diff --git a/BenchmarkCorpus/planning_document.txt b/BenchmarkCorpus/planning_document.txt
deleted file mode 100644
index 25f30d48599487e391718931253b60cfd871be9e..0000000000000000000000000000000000000000
--- a/BenchmarkCorpus/planning_document.txt
+++ /dev/null
@@ -1,35 +0,0 @@
-Word Complexity Project:
-
-General hypothesis: The language that scientists and many science educators use online is more complex than language used by non-scientists and science deniers. 
-
-Problem: This leads to the most readable and findable information being potentially less accurate (especially regarding controversial issues), while the most accurate information is likely more difficult to find in searches and will have less impact.
-
-1. Text complexity vs. site ranking within and between searches 
-Are simpler texts ranking higher in Google?
-How do scientific texts fare within this ranking?
-a. For various scientific searches vs. various non-scientific searches
-i. Sci searches may be: Genetics, evolution, cancer, vaccine, GMO, climate change, photosynthesis
-ii. Non-sci searches may be: Soccer, culture, reality television, ???
-b. Also perhaps targeted comparisons of ideal educational websites vs average?
-
-2. Text complexity vs. text sentiment
-Are more neutral/factual websites more complex?
-a. Rank pro, anti, and neutral websites for text complexity
-i. Vaccines
-ii. GMOs
-iii. Climate change
-
-3. Case studies: Complexity of texts using scientific vs. non-scientific terms 
-Are scientists using overly complex (but more precise) language online?
-a. GMO vs. transgenics
-b. Global warming vs. climate change vs. anthropogenic climate change
-c. (though non-scientific, perhaps) Intelligent design vs. evolution
-d. Also perhaps targeted comparisons of scientist-led blogs vs. public-led blogs covering specific scientific subjects? *can’t be batch processed
-
-Additional questions:
--In Russell’s general search graphs, two clusters of websites seemed to fall out in the graphs. How do we figure out what is causing this?
-
-Issues to consider:
--Are the first few, super successful sites outliers? Should we run these with and without the first page of results to see the differences?
-
--If AAB sources come up in any of these, should they be automatically excluded?
diff --git a/CodeComplexity/results.txt b/CodeComplexity/results.txt
deleted file mode 100644
index 7493e88f825884d27c35c4c8368aea41e6786331..0000000000000000000000000000000000000000
--- a/CodeComplexity/results.txt
+++ /dev/null
@@ -1,44 +0,0 @@
-/opt/conda/lib/python3.6/site-packages/nltk/twitter/__init__.py:20: UserWarning: The twython library has not been installed. Some functionality from the twitter package will not be available.
-  warnings.warn("The twython library has not been installed. "
-('\n'
- '    1 - 5 A (low risk - simple block)\n'
- '    6 - 10 B (low risk - well structured and stable block)\n'
- '    11 - 20 C (moderate risk - slightly complex block)\n'
- '    21 - 30 D (more than moderate risk - more complex block)\n'
- '    31 - 40 E (high risk - complex block, alarming)\n'
- '    41+ F (very high risk - error-prone, unstable block)\n'
- '    ')
-'cognitive complexity of function <function text_proc at 0x7f26d29a8400> is: 9'
-'Good work keep writing modular, readable, and simple code.'
-('\n'
- '    1 - 5 A (low risk - simple block)\n'
- '    6 - 10 B (low risk - well structured and stable block)\n'
- '    11 - 20 C (moderate risk - slightly complex block)\n'
- '    21 - 30 D (more than moderate risk - more complex block)\n'
- '    31 - 40 E (high risk - complex block, alarming)\n'
- '    41+ F (very high risk - error-prone, unstable block)\n'
- '    ')
-('cognitive complexity of function <function scrapelandtext at 0x7f26d29a8378> '
- 'is: 5')
-'Good work keep writing modular, readable, and simple code.'
-('\n'
- '    1 - 5 A (low risk - simple block)\n'
- '    6 - 10 B (low risk - well structured and stable block)\n'
- '    11 - 20 C (moderate risk - slightly complex block)\n'
- '    21 - 30 D (more than moderate risk - more complex block)\n'
- '    31 - 40 E (high risk - complex block, alarming)\n'
- '    41+ F (very high risk - error-prone, unstable block)\n'
- '    ')
-'cognitive complexity of function <function slat_ at 0x7f26d29a82f0> is: 7'
-'Good work keep writing modular, readable, and simple code.'
-('\n'
- '    1 - 5 A (low risk - simple block)\n'
- '    6 - 10 B (low risk - well structured and stable block)\n'
- '    11 - 20 C (moderate risk - slightly complex block)\n'
- '    21 - 30 D (more than moderate risk - more complex block)\n'
- '    31 - 40 E (high risk - complex block, alarming)\n'
- '    41+ F (very high risk - error-prone, unstable block)\n'
- '    ')
-'cognitive complexity of function <function text_proc at 0x7f26d29a8400> is: 9'
-'Good work keep writing modular, readable, and simple code.'
-jovyan@1c8cb2f0367f:~/CodeComplexity$
diff --git a/CodeComplexity/test_complexity.py b/CodeComplexity/test_complexity.py
deleted file mode 100644
index f43ddb71a8ac43d18b710780c3a0babc2471f3cb..0000000000000000000000000000000000000000
--- a/CodeComplexity/test_complexity.py
+++ /dev/null
@@ -1,51 +0,0 @@
-
-import inspect
-import types
-import pandas as pd
-import inspect, radon, pprint
-from radon.complexity import cc_rank, cc_visit
-    
-def ccomplexity_rater(other_function):
-    '''
-    This function calculates the radian cyclomatic complexity of other functions.
-    Radian complexity is used as a proxy for cognitive complexity, ie how hard is a code block to understand.
-    Inputs: Other Python functions.
-    Outputs: A positive integer value that is located in the interval 1-41. The scalar is used in conjunction
-    with a printed legend.
-
-    The program first uses introspection to convert other_function to a string representation of the
-    source code that the function was originally expressed in.
-    Subsequently another module radon that calculates cognitive complexity is called.
-    Dependencies: If the radon module is not installed consider executing ```pip install radon```
-    From:   http://radon.readthedocs.io/en/latest/api.html
-    https://www.guru99.com/cyclomatic-complexity.html
-
-    '''
-    f_source_code = "".join(inspect.getsourcelines(other_function)[0])
-    results = radon.complexity.cc_visit(f_source_code)
-    ranking = radon.complexity.sorted_results(results)
-    pp = pprint.PrettyPrinter(indent=4)
-    ranking_guide = '''
-    1 - 5 A (low risk - simple block)
-    ...
-    41+ F (very high risk - error-prone, unstable block)
-    '''
-    pp.pprint(ranking_guide)
-    actual_value = ranking[0][-1]
-    pp.pprint('cognitive complexity of function {0} is: {1}'.format(other_function,actual_value))
-    #df = pd.DataFrame(['cognitive complexity of function: '+str(other_function),actual_value])
-
-    if actual_value > 10:
-        pp.pprint('Consider rewriting your code it might be hard for you and others to understand, and therefore maintain')
-    return actual_value
-
-def is_function(object):
-    return isinstance(object, types.FunctionType) 
-
-def rank_all_sub_module_functions(provided_module):
-    sc_objects = [v for k,v in inspect.getmembers(provided_module) ]
-    ranks = []
-    for sc in sc_objects:
-        if is_function(sc):
-            ranks.append(ccomplexity_rater(sc))
-    return ranks
diff --git a/Examples/author_vs_distr.png b/Examples/author_vs_distr.png
new file mode 100644
index 0000000000000000000000000000000000000000..30df604ecf2b02d9fd62df83cdf3012dcf542cbc
Binary files /dev/null and b/Examples/author_vs_distr.png differ
diff --git a/Examples/figure_joss.png b/Examples/figure_joss.png
new file mode 100644
index 0000000000000000000000000000000000000000..3201328ea53a18e0b45ca77d3ed3239f3a952c5e
Binary files /dev/null and b/Examples/figure_joss.png differ
diff --git a/Examples/for_joss.ipynb b/Examples/for_joss.ipynb
index 816dd301634cb1f4defea48b6d224eea9f183867..411cd4e0f5b3efcd6b4331ae82bfa7398630445f 100644
--- a/Examples/for_joss.ipynb
+++ b/Examples/for_joss.ipynb
@@ -164,7 +164,7 @@
     {
      "data": {
       "text/plain": [
-       "[6.484375, 8.484375, 14.484375, 16.484375, 8.484375]"
+       "[6.484375, 8.484375, 14.484375, 16.484375, 8.484375, 16.484375]"
       ]
      },
      "execution_count": 6,
@@ -173,7 +173,7 @@
     },
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]
@@ -185,7 +185,7 @@
     },
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 800x640 with 1 Axes>"
       ]
@@ -298,7 +298,7 @@
     {
      "data": {
       "text/plain": [
-       "[0.21542776998597474, 0.017952314165497897, 0.008976157082748949]"
+       "[0.21449518223711772, 0.01787459851975981, 0.008937299259879905]"
       ]
      },
      "execution_count": 9,
@@ -320,9 +320,9 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "5\n",
+      "6\n",
       "64\n",
-      "5\n"
+      "6\n"
      ]
     }
    ],
@@ -342,7 +342,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "5\n"
+      "6\n"
      ]
     }
    ],
@@ -357,7 +357,7 @@
    "outputs": [
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 900x900 with 2 Axes>"
       ]
@@ -410,13 +410,9 @@
     "\n",
     "legendMain=ax.legend(labels=[str(\"Google scholar author relative to ART Corpus distribution. Total docs: \")+str(len(trainingDats))], prop=legend_properties,loc='upper left')\n",
     "\n",
-    "\n",
-    "#for i,j,k in zip(author_stats_grid,heights,[str(NAME)+' mean',str(NAME)+' min',str(NAME)+' max']):\n",
-    "\n",
     "xinterval = author_stats_grid\n",
     "xinterval.extend(bmark_stats_items_grid)\n",
-    "#xinterval.append(other)\n",
-    "#coords = copy.copy([item for item in ax.get_xticklabels()])\n",
+    "\n",
     "x1,y1,z1 = (mwp_distance[0],mwp_height[0],str('mean wikipedia'))\n",
     "xinterval.insert(4,x1)\n",
     "ax.set_xticks(xinterval)\n",
@@ -428,13 +424,11 @@
     "x,y,z = (worst_distance[0],worst_height,other_name)\n",
     "\n",
     "data3 = pd.DataFrame({\n",
-    "'Standard Reading Level': [x,x1],\n",
-    "    'CDF': [y,y1]\n",
+    "'Standard Reading Level': [x1],\n",
+    "    'CDF': [y1]\n",
     "    })\n",
     "ax = sns.regplot(data=data3, x='Standard Reading Level', y=\"CDF\", fit_reg=False, marker=\"o\", color=\"green\")\n",
     "\n",
-    "#new_categories.append(other_name)\n",
-    "\n",
     "ax.set_xticklabels(new_categories, minor=False, rotation=90)\n",
     "ax.set_xticklabels(new_categories, minor=True, rotation=0)\n",
     "\n",
@@ -444,16 +438,9 @@
     "xticks = list(range(0,45,5))\n",
     "\n",
     "plt.xticks(xticks)\n",
-    "#ax2.xaxis.set_minor_locator(AutoMinorLocator(4))\n",
     "plt.tick_params(axis=\"y\", labelcolor=\"r\", pad=8)\n",
     "\n",
-    "\n",
-    "\n",
-    "\n",
-    "ax.text(x-0.25,y+0.005,z, rotation=90)  \n",
-    "\n",
-    "#ax.text(x1,y1,z1, rotation=90)  \n",
-    "\n",
+    "ax.set(xlabel='Reading Grade', ylabel='Normalized Number of Texts at Reading Grade', title='some title')\n",
     "    \n",
     "plt.savefig(str(NAME)+'_author_readability.png')\n",
     "plt.show()\n",
@@ -520,7 +507,8 @@
        " 8.421875,\n",
        " 14.234375,\n",
        " 16.171875,\n",
-       " 8.421875]"
+       " 8.421875,\n",
+       " 16.171875]"
       ]
      },
      "execution_count": 15,
@@ -567,7 +555,7 @@
        "    <tr>\n",
