Bug fix for duplicate words

This word list is still pretty bad ...

thesethreewords.py

@@ -11,9 +11,8 @@ import geohash
 def get_google_words():
     lines = open("words/google-ngram-list")
     words = []
-    for line in lines:
-        _, word = line.split()
-        words.append(word)
+    for word in lines:
+        words.append(word.strip())
 
     lines.close()
     random.seed(634634)

words/README.md

@@ -21,7 +21,7 @@ To get a list of the top 300 words:
 
 To create the wordlist used by `These3Words` run:
 
-   sort -n googlebooks-eng-all-1gram-20120701-*-filtered | python normalise-words.py | tail -n32768 > google-ngram-list
+   sort -n googlebooks-eng-all-1gram-20120701-*-filtered | python normalise-words.py | sort | uniq | tail -n32768 > google-ngram-list
 
 Check that your list is long enough by counting the lines
 in `google-ngram-list`, you need exactly 32768 words

words/normalise-words.py

@@ -109,4 +109,4 @@ for line in fileinput.input():
     if len(word) == 4:
         continue
     
-    print count, word.lower()
+    print word.lower()