ -19,6 +19,9 @@ module Ebooks
       @bigrams = {}
 
       @sentences.each_with_index do |tikis, i|
+        if (i % 10000 == 0) then
+          log ("Building: sentence #{i} of #{sentences.length}")
+        end
         last_tiki = INTERIM
         tikis.each_with_index do |tiki, j|
           @unigrams[last_tiki] ||= []