        "      <td>0</td>\n",
        "      <td>18.109375</td>\n",
-       "      <td>0.215428</td>\n",
+       "      <td>0.214495</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
@@ -575,7 +563,7 @@
       ],
       "text/plain": [
        "   Standard Reading Level       CDF\n",
-       "0               18.109375  0.215428"
+       "0               18.109375  0.214495"
       ]
      },
      "execution_count": 16,
@@ -621,17 +609,17 @@
        "    <tr>\n",
        "      <td>0</td>\n",
        "      <td>18.109375</td>\n",
-       "      <td>0.215428</td>\n",
+       "      <td>0.214495</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <td>1</td>\n",
        "      <td>13.265625</td>\n",
-       "      <td>0.017952</td>\n",
+       "      <td>0.017875</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <td>2</td>\n",
        "      <td>25.859375</td>\n",
-       "      <td>0.008976</td>\n",
+       "      <td>0.008937</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
@@ -639,9 +627,9 @@
       ],
       "text/plain": [
        "   mean, min, maximum       CDF\n",
-       "0           18.109375  0.215428\n",
-       "1           13.265625  0.017952\n",
-       "2           25.859375  0.008976"
+       "0           18.109375  0.214495\n",
+       "1           13.265625  0.017875\n",
+       "2           25.859375  0.008937"
       ]
      },
      "execution_count": 17,
@@ -661,7 +649,7 @@
     {
      "data": {
       "text/plain": [
-       "[6.484375, 8.484375, 14.484375, 16.484375, 8.484375]"
+       "[6.484375, 8.484375, 14.484375, 16.484375, 8.484375, 16.484375]"
       ]
      },
      "execution_count": 18,
@@ -690,18 +678,1027 @@
     {
      "data": {
       "text/plain": [
-       "{'link': 'local_resource',\n",
-       " 'wcount': 299,\n",
-       " 'english': True,\n",
+       "{'link': 'nicholas',\n",
+       " 'page_rank': 'nicholas',\n",
+       " 'wcount': 3593,\n",
+       " 'tokens': ['understanding',\n",
+       "  'climate',\n",
+       "  'change',\n",
+       "  'gregory',\n",
+       "  'nicholas',\n",
+       "  'v',\n",
+       "  '1',\n",
+       "  '30',\n",
+       "  'aug',\n",
+       "  '2019',\n",
+       "  '1',\n",
+       "  '.',\n",
+       "  'introduction',\n",
+       "  'in',\n",
+       "  '2019',\n",
+       "  ',',\n",
+       "  'newspapers',\n",
+       "  'magazines',\n",
+       "  'publishing',\n",
+       "  'reports',\n",
+       "  'subject',\n",
+       "  'climate',\n",
+       "  'change',\n",
+       "  '.',\n",
+       "  'it',\n",
+       "  'appears',\n",
+       "  'scientific',\n",
+       "  'meteorological',\n",
+       "  'bodies',\n",
+       "  'see',\n",
+       "  'climate',\n",
+       "  'change',\n",
+       "  'major',\n",
+       "  'problem',\n",
+       "  'humankind',\n",
+       "  '(',\n",
+       "  'species',\n",
+       "  ')',\n",
+       "  ',',\n",
+       "  'concerted',\n",
+       "  'action',\n",
+       "  'needed',\n",
+       "  'immediately',\n",
+       "  'prevent',\n",
+       "  'disaster',\n",
+       "  ',',\n",
+       "  'principally',\n",
+       "  'reducing',\n",
+       "  'emissions',\n",
+       "  'greenhouse',\n",
+       "  'gases',\n",
+       "  '.',\n",
+       "  'however',\n",
+       "  'politicians',\n",
+       "  'administrative',\n",
+       "  'structures',\n",
+       "  'difficulty',\n",
+       "  'understanding',\n",
+       "  'nature',\n",
+       "  'problem',\n",
+       "  '(',\n",
+       "  'except',\n",
+       "  'perhaps',\n",
+       "  'europeans',\n",
+       "  ')',\n",
+       "  'seem',\n",
+       "  'hurry',\n",
+       "  'follow',\n",
+       "  'recommendations',\n",
+       "  'scientists',\n",
+       "  '.',\n",
+       "  'i',\n",
+       "  'believe',\n",
+       "  'people',\n",
+       "  'clear',\n",
+       "  'understanding',\n",
+       "  'problem',\n",
+       "  'scientific',\n",
+       "  'community',\n",
+       "  'somehow',\n",
+       "  'unable',\n",
+       "  '?',\n",
+       "  'rephrase',\n",
+       "  'explain',\n",
+       "  'reasoning',\n",
+       "  'form',\n",
+       "  'understood',\n",
+       "  'man',\n",
+       "  'eg',\n",
+       "  '‘',\n",
+       "  'lay',\n",
+       "  'street',\n",
+       "  '.',\n",
+       "  'they',\n",
+       "  'expect',\n",
+       "  'us',\n",
+       "  'take',\n",
+       "  'word',\n",
+       "  'without',\n",
+       "  'understanding',\n",
+       "  'audience',\n",
+       "  '’',\n",
+       "  'thisextremelyseriousmatterourselves',\n",
+       "  '.',\n",
+       "  'when',\n",
+       "  'one',\n",
+       "  'reads',\n",
+       "  'technical',\n",
+       "  'literature',\n",
+       "  'subject',\n",
+       "  ',',\n",
+       "  'much',\n",
+       "  'seems',\n",
+       "  'transparent',\n",
+       "  'scientific',\n",
+       "  'background',\n",
+       "  '.',\n",
+       "  'i',\n",
+       "  'therefore',\n",
+       "  'attempting',\n",
+       "  'essay',\n",
+       "  'go',\n",
+       "  'back',\n",
+       "  'scratch',\n",
+       "  'build',\n",
+       "  'component',\n",
+       "  'knowledge',\n",
+       "  'arguments',\n",
+       "  'form',\n",
+       "  'easily',\n",
+       "  'understood',\n",
+       "  '.',\n",
+       "  'be',\n",
+       "  'warned',\n",
+       "  ',',\n",
+       "  'however',\n",
+       "  '!',\n",
+       "  'it',\n",
+       "  'simple',\n",
+       "  'task',\n",
+       "  ',',\n",
+       "  'many',\n",
+       "  'scientific',\n",
+       "  'concepts',\n",
+       "  'involved',\n",
+       "  '.',\n",
+       "  'the',\n",
+       "  'literature',\n",
+       "  'subject',\n",
+       "  ',',\n",
+       "  'much',\n",
+       "  'available',\n",
+       "  'reputable',\n",
+       "  'internet',\n",
+       "  'sites',\n",
+       "  ',',\n",
+       "  'falls',\n",
+       "  'following',\n",
+       "  'categories',\n",
+       "  ':',\n",
+       "  'international',\n",
+       "  'bodies',\n",
+       "  '(',\n",
+       "  'international',\n",
+       "  'panel',\n",
+       "  'climate',\n",
+       "  'change',\n",
+       "  ')',\n",
+       "  ';',\n",
+       "  'national',\n",
+       "  'bodies',\n",
+       "  '(',\n",
+       "  'australian',\n",
+       "  'meteorological',\n",
+       "  'bureau',\n",
+       "  ',',\n",
+       "  'csiro',\n",
+       "  ',',\n",
+       "  'climate',\n",
+       "  'change',\n",
+       "  'authority',\n",
+       "  ')',\n",
+       "  ';',\n",
+       "  'state',\n",
+       "  'government',\n",
+       "  'bodies',\n",
+       "  'universities',\n",
+       "  ';',\n",
+       "  'miscellaneous',\n",
+       "  'websites',\n",
+       "  ',',\n",
+       "  'dubious',\n",
+       "  '.',\n",
+       "  'in',\n",
+       "  'study',\n",
+       "  'i',\n",
+       "  'drawn',\n",
+       "  'selection',\n",
+       "  'reputable',\n",
+       "  'web',\n",
+       "  'based',\n",
+       "  'material',\n",
+       "  'also',\n",
+       "  'readily',\n",
+       "  'available',\n",
+       "  'textbooks',\n",
+       "  '.',\n",
+       "  '2',\n",
+       "  '.',\n",
+       "  'setting',\n",
+       "  'scene',\n",
+       "  '–',\n",
+       "  'what',\n",
+       "  'makes',\n",
+       "  'planet',\n",
+       "  'special',\n",
+       "  '?',\n",
+       "  'let',\n",
+       "  'us',\n",
+       "  'start',\n",
+       "  'basics',\n",
+       "  'earth',\n",
+       "  '’',\n",
+       "  'climate',\n",
+       "  ',',\n",
+       "  'looking',\n",
+       "  'firstly',\n",
+       "  'planet',\n",
+       "  '’',\n",
+       "  'rotation',\n",
+       "  'orbit',\n",
+       "  'around',\n",
+       "  'sun',\n",
+       "  '(',\n",
+       "  'see',\n",
+       "  'diagram',\n",
+       "  ')',\n",
+       "  '.',\n",
+       "  'orbit',\n",
+       "  'orientation',\n",
+       "  'earth',\n",
+       "  'our',\n",
+       "  'earth',\n",
+       "  'rotates',\n",
+       "  'axis',\n",
+       "  ',',\n",
+       "  'giving',\n",
+       "  'us',\n",
+       "  'cycle',\n",
+       "  'day',\n",
+       "  'night',\n",
+       "  '.',\n",
+       "  'it',\n",
+       "  'also',\n",
+       "  'moves',\n",
+       "  'space',\n",
+       "  'around',\n",
+       "  'sun',\n",
+       "  'approximately',\n",
+       "  'circular',\n",
+       "  'orbit',\n",
+       "  '.',\n",
+       "  'because',\n",
+       "  'its',\n",
+       "  'axis',\n",
+       "  'tilted',\n",
+       "  ',',\n",
+       "  'illuminated',\n",
+       "  'area',\n",
+       "  'shifts',\n",
+       "  'north',\n",
+       "  'south',\n",
+       "  'back',\n",
+       "  'earth',\n",
+       "  '’',\n",
+       "  'passage',\n",
+       "  'around',\n",
+       "  'sun',\n",
+       "  ',',\n",
+       "  'experience',\n",
+       "  'seasons',\n",
+       "  '.',\n",
+       "  'the',\n",
+       "  'intensity',\n",
+       "  'sun',\n",
+       "  '’',\n",
+       "  'rays',\n",
+       "  'strongest',\n",
+       "  'tropics',\n",
+       "  'directly',\n",
+       "  'face',\n",
+       "  'sun',\n",
+       "  ',',\n",
+       "  'weakest',\n",
+       "  'polar',\n",
+       "  'areas',\n",
+       "  '.',\n",
+       "  'in',\n",
+       "  'solar',\n",
+       "  'system',\n",
+       "  ',',\n",
+       "  'planet',\n",
+       "  'earth',\n",
+       "  'placed',\n",
+       "  'fortunate',\n",
+       "  'location',\n",
+       "  '.',\n",
+       "  'the',\n",
+       "  'next',\n",
+       "  'outermost',\n",
+       "  'planet',\n",
+       "  'mars',\n",
+       "  'barren',\n",
+       "  ',',\n",
+       "  'thin',\n",
+       "  'atmosphere',\n",
+       "  ',',\n",
+       "  'without',\n",
+       "  'water',\n",
+       "  'cold',\n",
+       "  'surface',\n",
+       "  '.',\n",
+       "  'the',\n",
+       "  'next',\n",
+       "  'innermost',\n",
+       "  'planet',\n",
+       "  'venus',\n",
+       "  'dense',\n",
+       "  'poisonous',\n",
+       "  'atmosphere',\n",
+       "  ',',\n",
+       "  'surface',\n",
+       "  'extremely',\n",
+       "  'hot',\n",
+       "  '.',\n",
+       "  'but',\n",
+       "  'earth',\n",
+       "  'blessed',\n",
+       "  'atmosphere',\n",
+       "  'able',\n",
+       "  'support',\n",
+       "  'life',\n",
+       "  ',',\n",
+       "  'existence',\n",
+       "  'large',\n",
+       "  'quantities',\n",
+       "  'water',\n",
+       "  'form',\n",
+       "  'oceans',\n",
+       "  '.',\n",
+       "  'its',\n",
+       "  'surface',\n",
+       "  'temperature',\n",
+       "  'determined',\n",
+       "  'incoming',\n",
+       "  'radiation',\n",
+       "  'sun',\n",
+       "  ',',\n",
+       "  'surface',\n",
+       "  'characteristics',\n",
+       "  'land',\n",
+       "  'masses',\n",
+       "  'seas',\n",
+       "  ',',\n",
+       "  'nature',\n",
+       "  'gases',\n",
+       "  'make',\n",
+       "  'atmosphere',\n",
+       "  '(',\n",
+       "  'nitrogen',\n",
+       "  ',',\n",
+       "  'oxygen',\n",
+       "  'others',\n",
+       "  ')',\n",
+       "  '.',\n",
+       "  'conditions',\n",
+       "  'earth',\n",
+       "  '’',\n",
+       "  'surface',\n",
+       "  'largely',\n",
+       "  'determined',\n",
+       "  'atmosphere',\n",
+       "  '.',\n",
+       "  'the',\n",
+       "  'atmosphere',\n",
+       "  'made',\n",
+       "  'layers',\n",
+       "  ',',\n",
+       "  'shown',\n",
+       "  'diagram',\n",
+       "  '.',\n",
+       "  'the',\n",
+       "  'troposphere',\n",
+       "  ',',\n",
+       "  'average',\n",
+       "  '11',\n",
+       "  'km',\n",
+       "  'thick',\n",
+       "  '(',\n",
+       "  '16',\n",
+       "  'km',\n",
+       "  'thick',\n",
+       "  'equator',\n",
+       "  '8',\n",
+       "  'km',\n",
+       "  'poles',\n",
+       "  ')',\n",
+       "  '.this',\n",
+       "  'region',\n",
+       "  'contains',\n",
+       "  '90',\n",
+       "  '%',\n",
+       "  'atmosphere',\n",
+       "  '’',\n",
+       "  'mass',\n",
+       "  '.',\n",
+       "  'it',\n",
+       "  'region',\n",
+       "  'pressure',\n",
+       "  'differences',\n",
+       "  ',',\n",
+       "  'winds',\n",
+       "  ',',\n",
+       "  'vertical',\n",
+       "  'convection',\n",
+       "  'currents',\n",
+       "  ',',\n",
+       "  'water',\n",
+       "  'evaporation',\n",
+       "  'condensation',\n",
+       "  'take',\n",
+       "  'place',\n",
+       "  '.',\n",
+       "  'the',\n",
+       "  'stratosphere',\n",
+       "  ',',\n",
+       "  'extends',\n",
+       "  '50',\n",
+       "  'km',\n",
+       "  '.',\n",
+       "  'a',\n",
+       "  'brief',\n",
+       "  'sentence',\n",
+       "  'mesosphere',\n",
+       "  'thermosphere',\n",
+       "  'would',\n",
+       "  'helpful',\n",
+       "  'mentioned',\n",
+       "  'diagram',\n",
+       "  '.',\n",
+       "  'beyond',\n",
+       "  'stratosphere',\n",
+       "  'atmosphere',\n",
+       "  'becomes',\n",
+       "  'thinner',\n",
+       "  'thinner',\n",
+       "  'increasing',\n",
+       "  'altitude',\n",
+       "  ',',\n",
+       "  'definite',\n",
+       "  'boundary',\n",
+       "  'atmosphere',\n",
+       "  'outer',\n",
+       "  'space',\n",
+       "  '.',\n",
+       "  'earth',\n",
+       "  '’',\n",
+       "  'magnetic',\n",
+       "  'field',\n",
+       "  'serves',\n",
+       "  'deflect',\n",
+       "  'solar',\n",
+       "  'wind',\n",
+       "  ',',\n",
+       "  'whose',\n",
+       "  'charged',\n",
+       "  'particles',\n",
+       "  'would',\n",
+       "  'otherwise',\n",
+       "  'strip',\n",
+       "  'away',\n",
+       "  'ozone',\n",
+       "  'layer',\n",
+       "  ',',\n",
+       "  'extends',\n",
+       "  '15',\n",
+       "  'km',\n",
+       "  '35',\n",
+       "  'km',\n",
+       "  'altitude',\n",
+       "  'protects',\n",
+       "  'earth',\n",
+       "  'harmful',\n",
+       "  'x',\n",
+       "  'rays',\n",
+       "  'ultraviolet',\n",
+       "  'radiation',\n",
+       "  '.',\n",
+       "  'the',\n",
+       "  'atmosphere',\n",
+       "  'earth',\n",
+       "  'protects',\n",
+       "  'life',\n",
+       "  'earth',\n",
+       "  'creating',\n",
+       "  'sufficient',\n",
+       "  'pressure',\n",
+       "  'allow',\n",
+       "  'liquid',\n",
+       "  'water',\n",
+       "  'exist',\n",
+       "  'earth',\n",
+       "  '’',\n",
+       "  'surface',\n",
+       "  ';',\n",
+       "  'absorbing',\n",
+       "  'ultraviolet',\n",
+       "  'radiation',\n",
+       "  ';',\n",
+       "  'warming',\n",
+       "  'surface',\n",
+       "  'heat',\n",
+       "  'retention',\n",
+       "  'reducing',\n",
+       "  'temperature',\n",
+       "  'extremes',\n",
+       "  'day',\n",
+       "  'night',\n",
+       "  '.',\n",
+       "  '3',\n",
+       "  '.',\n",
+       "  'energy',\n",
+       "  'flow',\n",
+       "  'earth',\n",
+       "  'atmosphere',\n",
+       "  'the',\n",
+       "  'earth',\n",
+       "  '’',\n",
+       "  'climate',\n",
+       "  'determined',\n",
+       "  'primarily',\n",
+       "  'energy',\n",
+       "  'flows',\n",
+       "  'sun',\n",
+       "  ',',\n",
+       "  'earth',\n",
+       "  '’',\n",
+       "  'surface',\n",
+       "  'atmosphere',\n",
+       "  '.',\n",
+       "  'it',\n",
+       "  'important',\n",
+       "  'note',\n",
+       "  ',',\n",
+       "  'stable',\n",
+       "  'temperature',\n",
+       "  'conditions',\n",
+       "  ',',\n",
+       "  'total',\n",
+       "  'energy',\n",
+       "  'received',\n",
+       "  'radiant',\n",
+       "  'form',\n",
+       "  'sun',\n",
+       "  'must',\n",
+       "  'balanced',\n",
+       "  'total',\n",
+       "  'energy',\n",
+       "  'outgoing',\n",
+       "  'radiation',\n",
+       "  'emitted',\n",
+       "  'space',\n",
+       "  '.',\n",
+       "  'in',\n",
+       "  'diagram',\n",
+       "  ',',\n",
+       "  'energy',\n",
+       "  'flows',\n",
+       "  'average',\n",
+       "  'values',\n",
+       "  'surface',\n",
+       "  'earth',\n",
+       "  'period',\n",
+       "  'year',\n",
+       "  '.',\n",
+       "  'but',\n",
+       "  'need',\n",
+       "  'concern',\n",
+       "  'unduly',\n",
+       "  'numerical',\n",
+       "  'values',\n",
+       "  'diagram',\n",
+       "  '!',\n",
+       "  'section',\n",
+       "  '7',\n",
+       "  'details',\n",
+       "  'range',\n",
+       "  'electromagnetic',\n",
+       "  'radiation',\n",
+       "  'wavelengths',\n",
+       "  'involved',\n",
+       "  '.',\n",
+       "  'in',\n",
+       "  'general',\n",
+       "  'terms',\n",
+       "  ',',\n",
+       "  'radiation',\n",
+       "  'sun',\n",
+       "  ',',\n",
+       "  'high',\n",
+       "  'temperature',\n",
+       "  ',',\n",
+       "  'shortwave',\n",
+       "  'category',\n",
+       "  ',',\n",
+       "  'i.e',\n",
+       "  '.',\n",
+       "  'short',\n",
+       "  'wavelength',\n",
+       "  ',',\n",
+       "  'high',\n",
+       "  'energy',\n",
+       "  'content',\n",
+       "  '.',\n",
+       "  'the',\n",
+       "  'surface',\n",
+       "  'land',\n",
+       "  'areas',\n",
+       "  'oceans',\n",
+       "  'lower',\n",
+       "  'temperature',\n",
+       "  ',',\n",
+       "  'therefore',\n",
+       "  'emit',\n",
+       "  'radiation',\n",
+       "  'longer',\n",
+       "  'wavelengths',\n",
+       "  '(',\n",
+       "  'longwave',\n",
+       "  'radiation',\n",
+       "  ')',\n",
+       "  'lower',\n",
+       "  'energy',\n",
+       "  'content',\n",
+       "  '.',\n",
+       "  'note',\n",
+       "  'made',\n",
+       "  'amount',\n",
+       "  'infrared',\n",
+       "  'radiation',\n",
+       "  '(',\n",
+       "  'shown',\n",
+       "  'back',\n",
+       "  'radiation',\n",
+       "  ')',\n",
+       "  'directed',\n",
+       "  'atmosphere',\n",
+       "  'back',\n",
+       "  'surface',\n",
+       "  ',',\n",
+       "  'keeping',\n",
+       "  'us',\n",
+       "  'warm',\n",
+       "  '.',\n",
+       "  'average',\n",
+       "  'global',\n",
+       "  'energy',\n",
+       "  'flows',\n",
+       "  'on',\n",
+       "  'short',\n",
+       "  'timescale',\n",
+       "  ',',\n",
+       "  'distribution',\n",
+       "  'incoming',\n",
+       "  'energy',\n",
+       "  'sun',\n",
+       "  'influenced',\n",
+       "  'patterns',\n",
+       "  'circulating',\n",
+       "  'wind',\n",
+       "  'currents',\n",
+       "  'around',\n",
+       "  'globe',\n",
+       "  '.',\n",
+       "  'these',\n",
+       "  'patterns',\n",
+       "  'determined',\n",
+       "  'rotation',\n",
+       "  'earth',\n",
+       "  'around',\n",
+       "  'axis',\n",
+       "  '(',\n",
+       "  'see',\n",
+       "  'diagram',\n",
+       "  ')',\n",
+       "  '.',\n",
+       "  'it',\n",
+       "  'apparent',\n",
+       "  'atmospheric',\n",
+       "  'wind',\n",
+       "  'patterns',\n",
+       "  'northern',\n",
+       "  'hemisphere',\n",
+       "  'essentially',\n",
+       "  'separated',\n",
+       "  'southern',\n",
+       "  'hemisphere',\n",
+       "  '.',\n",
+       "  'on',\n",
+       "  'longer',\n",
+       "  'timescale',\n",
+       "  ',',\n",
+       "  'energy',\n",
+       "  'transferred',\n",
+       "  'patterns',\n",
+       "  'ocean',\n",
+       "  'currents',\n",
+       "  '.',\n",
+       "  'whilst',\n",
+       "  'also',\n",
+       "  'driven',\n",
+       "  'rotation',\n",
+       "  'earth',\n",
+       "  ',',\n",
+       "  'contained',\n",
+       "  'boundaries',\n",
+       "  'land',\n",
+       "  'masses',\n",
+       "  'depths',\n",
+       "  'oceans',\n",
+       "  '.',\n",
+       "  'major',\n",
+       "  'wind',\n",
+       "  'patterns',\n",
+       "  'major',\n",
+       "  'ocean',\n",
+       "  'currents',\n",
+       "  '4',\n",
+       "  '.',\n",
+       "  'the',\n",
+       "  'constituents',\n",
+       "  'atmosphere',\n",
+       "  'let',\n",
+       "  'us',\n",
+       "  'look',\n",
+       "  'component',\n",
+       "  'gases',\n",
+       "  'make',\n",
+       "  'atmosphere',\n",
+       "  '.',\n",
+       "  'the',\n",
+       "  'atmosphere',\n",
+       "  '(',\n",
+       "  'dry',\n",
+       "  'air',\n",
+       "  ')',\n",
+       "  'following',\n",
+       "  'major',\n",
+       "  'constituent',\n",
+       "  'gases',\n",
+       "  ':',\n",
+       "  'nitrogen',\n",
+       "  ':',\n",
+       "  '781',\n",
+       "  ',',\n",
+       "  '000',\n",
+       "  'parts',\n",
+       "  'per',\n",
+       "  'million',\n",
+       "  'volume',\n",
+       "  '(',\n",
+       "  'ppm',\n",
+       "  'average',\n",
+       "  ')',\n",
+       "  'oxygen',\n",
+       "  ':',\n",
+       "  '209,000',\n",
+       "  'ppm',\n",
+       "  'av',\n",
+       "  'argon',\n",
+       "  ':',\n",
+       "  '9,300',\n",
+       "  'ppm',\n",
+       "  'av',\n",
+       "  'carbon',\n",
+       "  'dioxide',\n",
+       "  ':',\n",
+       "  '400',\n",
+       "  'ppm',\n",
+       "  'av',\n",
+       "  'neon',\n",
+       "  ':',\n",
+       "  '18',\n",
+       "  'ppm',\n",
+       "  'av',\n",
+       "  'helium',\n",
+       "  ':',\n",
+       "  '5',\n",
+       "  'ppm',\n",
+       "  'av',\n",
+       "  'methane',\n",
+       "  ':',\n",
+       "  '2',\n",
+       "  'ppm',\n",
+       "  'av',\n",
+       "  'these',\n",
+       "  'concentrations',\n",
+       "  'maintained',\n",
+       "  'steady',\n",
+       "  '10',\n",
+       "  'km',\n",
+       "  'altitude',\n",
+       "  '.',\n",
+       "  'the',\n",
+       "  'original',\n",
+       "  'presence',\n",
+       "  'nitrogen',\n",
+       "  'carbon',\n",
+       "  'dioxide',\n",
+       "  'atmosphere',\n",
+       "  'thought',\n",
+       "  'arisen',\n",
+       "  'volcanism',\n",
+       "  ',',\n",
+       "  'together',\n",
+       "  'impact',\n",
+       "  'huge',\n",
+       "  'asteroids',\n",
+       "  '(',\n",
+       "  'carrying',\n",
+       "  'ammonia',\n",
+       "  ')',\n",
+       "  'early',\n",
+       "  'earth',\n",
+       "  '’',\n",
+       "  'history',\n",
+       "  '.',\n",
+       "  'the',\n",
+       "  'oxygen',\n",
+       "  'originates',\n",
+       "  'form',\n",
+       "  'living',\n",
+       "  'organisms',\n",
+       "  '(',\n",
+       "  'algae',\n",
+       "  ',',\n",
+       "  'plankton',\n",
+       "  'plant',\n",
+       "  'life',\n",
+       "  'thre',\n",
+       "  'land',\n",
+       "  'areas',\n",
+       "  ')',\n",
+       "  ',',\n",
+       "  'via',\n",
+       "  'process',\n",
+       "  'photosynthesis',\n",
+       "  '.',\n",
+       "  'oxygen',\n",
+       "  'thought',\n",
+       "  'first',\n",
+       "  'produced',\n",
+       "  'around',\n",
+       "  '2.8',\n",
+       "  'billion',\n",
+       "  'years',\n",
+       "  'ago',\n",
+       "  '.',\n",
+       "  'in',\n",
+       "  'addition',\n",
+       "  ',',\n",
+       "  'water',\n",
+       "  'vapour',\n",
+       "  'present',\n",
+       "  '0.25',\n",
+       "  '%',\n",
+       "  'mass',\n",
+       "  'full',\n",
+       "  'atmosphere',\n",
+       "  '.',\n",
+       "  'locally',\n",
+       "  'vary',\n",
+       "  '100',\n",
+       "  'ppm',\n",
+       "  '4200ppm',\n",
+       "  'volume',\n",
+       "  ',',\n",
+       "  'depending',\n",
+       "  'local',\n",
+       "  'temperature',\n",
+       "  '.',\n",
+       "  'its',\n",
+       "  'concentration',\n",
+       "  'highest',\n",
+       "  'tropical',\n",
+       "  'latitudes',\n",
+       "  ',',\n",
+       "  'varies',\n",
+       "  'strongly',\n",
+       "  'locally',\n",
+       "  'throughout',\n",
+       "  '.',\n",
+       "  'the',\n",
+       "  'presence',\n",
+       "  'water',\n",
+       "  'vapour',\n",
+       "  'air',\n",
+       "  'naturally',\n",
+       "  'dilutes',\n",
+       "  'displaces',\n",
+       "  'air',\n",
+       "  'components',\n",
+       "  'concentration',\n",
+       "  'increases',\n",
+       "  '.',\n",
+       "  'water',\n",
+       "  'vapour',\n",
+       "  'lower',\n",
+       "  'density',\n",
+       "  'air',\n",
+       "  'therefore',\n",
+       "  'buoyant',\n",
+       "  'atmosphere',\n",
+       "  '.',\n",
+       "  'its',\n",
+       "  'mean',\n",
+       "  'global',\n",
+       "  'content',\n",
+       "  'roughly',\n",
+       "  'sufficient',\n",
+       "  'cover',\n",
+       "  'entire',\n",
+       "  'surface',\n",
+       "  'planet',\n",
+       "  'layer',\n",
+       "  'liquid',\n",
+       "  'depth',\n",
+       "  '25',\n",
+       "  'mm',\n",
+       "  '.',\n",
+       "  'approximately',\n",
+       "  '99',\n",
+       "  '%',\n",
+       "  'earth',\n",
+       "  '’',\n",
+       "  'water',\n",
+       "  'vapour',\n",
+       "  'contained',\n",
+       "  'within',\n",
+       "  'troposphere',\n",
+       "  '.',\n",
+       "  'its',\n",
+       "  'condensation',\n",
+       "  'liquid',\n",
+       "  'solid',\n",
+       "  'form',\n",
+       "  'responsible',\n",
+       "  'clouds',\n",
+       "  ',',\n",
+       "  'rain',\n",
+       "  ',',\n",
+       "  'snow',\n",
+       "  'precipitation',\n",
+       "  ',',\n",
+       "  'count',\n",
+       "  'amongst',\n",
+       "  'significant',\n",
+       "  'elements',\n",
+       "  'experience',\n",
+       "  'weather',\n",
+       "  '.',\n",
+       "  'less',\n",
+       "  'obviously',\n",
+       "  ',',\n",
+       "  'latent',\n",
+       "  'heat',\n",
+       "  'vaporisation',\n",
+       "  ',',\n",
+       "  'released',\n",
+       "  'atmosphere',\n",
+       "  'whenever',\n",
+       "  'condensation',\n",
+       "  'occurs',\n",
+       "  ',',\n",
+       "  'one',\n",
+       "  'important',\n",
+       "  'terms',\n",
+       "  'atmospheric',\n",
+       "  'energy',\n",
+       "  'budget',\n",
+       "  'local',\n",
+       "  'global',\n",
+       "  'scales',\n",
+       "  '.',\n",
+       "  'for',\n",
+       "  'example',\n",
+       "  ',',\n",
+       "  'latent',\n",
+       "  'heat',\n",
+       "  'release',\n",
+       "  ...],\n",
+       " 'perplexity': None,\n",
+       " 'publication': {},\n",
+       " 'clue_words': ['volume', 'nature', 'article'],\n",
+       " 'clue_links': [],\n",
        " 'science': False,\n",
-       " 'uniqueness': 0.5913978494623656,\n",
-       " 'info_density': 0.012151898734177215,\n",
-       " 'scaled_info_density': -4.064180178654587e-05,\n",
-       " 'sp': 0.1776315789473684,\n",
-       " 'ss': 0.5230263157894737,\n",
-       " 'gf': 18.021016949152543,\n",
-       " 'standard': 8.0,\n",
-       " 'penalty': 8.700657894736842}"
+       " 'wiki': False,\n",
+       " 'uniqueness': 0.37306172088780787,\n",
+       " 'info_density': 0.0006726645926175061,\n",
+       " 'sp': 0.0730123687858983,\n",
+       " 'ss': 0.4008192815898699,\n",
+       " 'sp_norm': 0.0730123687858983,\n",
+       " 'ss_norm': 0.4008192815898699,\n",
+       " 'gf': 15.72,\n",
+       " 'standard': 16.0,\n",
+       " 'scaled': 0.004453103256331756,\n",
+       " 'conciseness': 8.03688862445894,\n",
+       " 'penalty': 8.01378057590509}"
       ]
      },
      "execution_count": 20,
diff --git a/Publication/manuscript.docx b/Publication/manuscript.docx
deleted file mode 100644
index e9b0360035ec9e7bd990d5dbf14300d457aecf0e..0000000000000000000000000000000000000000
Binary files a/Publication/manuscript.docx and /dev/null differ
diff --git a/Publication/project_plan.docx b/Publication/project_plan.docx
deleted file mode 100644
index 0342955752165d62134c3c2307266c5fcfb704cc..0000000000000000000000000000000000000000
Binary files a/Publication/project_plan.docx and /dev/null differ
diff --git a/Publication/tex_manuscript/alocal.sty b/Publication/tex_manuscript/alocal.sty
deleted file mode 100644
index f9df2658b359af7d1b8b14fe5a867912afbe5f96..0000000000000000000000000000000000000000
--- a/Publication/tex_manuscript/alocal.sty
+++ /dev/null
@@ -1,20 +0,0 @@
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-% alocal.sty
-% for Sandra Kuebler
-% 2011/05/29 02:20 CET
-% Klaus Lagally
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-% patches for use with class CLV2
-% put this file into the working directory
-
-\gdef\n@tr@ce
-{\edef \dotr@ce {\nxp \tracingmacros \the \tracingmacros \relax }%
-	\tracingmacros 0\relax } 
-
-\gdef\notr@ce
-{\protect\n@tr@ce}
-
-\long\gdef \@gobble #1{}
-
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-\endinput
diff --git a/Publication/tex_manuscript/clv3.cls b/Publication/tex_manuscript/clv3.cls
deleted file mode 100644
index fcf5ad46b85e3eb3215975a854f5a7ef000e3823..0000000000000000000000000000000000000000
--- a/Publication/tex_manuscript/clv3.cls
+++ /dev/null
@@ -1,2063 +0,0 @@
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-%       This File               : clv3.cls
-%       Version         : 3
-%
-%       Developed By    : Krishan Gopal Goswami
-%       Date                    : 21-Oct-2005
-%
-%       Developed for   : SPI Publisher Services
-%       Copyright (c)   : 
-%
-%       Remarks         : This is based on MIT - Computational Linguistics
-%                         Standard Typesetting Instructions
-%
-%       Note            : 
-%
-%       1) Do not make any change in this file with out prior information
-%       2) Update History for the changes in the format given below
-%       
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-%       Version : V1.2
-%
-%       Updated By      : Narayan Piyush
-%       Date            : 26-Nov-2005
-%
-%       Updated for     : 
-%
-%       1) Italic Greek (Lowercase & Uppercase) Characters
-%       
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-%       Version : V3
-%
-%       Updated By      : Daniel Gildea
-%       Date            : 15-Jun-2016
-%
-%       Updated for     : 
-%
-%       1) use natbib.  Fixes incompatibility with hyperref, for clickable pdfs.
-%		allows for more flexible citation commands.  Requires use
-%		use compling.bst in place of fullname.bst.
-%	2) use amsthm.  Fixes problem with no demarcation of the end
-%		of a theorem/lemma/proposition.  Fixes problem with small extra 
-%		space at beginning of a theorem/lemma/proposition
-%		if the \begin{theorem} is followed by a \label command.
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-% Fonts: 
-%
-% Palatino
-% Helvetica
-%       
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-% Standard Sizes: 
-%
-% normalsize -- 10/12
-% footnotesize -- 8/9
-% small -- 9/10
-% scriptsize -- 7/8
-% tiny -- 5/6
-% large -- 12/13
-% Large -- 16/20
-% LARGE -- 17/20
-% huge -- 20/25
-% Huge -- 25/30
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-\NeedsTeXFormat{LaTeX2e}
-\ProvidesClass{clv3}[2016/06/15 v2 LaTeX document class for MIT - Computational Linguistics Journals]
-%
-%\xdef\jobtag{MIT --- Computational Linguistics\qquad (Typeset by spi publisher services, Delhi)}%
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Job Options %%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-\newdimen\draftrule\draftrule0pt
-\newdimen\trimrule\trimrule.1pt
-%
-\newif\ifindex
-\newif\ifdiscussion
-\newif\ifbookreview
-\newif\ifbrief
-\newif\ifpubrec
-\newif\ifshortpaper
-\newif\ifmanuscript
-%
-\DeclareOption{manuscript}{\AtBeginDocument{\setlength{\baselineskip}{2\baselineskip}\global\manuscripttrue}}
-\DeclareOption{discussion}{\discussiontrue}
-\DeclareOption{bookreview}{\bookreviewtrue}
-\DeclareOption{brief}{\brieftrue\shortpapertrue}
-\DeclareOption{pubrec}{\pubrectrue\shortpapertrue}
-\DeclareOption{index}{\indextrue}
-\DeclareOption{final}{}
-\ExecuteOptions{final}
-\ProcessOptions
-%
-\@twosidetrue\@mparswitchtrue\ifshortpaper\@twocolumntrue\else\@twocolumnfalse\fi
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Variable Declarations %%%%%%%%%%%%%%%%%%%%%
-%
-% \newlength Declarations
-%
-\newlength\trimwidth
-\newlength\trimheight
-\newlength\typewidth
-\newlength\typeheight
-\newlength\normaltextheight
-\newlength\blindfoliodrop
-\newlength\figheight
-\newlength\figwidth
-\newlength\tabledim
-%
-%%%%% \newdimen Declarations %%%%%
-%
-\newdimen\tempdimen
-\newdimen\enumdim
-\newdimen\mathindent
-\newdimen\emathindent
-\newdimen\bibindent
-%
-% \newskip Declarations
-%
-\newskip\normalbaselineskip
-\newskip\tableleftskip
-\newskip\tablerightskip
-%
-%%%%% \newbox Declarations %%%%%
-%
-\newbox\tempbox
-%
-%%%%% \newif Declarations %%%%%
-%
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%% Variable Initialization %%%%%%%%%%%%%%%%%%%%%%
-%
-%
-%%
-\setlength\trimheight{10in}
-\setlength\trimwidth{6.75in}
-%
-\setlength\typeheight{56pc}
-%
-\setlength\headheight{6.5\p@}%
-\setlength\headsep   {31pt}%
-\setlength\topskip   {7\p@}%
-%
-\setlength\textheight{49.61pc}%
-%\addtolength\textheight{\topskip}% \textheight = 49\baselineskip + \topskip
-%
-\setlength\normaltextheight{\textheight}
-\setlength\textwidth{32pc}
-%
-\setlength\typewidth{\textwidth}
-%
-\setlength\topmargin{26.7pt}%25.35pt}
-\setlength\oddsidemargin{54pt}
-%
-\setlength\evensidemargin\trimwidth
-\addtolength\evensidemargin{-\textwidth}
-\addtolength\evensidemargin{-\oddsidemargin}
-%\addtolength\oddsidemargin{4.5pc}
-\setlength\footskip{36pt}
-\setlength\maxdepth{4\p@}
-%
-\setlength\blindfoliodrop{\trimheight}
-\addtolength\blindfoliodrop{-\typeheight}
-\addtolength\blindfoliodrop{-\topmargin}
-%\addtolength\blindfoliodrop{-\footskip}
-\addtolength\blindfoliodrop{18pt}
-%
-\ifshortpaper
-   \setlength\parindent{9pt}
-\else
-   \setlength\parindent{18pt}
-\fi
-%
-\setlength\marginparwidth {5pc}
-\setlength\marginparsep{6\p@}
-\setlength\marginparpush{5\p@}
-%
-\setlength\footnotesep{6.65\p@}
-\setlength{\skip\footins}{23\p@ \@plus 8\p@}% \@minus 8\p@}%
-\skip\@mpfootins = \skip\footins
-%
-\setlength\floatsep    {19\p@ \@plus 2\p@}% \@minus 2\p@}
-\setlength\textfloatsep{30\p@ \@plus 2\p@}% \@minus 4\p@}
-\setlength\intextsep   {12\p@ \@plus 2\p@}% \@minus 2\p@}
-\setlength\dblfloatsep    {12\p@ \@plus 2\p@ \@minus 2\p@}
-\setlength\dbltextfloatsep{20\p@ \@plus 2\p@ \@minus 4\p@}
-\setlength\@fptop{0\p@}
-\setlength\@fpsep{8\p@ \@plus 1fil}
-\setlength\@fpbot{0\p@ \@plus 1fil}
-\setlength\@dblfptop{0\p@}
-\setlength\@dblfpsep{8\p@ \@plus 1fil}
-\setlength\@dblfpbot{0\p@ \@plus 1fil}
-%
-\setlength\partopsep{0pt}
-\setlength\lineskip{1\p@}% check if it can be flexible
-\setlength\normallineskip{1\p@}%
-\renewcommand\baselinestretch{}
-\ifpubrec
-   \setlength\parskip{10pt plus.1pt}%{0\p@ \@plus \p@}
-\else
-   \setlength\parskip{\z@}%{0\p@ \@plus \p@}
-\fi
-\@lowpenalty   51
-\@medpenalty  151
-\@highpenalty 301
-%
-\@beginparpenalty -\@lowpenalty
-\@endparpenalty   -\@lowpenalty
-\@itempenalty     -\@lowpenalty
-%
-\clubpenalty\@M
-\widowpenalty\@M
-\hyphenpenalty400
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%% Fonts Size Definitions %%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-\def\@viiipt{8}
-\def\@ixpt{9}
-\def\@xpt{10}
-\def\@xhpt{10.5}
-\def\@xiipt{12}
-\def\@xvpt{15}
-\def\@xvipt{16}
-\def\@xviipt{17}
-\def\@xviiipt{18}
-\def\@xxivpt{24}
-%
-\normalbaselineskip12pt
-%
-\ifpubrec
- \renewcommand\normalsize{%
-   \@setfontsize\normalsize\@ixpt{10}%
-   \abovedisplayskip 11\p@ \@plus2\p@% \@minus\p@
-   \belowdisplayskip      \abovedisplayskip
-   \abovedisplayshortskip \abovedisplayskip
-   \belowdisplayshortskip \abovedisplayskip
-   \let\@listi\@listI}
-\else
-\ifbrief
- \renewcommand\normalsize{%
-   \@setfontsize\normalsize\@ixpt{11}%
-   \abovedisplayskip 11\p@ \@plus2\p@% \@minus\p@
-   \belowdisplayskip      \abovedisplayskip
-   \abovedisplayshortskip \abovedisplayskip
-   \belowdisplayshortskip \abovedisplayskip
-   \let\@listi\@listI}
-\else
- \renewcommand\normalsize{%
-   \@setfontsize\normalsize\@xpt{12}%
-   \abovedisplayskip 12\p@ \@plus2\p@% \@minus\p@
-   \belowdisplayskip      \abovedisplayskip
-   \abovedisplayshortskip \abovedisplayskip
-   \belowdisplayshortskip \abovedisplayskip
-   \let\@listi\@listI}
-\fi\fi
-\def\biggg#1{{\hbox{$\left#1\vbox to20.5\p@{}\right.\n@space$}}}
-\def\Biggg#1{{\hbox{$\left#1\vbox to23.5\p@{}\right.\n@space$}}}
-\normalsize
-%
-\newcommand\small{%
-   \@setfontsize\small\@ixpt{10}%
-   \abovedisplayskip 10\p@ \@plus2\p@% \@minus\p@
-   \belowdisplayskip      \abovedisplayskip
-   \abovedisplayshortskip \abovedisplayskip
-   \belowdisplayshortskip \abovedisplayskip
-   \def\@listi{\leftmargin\leftmargini
-               \topsep 5\p@ \@plus2\p@% \@minus\p@
-               \parsep 0\p@% \@plus\p@% \@minus\p@
-               \itemsep \parsep}%
-   \abovedisplayshortskip\abovedisplayskip
-   \belowdisplayshortskip\abovedisplayshortskip
-   \belowdisplayskip \abovedisplayskip
-   \setSmallDelims}
-%
-\def\setSmallDelims{%
-\def\big##1{{\hbox{$\left##1\vbox to7.5\p@{}\right.\n@space$}}}%
-\def\Big##1{{\hbox{$\left##1\vbox to10.5\p@{}\right.\n@space$}}}%
-\def\bigg##1{{\hbox{$\left##1\vbox to13.5\p@{}\right.\n@space$}}}%
-\def\Bigg##1{{\hbox{$\left##1\vbox to16.5\p@{}\right.\n@space$}}}%
-\def\biggg##1{{\hbox{$\left##1\vbox to19.5\p@{}\right.\n@space$}}}%
-\def\Biggg##1{{\hbox{$\left##1\vbox to22.5\p@{}\right.\n@space$}}}%
-}
-%
-\newcommand\footnotesize{%
-   \@setfontsize\footnotesize\@viiipt{9}%
-   \abovedisplayskip 9\p@ \@plus2\p@% \@minus\p@
-   \belowdisplayskip      \abovedisplayskip
-   \abovedisplayshortskip \abovedisplayskip
-   \belowdisplayshortskip \abovedisplayskip
-   \def\@listi{\leftmargin\leftmargini
-               \topsep 4.5\p@ \@plus\p@% \@minus\p@
-               \parsep 0\p@% \@plus\p@% \@minus\p@
-               \itemsep \parsep}%
-   \setFootnotesizeDelims}
-%
-\def\setFootnotesizeDelims{%
-\def\big##1{{\hbox{$\left##1\vbox to6.5\p@{}\right.\n@space$}}}%
-\def\Big##1{{\hbox{$\left##1\vbox to9.5\p@{}\right.\n@space$}}}%
-\def\bigg##1{{\hbox{$\left##1\vbox to12.5\p@{}\right.\n@space$}}}%
-\def\Bigg##1{{\hbox{$\left##1\vbox to15.5\p@{}\right.\n@space$}}}%
-\def\biggg##1{{\hbox{$\left##1\vbox to18.5\p@{}\right.\n@space$}}}%
-\def\Biggg##1{{\hbox{$\left##1\vbox to21.5\p@{}\right.\n@space$}}}%
-}
-%   
-\newcommand\tiny{\@setfontsize\tiny\@vpt{6}}
-\newcommand\scriptsize{\@setfontsize\scriptsize\@viipt{8}}
-\newcommand\large{\@setfontsize\large\@xiipt{13}}
-\newcommand\Large{\@setfontsize\Large\@xvipt{20}}
-\newcommand\LARGE{\@setfontsize\LARGE\@xviipt{20}}
-\newcommand\huge{\@setfontsize\huge\@xxpt{25}}
-\newcommand\Huge{\@setfontsize\Huge\@xxvpt{30}}
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-%% headings
-\def\rhfont{\small}
-\def\rffont{\footnotesize}
-\def\foliofont{\normalsize}
-\def\dropfoliofont{\normalsize}
-%% Title
-\def\sptitlefont{\LARGE\bfseries\ifdiscussion\rightskip4pc plus1fill\else\raggedright\fi\mathversion{bold}}
-\ifbrief
-   \def\titlefont{\fontsize{10}{11}\selectfont\bfseries\raggedright\mathversion{bold}}
-\else\ifbookreview
-   \def\titlefont{\fontsize{12}{12}\selectfont\bfseries\raggedright\mathversion{bold}}
-\else\ifdiscussion
-   \def\titlefont{\Large\bfseries\raggedright\mathversion{bold}}
-\else
-   \def\titlefont{\Large\bfseries\rightskip4pc plus1fill\mathversion{bold}}
-\fi\fi\fi
-\def\subtitlefont{\large\bfseries}
-%% Author/Affil
-\ifbrief
-   \def\authorfont{\small\bfseries\raggedright}
-\else\ifbookreview
-   \def\authorfont{\normalsize\bfseries\raggedright}
-\else
-   \def\authorfont{\large\raggedright}
-\fi\fi
-\ifbrief
-   \def\affilfont{\small\raggedright}
-\else
-   \def\affilfont{\normalsize\raggedright}
-\fi
-\ifbrief
-   \def\pubinfofont{\small\raggedright}
-\else
-   \def\pubinfofont{\normalsize\raggedright}
-\fi
-\def\reviewerfont{\normalsize\itshape}
-\ifbrief
-   \def\biofont{\normalsize}
-\else
-   \def\biofont{\small}%\itshape}
-\fi
-%%
-\def\abstractfont{\normalsize\setlength\baselineskip{13pt}\itshape}
-\def\keywordfont{\normalsize\raggedright}
-%% section
-\def\sectionfont{\normalsize\bfseries\mathversion{bold}}
-\def\sectionnumfont{\normalsize\bfseries}
-\def\subsectionfont{\normalsize\bfseries\mathversion{bold}}
-\def\subsectionnumfont{\normalsize\bfseries}
-\def\subsubsectionfont{\normalsize\bfseries\mathversion{bold}}
-\def\paragraphfont{\normalsize\itshape}%
-\def\subparagraphfont{}%
-\def\subsubparagraphfont{}%
-\def\xheadfont{\normalsize\bfseries}
-%% figure
-\def\figcaptionfont{\small\raggedright}%\mathversion{sfnormal}}
-\def\figcaptionnumfont{\small\bfseries}
-\def\figsourcefont{\reset@font\fontsize{8.5}{10}\selectfont}
-%% table
-\def\tablefont{\small}%\leftskip\tableleftskip\rightskip\tablerightskip use plus 1fill if needed
-\def\tablecaptionfont{\small\raggedright}%\mathversion{sfnormal}}%
-\def\tablecaptionnumfont{\small\bfseries}%
-\def\TCHfont{\small}%
-\def\tabnotefont{\leftskip\tableleftskip\rightskip\tablerightskip}% use plus 1fill if needed
-%% BM
-\def\indexfont{\fontsize{9}{10}\selectfont\raggedright}
-\def\ackfont{\small\raggedright}
-%
-\def\listfont{\raggedright}
-\def\listdevicefont{}
-%
-\ifbrief
-   \def\extractfont{\small\leftskip2pc}
-\else\ifbookreview
-   \def\extractfont{\normalsize\itshape}
-\else
-   \def\extractfont{\small\leftskip1.5pc\rightskip1.5pc plus1fill}%
-\fi\fi
-\def\sourcefont{\reset@font\normalsize}
-%
-%
-%%%%%%%  For Times family  %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-\DeclareOldFontCommand{\rm}{\normalfont\rmfamily}{\mathrm}
-\DeclareOldFontCommand{\sf}{\normalfont\sffamily}{\mathsf}
-\DeclareOldFontCommand{\tt}{\normalfont\ttfamily}{\mathtt}
-\DeclareOldFontCommand{\bf}{\normalfont\bfseries}{\mathbf}
-\DeclareOldFontCommand{\it}{\normalfont\itshape}{\mathit}
-\DeclareOldFontCommand{\sl}{\normalfont\slshape}{\@nomath\sl}
-\DeclareOldFontCommand{\sc}{\normalfont\scshape}{\@nomath\sc}
-\DeclareOldFontCommand{\bi}{\bfseries\itshape}{\bfseries\itshape}
-\newcommand{\cal}{\protect\pcal}
-\newcommand{\pcal}{\@fontswitch{\relax}{\mathcal}}
-\newcommand{\mit}{\protect\pmit}
-\newcommand{\pmit}{\@fontswitch{\relax}{\mathnormal}}
-%
-%\renewcommand\rmdefault{Times}
-%\newcommand\rmmathdefault{TimesMath}
-%
-%\renewcommand\sfdefault{Officina}
-%\newcommand\sfmathdefault{HelveticaMath}
-%\renewcommand{\ttdefault}{Courier}
-%%
-\def\scitdefault{scit}% All scit
-\def\capsitdefault{capsit}% all capsit 
-%
-\def\capsdefault{caps}% all caps
-\DeclareRobustCommand\capsshape
-        {\not@math@alphabet\capsshape\mathrm
-         \ifx\f@shape\itdefaultabbrev\fontshape\capsitdefault\else\fontshape\capsdefault\fi\selectfont}
-%
-\def\itdefaultabbrev{it}
-\DeclareRobustCommand\itshape
-        {\not@math@alphabet\itshape\mathit
-         \ifx\f@shape\scdefaultabbrev\fontshape\scitdefault\else\ifx\f@shape\capsdefault\fontshape\capsitdefault\else\fontshape\itdefault\fi\fi\selectfont}
-%
-\def\scdefaultabbrev{sc}
-\DeclareRobustCommand\scshape
-        {\not@math@alphabet\scshape\relax
-         \ifx\f@shape\itdefaultabbrev\fontshape\scitdefault\else\fontshape\scdefault\fi\selectfont}
-%
-%%%%%%%%%%%%%%%%%%%% Times %%%%%%%%%%%%%%%%%%
-%
-\DeclareFontFamily{OML}{TimesMath}{\skewchar\font127 }
-\DeclareFontShape{OML}{TimesMath}{m}{it}{<-> MTMI }{}
-\DeclareFontShape{OML}{TimesMath}{bx}{it}{<->  MTMIB }{}
-\DeclareFontShape{OML}{TimesMath}{b}{it}{<->ssub*TimesMath/bx/it}{}
-%
-\DeclareFontFamily{OMS}{TimesMath}{\skewchar\font48 }
-\DeclareFontShape{OMS}{TimesMath}{m}{n}{<->  MTSY }{}
-\DeclareFontShape{OMS}{TimesMath}{bx}{n}{<->  MTSYB }{}
-%
-\DeclareFontFamily{OMX}{TimesMath}{}
-\DeclareFontShape{OMX}{TimesMath}{m}{n}{<->  MTEX }{}
-%
-\DeclareFontFamily{OT1}{Times}{}
-\DeclareFontShape{OT1}{Times}{m}{n}{ <-> ptmr }{}
-\DeclareFontShape{OT1}{Times}{m}{it}{ <-> ptmri }{}
-%
-\DeclareFontShape{OT1}{Times}{bx}{n}{ <-> ptmb }{}
-\DeclareFontShape{OT1}{Times}{bx}{it}{ <-> ptmbi }{}
-%
-%%%%%%%%%%%%%%%%%%%% Helvetica %%%%%%%%%%%%%%%%%%
-%
-\DeclareFontFamily{OML}{HelveticaMath}{\skewchar\font127 }
-\DeclareFontShape{OML}{HelveticaMath}{m}{it}{<-> HelMTMI }{}
-\DeclareFontShape{OML}{HelveticaMath}{bx}{it}{<->  HelMTMIB }{}
-\DeclareFontShape{OML}{HelveticaMath}{b}{it}{<->ssub*HelveticaMath/bx/it}{}
-%
-\DeclareFontFamily{OMS}{HelveticaMath}{\skewchar\font48 }
-\DeclareFontShape{OMS}{HelveticaMath}{m}{n}{<->  MTSY }{}
-\DeclareFontShape{OMS}{HelveticaMath}{bx}{n}{<->  MTSYB }{}
-%
-\DeclareFontFamily{OMX}{HelveticaMath}{}
-\DeclareFontShape{OMX}{HelveticaMath}{m}{n}{<->  MTEX }{}
-%
-\DeclareFontFamily{OT1}{Helvetica}{}
-\DeclareFontShape{OT1}{Helvetica}{m}{n}{ <-> phvr }{}
-\DeclareFontShape{OT1}{Helvetica}{m}{it}{ <-> phvro }{}
-\DeclareFontShape{OT1}{Helvetica}{bx}{n}{ <-> phvb }{}
-\DeclareFontShape{OT1}{Helvetica}{bx}{it}{ <-> phvbo }{}
-%
-%
-\DeclareFontFamily{OT1}{ams}{}
-\DeclareFontShape{OT1}{ams}{m}{n}{ <-> msam10 }{}
-\DeclareFontShape{OT1}{ams}{m}{it}{ <-> msam10 }{}
-\DeclareFontShape{OT1}{ams}{bx}{n}{ <-> msbm10 }{}
-\DeclareFontShape{OT1}{ams}{bx}{it}{ <-> msbm10 }{}
-%
-%
-\DeclareFontShape{OMS}{cmsy}{m}{n}{ <-> cmsy10 }{}
-\DeclareFontShape{OMS}{cmsy}{b}{n}{ <-> cmbsy10 }{}
-%
-%
-\newcommand\sfboldmath{\@nomath\sfboldmath\mathversion{sfbold}}
-%%%%% define bold math font %%%%%
-\newcommand\bm[1]{\mathchoice
-   {\mbox{\boldmath$\displaystyle#1$}}%
-   {\mbox{\boldmath$#1$}}%
-   {\mbox{\boldmath$\scriptstyle#1$}}%
-   {\mbox{\boldmath$\scriptscriptstyle#1$}}}
-%
-\usepackage[T1]{fontenc}
-\usepackage{textcomp}
-\usepackage{palatino,helvet}%,zeupplv1}%,mathptm}%
-%
-\DeclareFontFamily{OML}{euppl}{\skewchar \font =127}
-\DeclareFontShape{OML}{euppl}{m}{it}{<8-> pplre7m <6-8> pplre7m7 <-6> pplre7m5}{}
-\DeclareFontShape{OML}{euppl}{m}{sl}{<->ssub * euppl/m/it}{}
-\DeclareFontShape{OML}{euppl}{b}{it}{<-> pplbe7m}{}
-\DeclareFontShape{OML}{euppl}{bx}{it}{<->ssub * euppl/b/it}{}
-%
-\DeclareFontFamily{OML}{euphv}{\skewchar \font =127}
-\DeclareFontShape{OML}{euphv}{m}{it}{<8-> phvre7m <6-8> phvre7m7 <-6> phvre7m5}{}
-\DeclareFontShape{OML}{euphv}{m}{sl}{<->ssub * euphv/m/it}{}
-\DeclareFontShape{OML}{euphv}{b}{it}{<-> phvbe7m}{}
-\DeclareFontShape{OML}{euphv}{bx}{it}{<->ssub * euphv/b/it}{}
-%
-\DeclareFontFamily{U}{bbnum}{}
-\DeclareFontShape{U}{bbnum}{m}{n}{<-> CASLOB}{}
-\DeclareMathAlphabet{\mathbbnum}{U}{bbnum}{m}{n}
-%
-%\DeclareMathSymbol{,}{\mathpunct}{operators}{`,}
-%
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%% General Commands %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-\newcommand\contentsname{Contents}
-\newcommand\listfigurename{List of Figures}
-\newcommand\listtablename{List of Tables}
-\newcommand\bibname{References}
-\newcommand\indexname{Index}
-\newcommand\figurename{Figure}
-\newcommand\tablename{Table}
-\newcommand\appendixname{Appendix}
-%
-\newcommand\today{\ifcase\month\or
-  January\or February\or March\or April\or May\or June\or
-  July\or August\or September\or October\or November\or December\fi
-  \space\number\day, \number\year}
-%
-\newcount\minute
-\newcount\hour
-\def\currenttime{%
-     \minute\time
-     \hour\minute
-     \divide\hour60
-     \the\hour:\multiply\hour60\advance\minute-\hour\the\minute}
-%
-\def\spreadoutfactor{1}
-%
-\def\spreadout#1{%
-  \gdef\temp{#1}\dimen0 = \spreadoutfactor pt
-  \expandafter\dospreadout\temp\endmark\kern-\dimen0}
-%
-\def\dospreadout{%
-  \afterassignment\findospreadout
-  \let\next= }
-%
-\def\findospreadout{%
-   \ifx\next\endmark
-      \let\nextaction = \relax
-   \else
-      \let\nextaction = \dospreadout
-      \next
-      \kern\dimen0
-   \fi
-   \nextaction}
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%% Make Title %%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-\def\jname#1{\gdef\@jname{#1}}
-\def\jinfo#1{\gdef\@jinfo{#1}}
-\def\jvol#1{\gdef\@jvol{#1}}\def\@jvol{xx}
-\def\jnum#1{\gdef\@jnum{#1}}\def\@jnum{xx}
-\def\jyear#1{\gdef\@jyear{#1}}\def\@jyear{2005}
-\def\rtitle#1{\gdef\@rtitle{#1}}\def\@rtitle{Running Title}
-\def\rauthor#1{\gdef\@rauthor{#1}}\def\@rauthor{Running Author}
-%
-\def\issue #1#2#3{\jvol{#1}\jnum{#2}\jyear{#3}}
-\def\runningtitle#1{\rtitle{#1}}
-\def\runningauthor#1{\rauthor{#1}}
-%
-\jname{Computational Linguistics}
-\jinfo{Volume \@jvol, Number \@jnum}
-%
-\def\@sptitle{}
-\def\sptitle#1{\gdef\@sptitle{#1}}
-%
-\def\@title{}
-\def\title{\@dblarg{\@@title}}
-\def\@@title[#1]#2{\gdef\@title{#2}}
-%
-\newcount\aucount
-\newcount\tempcount
-%
-\def\author{\@dblarg{\@author}}
-\def\@author[#1]#2{\global\advance\aucount\@ne
-   \expandafter\gdef\csname author\romannumeral\aucount\endcsname{#2}}
-%
-\def\affil#1{\expandafter\gdef\csname affil\romannumeral\aucount\endcsname{#1}}
-%
-\def\printauthors{%
-   \ifnum\aucount=\z@
-      \gdef\@authors{}
-   \else
-      \gdef\@authors{%
-         \tempcount\@ne
-         \@whilenum\aucount>\z@
-         \do{%
-            \ifbrief
-               {\authorfont\csname author\romannumeral\tempcount\endcsname\vphantom{pl}\par}
-               {\reset@font\affilfont\csname affil\romannumeral\tempcount\endcsname\vphantom{pl}\par}%
-            \else\ifbookreview
-               {\authorfont\csname author\romannumeral\tempcount\endcsname\vphantom{pl}\par
-               {\reset@font\affilfont\csname affil\romannumeral\tempcount\endcsname\vphantom{pl}\par}}%
-            \else
-               \ifnum\tempcount=\@ne\noindent\else\ifodd\tempcount\vskip13.5pt\noindent\else\hskip2pc\fi\fi
-               \parbox[t]{15pc}{\authorfont
-               \csname author\romannumeral\tempcount\endcsname\vphantom{pl}\par
-               {\affilfont\setlength\baselineskip{13pt}\csname affil\romannumeral\tempcount\endcsname\vphantom{pl}\par}}%
-            \fi\fi
-            \advance\aucount\m@ne\advance\tempcount\@ne
-         }%
-      }
-   \fi
-}
-%
-\def\@historydates{}
-\def\historydates#1{\gdef\@historydates{#1}}%\thanks{#1}}
-%
-\def\@pubinfo{}
-\def\pubinfo#1{\gdef\@pubinfo{#1}}
-%
-\def\@reviewer{}
-\def\reviewer#1{\gdef\@reviewer{#1}}
-%
-\def\endbody{\ifx\@reviewer\@empty\else\unskip---{\itshape\@reviewer}\fi}%\par\addvspace{10pt}}
-%
-\def\@biography{}
-\def\biography#1{\gdef\@biography{#1}}
-%
-\ifpubrec
-  \def\@maketitle{\cleardoublepage%
-   \thispagestyle{pageonly}%titlepage
-   \parindent\z@
-   \ifx\@sptitle\@empty\else
-      \null\vskip -56.3pt
-      {\sptitlefont\@sptitle\vphantom{y}\par}%
-      \vskip 16.5pt
-   \fi
-}
-\else
-\ifbrief
-  \def\@maketitle{\cleardoublepage%
-   \thispagestyle{pageonly}%titlepage
-   \parindent\z@
-   \ifx\@sptitle\@empty\else
-      \null\vskip -56.3pt
-      {\sptitlefont\@sptitle\vphantom{y}\par}%
-      \vskip 17.5pt
-   \fi
-   \gdef\mktout@after@twocol{\parindent\z@%
-   \ifx\@title\@empty\else
-      {\titlefont\@title\vphantom{y}\par}
-   \fi
-   \printauthors
-   \ifx\@authors\@empty\else\vskip 6pt
-      {\authorfont\@authors\par}
-   \fi
-   \ifx\@pubinfo\@empty\else\vskip 6pt
-      {\pubinfofont\@pubinfo\par}
-   \fi
-%   \ifx\@reviewer\@empty\else\vskip 12pt
-%      {\reviewerfont Reviewed by\par\@reviewer\par}
-%   \fi
-   \par\addvspace{13pt}%
-   }\aftergroup\mktout@after@twocol
-}
-\else\ifbookreview
-  \def\@maketitle{\cleardoublepage%
-   \thispagestyle{empty}%titlepage
-   \parindent\z@
-   \ifx\@sptitle\@empty\else
-      \null\vskip -52.5pt
-      {\sptitlefont\@sptitle\vphantom{y}\par}%
-      \vskip 42.5pt
-   \fi
-   \ifx\@title\@empty\else
-      {\titlefont\@title\vphantom{y}\par}\vskip 12pt
-   \fi
-   \printauthors
-   \ifx\@authors\@empty\else
-      {\authorfont\@authors\par}
-   \fi
-   \ifx\@pubinfo\@empty\else\vskip 12pt
-      {\pubinfofont\@pubinfo\par}
-   \fi
-   \ifx\@reviewer\@empty\else\vskip 12pt
-      {\reviewerfont Reviewed by\par\@reviewer\par}
-   \fi
-   \par\addvspace{12pt}%
-   \gdef\@reviewer{}
-}
-\else
-  \def\@maketitle{\cleardoublepage%
-   \thispagestyle{titlepage}%
-   \parindent\z@
-   \ifx\@sptitle\@empty\else
-      \null\vskip -52.5pt
-      {\sptitlefont\@sptitle\vphantom{y}\par}%
-      \vskip 39.5pt
-   \fi
-   \ifx\@title\@empty\else
-      {\titlefont\@title\vphantom{y}\par}\vskip 29pt
-   \fi
-   \printauthors
-   \ifx\@authors\@empty\else
-      {\authorfont\@authors\par}
-   \fi
-   \par\addvspace{31pt}%
-}
-\fi\fi\fi
-%
-\def\maketitle{\par
- \begingroup
-   \def\thefootnote{\fnsymbol{footnote}}
-   \if@twocolumn 
-      \twocolumn[\@maketitle] 
-   \else
-      \@maketitle 
-   \fi
-   \@thanks
-   \ifx\@historydates\@empty\else\let\domkfnmark\relax\def\dofnformat{\vskip\baselineskip}\@footnotetext{\@historydates}\fi
- \endgroup
- \setcounter{footnote}{0}
- \let\maketitle\relax
- \let\@maketitle\relax
- \gdef\@thanks{}\gdef\@authors{}\gdef\@title{}\gdef\@pubinfo{}\let\thanks\relax\@afterindentfalse\@afterheading}
-%
-\newenvironment{abstract}{\par\abstractfont\noindent\ignorespaces}{\par\addvspace{12pt}\@afterindentfalse\@afterheading}
-\newcommand{\keywords}[1]{{\keywordfont Keywords: #1\par\addvspace{12pt}}}
-%
-\def\bookinfospace{\vskip10pt plus1pt}
-\long\def\bookinfo#1#2{{\raggedright{\bfseries #1}\\#2\par\bookinfospace}}
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Sectioning commands %%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-\setcounter{secnumdepth}{3}
-\newcounter {part}
-\newcounter {chapter}
-\newcounter {section}[chapter]
-\newcounter {subsection}[section]
-\newcounter {subsubsection}[subsection]
-\newcounter {paragraph}[subsubsection]
-\newcounter {subparagraph}[paragraph]
-\renewcommand\thepart          {\Roman{part}}
-\renewcommand\thechapter       {\arabic{chapter}}
-\renewcommand\thesection       {\arabic{section}}
-\renewcommand\thesubsection    {\thesection.\arabic{subsection}}
-\renewcommand\thesubsubsection {\thesubsection .\arabic{subsubsection}}
-\renewcommand\theparagraph     {\thesubsubsection.\arabic{paragraph}}
-\renewcommand\thesubparagraph  {\theparagraph.\arabic{subparagraph}}
-%
-\def\@sectioncntformat#1{{\csname #1numfont\endcsname\csname the#1\endcsname.}\nobreakspace}
-\def\@seccntformat#1{\csname #1numfont\endcsname\csname the#1\endcsname\nobreakspace}
-%
-\newcommand\section{\@startsection {section}{1}{\z@}{-12pt \@plus -2pt}{12pt}{\def\@afterhead{}\sectionfont}}%
-\newcommand\subsection{\@startsection{subsection}{2}{\z@}{-12pt \@plus -2pt}{12pt}{\def\@afterhead{}\subsectionfont}}%
-\newcommand\subsubsection{\@startsection{subsubsection}{3}{\z@}{-12pt \@plus -2pt}{0pt}{\def\@afterhead{. }\subsubsectionfont}}%
-\newcommand\paragraph{\def\@afterhead{.}\@startsection{paragraph}{4}{\z@}{-12pt \@plus -2pt}{-0pt}{\def\@afterhead{. }\paragraphfont}}%
-\newcommand\subparagraph{\@startsection{subparagraph}{5}{\parindent}{-6pt \@plus -2pt}{-1em}{\def\@afterhead{. }\subparagraphfont}}%
-\newcommand\subsubparagraph{\@startsection{subsubparagraph}{6}{\parindent}{-6pt \@plus -2pt}{-1em}{\def\@afterhead{. }\subsubparagraphfont}}%
-\newcommand\xhead{\@startsection{xhead}{7}{\z@}{-14pt \@plus -2pt}{2pt}%{0.0001pt}
-                                        {\def\@afterhead{}\xheadfont}}%
-\let\xheadmark\@gobble
-%
-\def\@startsection#1#2#3#4#5#6{%
-  \if@noskipsec \leavevmode \fi
-  \par
-  \@tempskipa #4\relax
-  \@afterindenttrue
-  \ifdim \@tempskipa <\z@
-    \@tempskipa -\@tempskipa \@afterindentfalse
-  \fi
-  \if@nobreak
-%    \ifnum#2=2\fi
-    \ifnum#2=3\vskip-12pt\fi
-    \everypar{}%
-  \else
-    \addpenalty\@secpenalty\addvspace\@tempskipa
-  \fi
-  \@ifstar
-    {\@ssect{#3}{#4}{#5}{#6}}%
-    {\@dblarg{\@sect{#1}{#2}{#3}{#4}{#5}{#6}}}}
-%
-\def\@afterhead{}
-\def\@sect#1#2#3#4#5#6[#7]#8{\ifnum #2>\c@secnumdepth
-     \let\@svsec\@empty\else
-     \refstepcounter{#1}%
-     \let\@@protect\protect
-     \def\protect{\noexpand\protect\noexpand}%
-     \ifnum#2=1
-        \edef\@svsec{\@sectioncntformat{#1}}%
-     \else
-        \edef\@svsec{\@seccntformat{#1}}%
-     \fi
-     \let\protect\@@protect\fi
-     \@tempskipa #5\relax
-      \ifdim \@tempskipa>\z@
-        \begingroup #6\relax
-          \@hangfrom{\hskip #3\relax\@svsec}%
-                    {\interlinepenalty \@M #8\@afterhead\par}%
-        \endgroup
-       \csname #1mark\endcsname{#7}\addcontentsline
-         {toc}{#1}{\ifnum #2>\c@secnumdepth \else
-                      \protect\numberline{\csname the#1\endcsname}\fi
-                    #7}\else
-        \def\@svsechd{#6\hskip #3\relax
-                   \@svsec #8\@afterhead\csname #1mark\endcsname
-                      {#7}\addcontentsline
-                           {toc}{#1}{\ifnum #2>\c@secnumdepth \else
-                           \protect\numberline{\csname the#1\endcsname}%
-                                     \fi
-                       #7}}\fi
-     \@xsect{#5}}
-%
-\def\@ssect#1#2#3#4#5{\@tempskipa #3\relax
-   \ifdim \@tempskipa>\z@
-     \begingroup #4\@hangfrom{\hskip #1}{\interlinepenalty \@M #5\par}\endgroup
-   \else \def\@svsechd{#4\hskip #1\relax #5\@afterhead\null}\fi
-    \@xsect{#3}}
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%  Lists Variable Initialisation %%%%%%%%%%%%%%%%%%%%%%%
-%
-\newskip\topsepi \topsepi6\p@ \@plus2\p@% \@minus.5\p@
-\newskip\topsepii \topsepii2pt% \@plus1\p@ 
-\newskip\topsepiii \topsepiii2pt% \@plus1\p@ 
-\newskip\itemsepi \itemsepi0pt
-\newskip\itemsepii \itemsepii0pt
-\newskip\itemsepiii \itemsepiii0pt
-\newdimen\LabelSep \LabelSep4.7pt
-%
-\def\@listI{\leftmargin\leftmargini
-            \labelwidth\leftmargini
-            \advance\labelwidth-\labelsep
-            \parsep 0\p@%
-            \topsep \topsepi
-            \itemsep\itemsepi}%
-\let\@listi\@listI
-\@listi
-\def\@listii {\leftmargin\leftmarginii
-              \labelwidth\leftmarginii
-              \advance\labelwidth-\labelsep
-              \topsep\topsepii
-              \parsep 0pt
-              \itemsep\itemsepii}
-\def\@listiii {\leftmargin\leftmarginiii
-              \labelwidth\leftmarginiii
-              \advance\labelwidth-\labelsep
-              \topsep\topsepiii
-              \parsep 0pt
-              \itemsep\itemsepiii}
-\def\@listiv {\leftmargin\leftmarginiv
-              \labelwidth\leftmarginiv
-              \advance\labelwidth-\labelsep}
-\def\@listv  {\leftmargin\leftmarginv
-              \labelwidth\leftmarginv
-              \advance\labelwidth-\labelsep}
-\def\@listvi {\leftmargin\leftmarginvi
-              \labelwidth\leftmarginvi
-              \advance\labelwidth-\labelsep}
-%
-\setlength\leftmargini  {2.5em}
-\setlength\leftmarginii  {2.2em}
-\setlength\leftmarginiii {1.87em}
-\setlength\leftmarginiv  {1.7em}
-\setlength\leftmarginv  {1em}
-\setlength\leftmarginvi {1em}
-\setlength\leftmargin    {\leftmargini}
-%
-\setlength  \labelsep  {\LabelSep}
-\setlength  \labelwidth{\leftmargini}
-\addtolength\labelwidth{-\labelsep}
-%
-\renewcommand\theenumi{\arabic{enumi}}
-\renewcommand\theenumii{\alph{enumii}}
-\renewcommand\theenumiii{\roman{enumiii}}
-\renewcommand\theenumiv{\Alph{enumiv}}
-\newcommand\labelenumi{\theenumi.}
-\newcommand\labelenumii{(\theenumii)}
-\newcommand\labelenumiii{\theenumiii.}
-\newcommand\labelenumiv{\theenumiv.}
-\renewcommand\p@enumii{\theenumi}
-\renewcommand\p@enumiii{\theenumi(\theenumii)}
-\renewcommand\p@enumiv{\p@enumiii\theenumiii}
-\font\lcir = lcircle10 at 12pt
-\newcommand\bulls{\hbox{\lcir\char'162}}
-\def\textbullet{\leavevmode\raise3.5pt\bulls\hskip-2pt}
-\def\textendash{{\bf--}}
-\def\textasteriskcentered{\leavevmode\raise-1.5pt\hbox{*}}
-\def\textperiodcentered{\leavevmode\raise1.5pt\hbox{\bulls}}
-\newcommand\labelitemi{\textbullet}
-\newcommand\labelitemii{\normalfont\bfseries \textendash}
-%
-\newenvironment{description}
-               {\list{}{\labelwidth\z@ \itemindent-\leftmargin
-                        \let\makelabel\descriptionlabel}}
-               {\endlist}
-\newcommand*\descriptionlabel[1]{\hspace\labelsep
-                                \normalfont\bfseries #1}
-%
-\newenvironment{verse}
-               {\let\\\@centercr
-                \list{}{\itemsep      \z@
-                        \itemindent   -1.5em%
-                        \listparindent\itemindent
-                        \rightmargin  \leftmargin
-                        \advance\leftmargin 1.5em}%
-                \item\relax}
-               {\endlist}
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%  Enumerate list %%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-\newskip\listtopsepi
-\newskip\listtopsepii
-\newskip\listtopsepiii
-\newskip\listitemsepi
-\newskip\listitemsepii
-\newskip\listitemsepiii
-\newlength\listleftmargini
-\newlength\listleftmarginii
-\newlength\listleftmarginiii
-\newlength\listlabelwidthi
-\newlength\listlabelwidthii
-\newlength\listlabelwidthiii
-\newlength\listlabelsepi
-\newlength\listlabelsepii
-\newlength\listlabelsepiii
-\newlength\listrightmargin
-%
-\newcount\listdepth
-%
-\newif\if@nomainitem
-\def\nomainitem{\global\@nomainitemtrue}
-\newlength\lpalignmaxwd
-\newif\if@lpalign
-\def\lpalign#1{\global\@lpaligntrue\global\settowidth\lpalignmaxwd{#1}}
-%
-\ifbrief
-   \listtopsepi 5pt plus2pt% minus1pt
-   \listtopsepii 0pt
-   \listtopsepiii 0pt
-   \listleftmargini 25pt
-   \listleftmarginii 25pt
-   \listleftmarginiii 25pt
-   \listlabelwidthi 12pt
-   \listlabelwidthii 12pt
-   \listlabelwidthiii 12pt
-   \listlabelsepi 13pt
-   \listlabelsepii 13pt
-   \listlabelsepiii 13pt
-   \listrightmargin 0pt
-   \listitemsepi 3pt
-   \listitemsepii 0pt
-   \listitemsepiii 0pt
-\else
-   \listtopsepi 12pt plus2pt% minus1pt
-   \listtopsepii 0pt
-   \listtopsepiii 0pt
-   \listleftmargini 36pt
-   \listleftmarginii 14pt
-   \listleftmarginiii 8pt
-   \listlabelwidthi 18pt
-   \listlabelwidthii 32pt
-   \listlabelwidthiii 40pt
-   \listlabelsepi 18pt
-   \listlabelsepii 18pt
-   \listlabelsepiii 18pt
-   \listrightmargin 18pt
-   \listitemsepi 6pt
-   \listitemsepii 0pt
-   \listitemsepiii 0pt
-\fi
-%
-\def\enumerate{\@ifnextchar[{\@enumerate}{\@enumerate[1]}}% 
-%
-\def\@enumerate[#1]{\par
-   \ifnum\@enumdepth >\thr@@\@toodeep\else
-      \advance\@enumdepth\@ne\advance\listdepth\@ne
-      \edef\@listcounter{enum\romannumeral\the\@enumdepth}%
-      \setcounter{\@listcounter}{1}%
-   \fi
-   \list{\csname labelenum\romannumeral\the\@enumdepth\endcsname}{\listfont%
-      \usecounter{\@listcounter}%
-      \topsep\csname listtopsep\romannumeral\the\listdepth\endcsname
-      \leftmargin\csname listleftmargin\romannumeral\the\listdepth\endcsname
-      \labelwidth\csname listlabelwidth\romannumeral\the\listdepth\endcsname
-      \labelsep\csname listlabelsep\romannumeral\the\listdepth\endcsname
-      \itemsep\csname listitemsep\romannumeral\the\listdepth\endcsname
-      \rightmargin\listrightmargin\advance\leftmargin\leftskip
-      \if@lpalign
-         \global\@lpalignfalse
-         \def\makelabel##1{\hbox to \labelwidth{\hss\hbox to\lpalignmaxwd{##1\hss}\hskip-.6pt}}
-      \else
-         \def\makelabel##1{\hbox to \labelwidth{\hss##1\hskip-.6pt}}
-      \fi
-      \if@nomainitem
-         \global\@nomainitemfalse
-         \leftmargin 10pt
-         \labelwidth 10pt
-         \labelsep 0pt
-         \def\makelabel##1{##1 }%
-      \fi
-      }%
-  }%
-%
-\def\endenumerate{\endlist}%
-%
-\newenvironment{arabiclist}{%
-  \def\theenumi{\arabic{enumi}}\def\labelenumi{\theenumi.}
-  \def\theenumii{\arabic{enumii}}\def\labelenumii{\theenumii.}%
-  \def\theenumiii{\arabic{enumiii}}\def\labelenumiii{\theenumiii.}%
-  \begin{enumerate}%
-}{%
-  \end{enumerate}}
-%
-\newenvironment{romanlist}{%
-  \def\theenumi{\roman{enumi}}\def\labelenumi{\theenumi.}%
-  \def\theenumii{\roman{enumii}}\def\labelenumii{\theenumii.}%
-  \def\theenumiii{\roman{enumiii}}\def\labelenumiii{\theenumiii.}%
-  \begin{enumerate}%
-}{%
-  \end{enumerate}}
-%
-\newenvironment{alphalist}{%
-  \def\theenumi{\alph{enumi}}\def\labelenumi{(\theenumi)}%
-  \def\theenumii{\alph{enumii}}\def\labelenumii{(\theenumii)}%
-  \def\theenumiii{\alph{enumiii}}\def\labelenumiii{(\theenumiii)}%
-  \begin{enumerate}%
-}{%
-  \end{enumerate}}
-%
-\newenvironment{Romanlist}{%
-  \def\theenumi{\Roman{enumi}}\def\labelenumi{\theenumi.}%
-  \def\theenumii{\Roman{enumii}}\def\labelenumii{\theenumii.}%
-  \def\theenumiii{\Roman{enumiii}}\def\labelenumiii{\theenumiii.}%
-  \begin{enumerate}%
-}{%
-  \end{enumerate}}
-%
-\newenvironment{Alphalist}{%
-  \def\theenumi{\Alph{enumi}}\def\labelenumi{(\theenumi)}%
-  \def\theenumii{\Alph{enumii}}\def\labelenumii{(\theenumii)}%
-  \def\theenumiii{\Alph{enumiii}}\def\labelenumiii{(\theenumiii)}%
-  \begin{enumerate}%
-}{%
-  \end{enumerate}}
-%
-\newenvironment{examples}{\begin{exlist}\item}{\end{exlist}}
-%
-
-\newcounter{eqnnosave}          % used in trick with equation number
-\newenvironment{exlist}{%         % define "example" environment
-   \listleftmargini 36pt
-   \listlabelwidthi 30pt
-   \listlabelsepi 6pt
-   \listitemsepi12pt
-   \def\labelenumi{(\theenumi)}
-   \def\theenumii{\arabic{enumii}}\def\labelenumii{\theenumii.}%
-   \begin{enumerate}%
-   \setcounter{enumi}{\arabic{eqnnosave}}%   % restores previous value
-}%
-{\end{enumerate}%
-\setcounter{eqnnosave}{\arabic{enumi}}%
-}
-%
-\newenvironment{exoutlist}{\par%
-   \listleftmarginii 20pt
-   \listlabelwidthii 32pt
-   \listlabelsepii 24pt
-   \def\labelenumi{(\theenumi)}
-   \def\theenumii{\arabic{enumii}}\def\labelenumii{\theenumii.}%
-   \begin{enumerate}%
-   \setcounter{enumi}{\arabic{eqnnosave}}%   % restores previous value
-}%
-{\end{enumerate}%
-\setcounter{eqnnosave}{\arabic{enumi}}%
-}
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%% itemize (bullet) %%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-\def\itemize{\par
-   \ifnum\@itemdepth >\thr@@\@toodeep\else
-      \advance\@itemdepth\@ne\advance\listdepth\@ne
-   \fi
-   \list{\csname labelitem\romannumeral\the\@itemdepth\endcsname}{\listfont%
-      \topsep\csname listtopsep\romannumeral\the\listdepth\endcsname
-      \labelwidth\csname listlabelwidth\romannumeral\the\listdepth\endcsname
-      \labelsep\csname listlabelsep\romannumeral\the\listdepth\endcsname
-      \leftmargin\csname listleftmargin\romannumeral\the\listdepth\endcsname
-      \itemsep\csname listitemsep\romannumeral\the\listdepth\endcsname
-      \rightmargin\listrightmargin\advance\leftmargin\leftskip
-      \def\makelabel##1{\hbox to \labelwidth{\hss##1}}}%
-   }%
-%
-\def\enditemize{\endlist}
-%%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%% enumroman (i) %%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-\newenvironment{bulletlist}{%
-   \renewcommand\labelitemi{\textbullet}\renewcommand\labelitemii{\textbullet}%
-   \begin{itemize}
-}{%
-   \end{itemize}}
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%  unnumlist %%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-\newenvironment{unnumlist}{\par%
-   \list{}{\listfont%
-      \topsep12pt plus2pt% minus1pt
-      \rightmargin18pt
-      \leftmargin36pt%\itemindent-18pt
-      \itemsep6pt\parsep0pt
-      \partopsep0pt}
-    \def\makelable##1{##1}%
-   }{\endlist}%
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%  Quotes %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-\def\@source{}
-\def\source#1{\gdef\@source{#1}}
-%
-\newenvironment{extract}{\par\ifbrief\addvspace{4pt plus2pt}\else\ifbookreview\addvspace{12pt plus2pt}\else\addvspace{14pt plus2pt}\fi\fi
-      \extractfont\parindent18pt\noindent\ignorespaces
-}{\par\ifx\@source\@empty\else{\sourcefont\noindent---\@source\par}\fi\gdef\@source{}\ifbrief\addvspace{3pt plus2pt}\else\addvspace{12pt plus2pt}\fi\@endparenv}
-%
-%%%%%%%%%%%%%%%%%%%%%% endpara and numberedpara %%%%%%%%%%%%%%%%%%%%%%%
-%
-\newenvironment{lastpara}{\par\addvspace{17pt plus2pt}%
-      \noindent\ignorespaces}{\par}
-%
-\newenvironment{numpara}{\par
-      \list{\arabic{enumi}}{\usecounter{enumi}\topsep\z@\itemsep\z@\leftmargin9pt\itemindent-9pt\labelwidth\z@\labelsep\z@\labelwidth\z@\listparindent12pt\def\makelabel##1{##1 }}}{\endlist}
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Theorems %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-\usepackage{amsthm}
-\newtheoremstyle{break}%
- {}{}%
- {}{}%
- {\bfseries}{}% % Note that final punctuation is omitted.
- {\newline}{}
-
-\theoremstyle{break}
-
-\def\examplename{Example}
-\newtheorem{example}{\examplename}
-%
-\def\theoremname{Theorem}
-\newtheorem{theorem}{\theoremname}%[section]
-%
-\def\lemmaname{Lemma}
-\newtheorem{lemma}{\lemmaname}%[section]
-%
-%\def\rulesname{Rule}
-%\newtheorem{rules}{\rulesname}%[section]
-%
-\def\propositionname{Proposition}
-\newtheorem{proposition}{\propositionname}%[section]
-%
-\def\corollaryname{Corollary}
-\newtheorem{corollary}{\corollaryname}%[section]
-%
-\def\notationname{Notation}
-\newtheorem{notation}{\notationname}%[section]
-%
-\def\assumptionname{Assumption}
-\newtheorem{assumption}{\assumptionname}%[section]
-%
-\def\remarkname{Remark}
-\newtheorem{remark}{\remarkname}%[section]
-%
-\newif\ifdefinition
-\def\numdefname{Definition}
-\newtheorem{numdef}{\numdefname}%[section]
-%
-\newtheorem{numtheorem}{Theorem}
-%
-\def\casename{Case}
-\newtheorem{case}{\casename}%[section]
-%
-%
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Proof %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-% proof* and proof are the same
-% proof* is provided for compatibility with older cl style files
-\newenvironment{proof*}{\begin{proof}}{\end{proof}}
-%
-\newenvironment{solution}{\begin{proof}[Solution.]}{\end{proof}}
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%  Floats %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-\setcounter{topnumber}{5}
-\renewcommand\topfraction{.9}
-\setcounter{bottomnumber}{5}
-\renewcommand\bottomfraction{.9}
-\setcounter{totalnumber}{10}
-\renewcommand\textfraction{.09}
-\renewcommand\floatpagefraction{.901}
-\setcounter{dbltopnumber}{1}
-\renewcommand\dbltopfraction{.9}
-\renewcommand\dblfloatpagefraction{.901}
-%
-\newlength\abovecaptionskip
-\newlength\belowcaptionskip
-\setlength\abovecaptionskip{4.5\p@}
-\setlength\belowcaptionskip{2.5\p@}
-%
-\def\FigName{figure}
-%
-\long\def\@makecaption#1#2{%
-    \ifx\FigName\@captype
-      \vskip\abovecaptionskip
-      \@makefigurecaption{#1}{#2}%
-    \else
-      \@maketablecaption{#1}{#2}%
-      \vskip\belowcaptionskip
-    \fi
-}
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%  Figures %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-\newcounter{figure}[chapter]
-\renewcommand\thefigure{\@arabic\c@figure}
-\def\fps@figure{tbp}
-\def\ftype@figure{1}
-\def\ext@figure{lof}
-\def\fnum@figure{\figurename\nobreakspace\thefigure}
-%
-\newenvironment{figure}
-               {\@float{figure}}
-               {\end@float}
-%
-\newenvironment{figure*}
-               {\@dblfloat{figure}}
-               {\end@dblfloat}
-%
-\def\figlabelsep{.5em}
-%
-\def\@makefigurecaption#1#2{%
-    {\figcaptionnumfont#1\par}
-    {\figcaptionfont#2\vphantom{y}\par}\vskip-2.6pt}
-%
-\def\ArtDir{art/}%
-%
-\usepackage{epsfig}
-\usepackage[figuresright]{rotating}
-%
-\newbox\figtempbox
-\def\ArtPiece#1{\epsfbox{\ArtDir#1}}%
-\let\figboxformat\leftline
-%
-\def\figurebox#1#2#3{%
-    \@ifnextchar[{\@figurebox{#1}{#2}{#3}}{\@figurebox{#1}{#2}{#3}[]}}
-%
-\def\@figurebox#1#2#3[#4]{%
-      \gdef\@figscale{#3}
-      \gdef\@frtharg{#4}
-      \ifx\@frtharg\empty
-         \global\figheight=#1
-         \global\figwidth=#2
-      \else 
-         \setbox\figtempbox=\hbox{\ifx\@figscale\empty\else\epsfxsize\@figscale\fi\epsfbox{\ArtDir#4}}%
-         \global\figwidth=\wd\figtempbox
-         \global\figheight=\ht\figtempbox
-      \fi
-      {\figboxformat{\figbox}}%%
-}%
-%
-\def\figbox{%
-     \ifx\@frtharg\empty  
-       \noindent\vbox{\hsize\figwidth%
-              \hrule\hbox to\figwidth{\vrule\hfill\vbox to\figheight{\hsize\figwidth\vfill}\vrule}\hrule}%
-     \else
-        \vbox{\vskip.8pt\hsize\figwidth
-              \hbox to\figwidth{\vbox to\figheight{\hsize\figwidth\box\figtempbox}}}%
-     \fi
-}
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%  Tables %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-\newcounter{table}[chapter]
-\renewcommand\thetable{\@arabic\c@table}
-\def\fps@table{tbp}
-\def\ftype@table{2}
-\def\ext@table{lot}
-\def\fnum@table{\tablename\nobreakspace\thetable}
-%
-\newenvironment{table}
-               {\@float{table}}
-               {\end@float}
-%
-\newenvironment{table*}
-               {\@dblfloat{table}}
-               {\end@dblfloat}
-%
-\def\@maketablecaption#1#2{%
-    \hrule height1pt\par\vskip12pt
-    {\tablecaptionnumfont#1\par}%
-    {\tablecaptionfont#2\vphantom{y}\par}}
-%
-\def\tbl#1#2{\tablefont%
-  \setbox\tempbox\hbox{\tablefont#2}%
-  \tabledim\hsize
-  \advance\tabledim-\wd\tempbox
-  \ifdim\tabledim>0pt
-        \divide\tabledim2
-  \else
-        \global\tabledim0pt
-  \fi
-  \global\tableleftskip\tabledim
-  \global\tablerightskip\tabledim
-  \caption{#1}%
-  {\box\tempbox}}%
-%
-\def\TCH#1{\TCHfont#1}%
-%
-\def\x{@{\extracolsep{\fill}}}
-\def\toprule{\Hline\\[-5.5pt]}
-\def\colrule{\\[-7.5pt]\Hline\\[-5pt]}
-\def\botrule{}
-\def\crule#1{\\[-7.5pt]\CLINE{#1}\\[-5pt]}
-%
-\def\Hline{%
-  \noalign{\ifnum0=`}\fi\hrule \@height .5pt \futurelet%\@height \arrayrulewidth
-   \@tempa\@xhline}
-%
-\newenvironment{tabnote}{\par\tabnotefont
-   }{\par}
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%  End Floats %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%  Math %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-\def\arraystretch{1}
-\setlength\arraycolsep{1.5\p@}
-\setlength\tabcolsep{6\p@}
-\setlength\arrayrulewidth{.4\p@}
-\setlength\doublerulesep{2\p@}
-\setlength\tabbingsep{\labelsep}
-\setlength\fboxsep{3\p@}
-\setlength\fboxrule{.4\p@}
-%
-\setlength\columnsep{24\p@}
-\setlength\columnseprule{0\p@}
-%
-\@addtoreset{equation}{chapter}
-\renewcommand\theequation{\arabic{equation}}
-\def\@eqnnum{{\reset@font\rmfamily\quad (\theequation)}}
-%
-\def\bstrut{\vrule width0pt depth6pt}
-\def\tstrut{\vrule width0pt height9pt}
-\jot=6pt
-%%
-%
-\def\text#1{\mathchoice
-        {\hbox{\fontsize{\tf@size}{\tf@size}\selectfont#1}}%
-        {\hbox{\fontsize{\tf@size}{\tf@size}\selectfont#1}}%
-        {\hbox{\fontsize{\sf@size}{\sf@size}\selectfont#1}}%
-        {\hbox{\fontsize{\ssf@size}{\ssf@size}\selectfont#1}}}
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%  End Math %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%  Footnote %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-\def\thanks#1{\footnotemark
-    \protected@xdef\@thanks{\@thanks\protect\footnotetext[\the\c@footnote]{#1}}}
-%
-\def\@makefnmark{\smash{\hbox{\@textsuperscript{\normalfont\@thefnmark}}}}
-%
-\renewcommand\footnoterule{%
-  \kern-4\p@
-  \hrule width 15pc height.5pt depth\z@
-  \kern 3.5\p@}
-%
-\@addtoreset{footnote}{chapter}
-\renewcommand\thefootnote{\arabic{footnote}}
-%
-\def\@fnsymbol#1{\ifcase#1\or \ensuremath{*}\or \ensuremath{**}\or\ensuremath{\dagger}\or\ensuremath{\ddagger}\or
-   \S\or\|\or\#\or**\or\ensuremath{\dagger\dagger}\or\ensuremath{\ddagger\ddagger} 
-   \or\S\S\or\|\hskip-1pt\|\or\#\#\or ***\or\ensuremath{\dagger\dagger\dagger}\or\ensuremath{\ddagger\ddagger\ddagger}\else\@ctrerr\fi\relax}
-%
-\newdimen\@footmax
-\def\footmax#1{%
-  \setbox\tempbox\hbox{\footnotesize#1}
-  \global\@footmax\wd\tempbox}
-%
-\footmax{00}
-%
-\def\domkfnmark{\noindent\hskip-12pt\hbox to 12pt{\hbox to \@footmax{\hss$\@thefnmark$}\hss}}
-\def\dofnformat{\parindent8pt\leftskip12pt\rightskip0pt plus1fill}
-%
-\long\def\@makefntext#1{\dofnformat%
-    \domkfnmark#1}
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%  End Footnote %%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%  Page styles  %%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-\def\cyanColor#1{\special{color push cmyk 1 0 0 0}#1\special{color pop}}
-\def\whiteColor#1{\special{color push cmyk 0 0 0 0}#1\special{color pop}}
-%
-\def\folio{{\foliofont\thepage}}
-%
-\def\bookrev#1{\gdef\@bookrev{#1}}
-\bookrev{Book Reviews}
-%
-\def\briefhead#1{\gdef\@briefhead{#1}}
-\briefhead{Briefly Noted}
-%
-\def\ps@headings{%
-    \def\@oddfoot{\hfill{\folio}}
-    \def\@evenfoot{{\folio}\hfill}
-    \def\@evenhead{{\rhfont\@jname\hfill\@jinfo}}%
-    \def\@oddhead{\ifbookreview{\rhfont\hfill\@bookrev}\else\ifshortpaper{\rhfont\hfill\@sptitle}\else{\rhfont\@rauthor\hfill\@rtitle}\fi\fi}%
-    \let\@mkboth\markboth
-}
-%
-\def\ps@empty{%
-      \def\@oddfoot{\hfill\lower\blindfoliodrop\hbox{\thepage}\hfill}
-      \let\@evenfoot\@oddfoot
-      \def\@evenhead{}%
-      \def\@oddhead{}%
-      \let\@mkboth\markboth
-}
-%
-\newcommand{\footmark}{\copyright\ \@jyear\ Association for Computational Linguistics}
-%
-\def\ps@titlepage{\let\@mkboth\@gobbletwo%
-  \def\@oddfoot{\footnotesize\footmark\hfill}
-  \def\@evenfoot{\footnotesize\hfill\footmark}
-  \def\@oddhead{}\let\@evenhead\@oddhead}
-%
-\def\ps@pageonly{\let\@mkboth\@gobbletwo%
-    \def\@oddfoot{\hfill{\folio}}
-    \def\@evenfoot{{\folio}\hfill}
-  \def\@oddhead{}\let\@evenhead\@oddhead}
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%  End Page styles  %%%%%%%%%%%%%%%%%%%%%%%%
-%
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%  Bibliography Section   %%%%%%%%%%%%%%%%%%
-%
-% Bibliography
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-%%% Citation forms:
-%%%
-%%% Macro       Output format
-%%% ----------- -----------------------------------------
-%%% \cite:      (Dewey, 1988)
-%%%             (Dewey, 1988, page 15)
-%%%             (Dewey, 1988; Cheatham, 1987; Howe, 1903)
-%%% \namecite:  Dewey (1988)
-%%%             Dewey (1988, page 15)
-%
-
-\newlength\bibleftmargin
-\newlength\biblabelsep
-\newlength\bibitemsep
-\newlength\bibparsep
-%
-\setlength\bibleftmargin{9pt}
-\setlength\biblabelsep  {3pt}
-\setlength\bibitemsep   {0pt}
-\setlength\bibparsep    {0pt}
-%
-\def\bibhead#1{\bibitem{}\null\par\bgroup\nobreak{\leftskip-\leftmargin\hskip\leftmargin\bfseries #1\par}\nobreak\egroup}
-%
-\newenvironment{thebibliography}[1]
-   {\par%
-   \ifx\withintwocol{true}\else%\starttwocolumn
-   \fi
-    \def\@tempa{#1}%
-    \ifx\@tempa\@empty
-       \list{}{%
-          \labelwidth0pt\labelsep0pt
-          \leftmargin\bibleftmargin
-          \itemindent-\bibleftmargin
-          \itemsep\bibitemsep
-          \parsep\bibparsep
-          \usecounter{enumiv}%
-          \let\p@enumiv\@empty}%
-    \else
-       \setbox\tempbox\hbox{\@tempa.}
-       \tempdimen\wd\tempbox
-       \def\@biblabel##1{\hbox to \tempdimen{\hfill##1.}}
-       \list{\@biblabel{\arabic{enumiv}}}%
-          {\settowidth\labelwidth{\@biblabel{#1}}%
-           \labelsep\biblabelsep\leftmargin\labelsep
-           \advance\leftmargin\labelwidth
-           \itemindent0pt
-           \itemsep\bibitemsep
-           \parsep\bibparsep
-           \usecounter{enumiv}%
-           \let\p@enumiv\@empty
-           \renewcommand\theenumiv{\arabic{enumiv}}}%
-    \fi
-    \sloppy\clubpenalty4000\widowpenalty4000\sfcode`\.=\@m
-   }{%
-    \def\@noitemerr{\@latex@warning{Empty `thebibliography' environment}}%
-    \endlist
-   }
-
-\def\@lbibitem[#1]#2{\item[]\if@filesw 
-      { \def\protect##1{\string ##1\space}\immediate
-        \write\@auxout{\string\bibcite{#2}{#1}}\fi\ignorespaces}}
-
-\def\@bibitem#1{\item\if@filesw \immediate\write\@auxout
-       {\string\bibcite{#1}{\the\c@enumi}}\fi\ignorespaces}
-
-
-\usepackage{natbib}
- \setcitestyle{aysep={}}	% no comma between author and year
- \renewcommand{\cite}{\citep}	% to get "(Author Year)" with natbib
- \newcommand{\namecite}{\citet}	% to get "Author (Year)" with natbib
- \def\bibfont{\small\raggedright}
- \def\bibsection{\xhead*{References}}
-
-% 
-\newcommand\newblock{}
-%
-% endnotes; same environment as bibliography:
-%
-\def\theendnotes{\small\parindent=0pt\par\vspace{14pt}{\bf Notes}\par\vspace{2pt}\everypar{\hangindent=1em\hangafter=1}\raggedright}
-\def\endtheendnotes{\par\vskip14pt plus4pt }
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%  End Bibliography Section %%%%%%%%%%%%%%%%%%%%%%%
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%  Contents List  %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-\def\dotfill{%
-  \leavevmode
-  \cleaders \hb@xt@ .44em{\hss.\hss}\hfill
-  \kern\z@}
-%
-\newcommand\@pnumwidth{1.55em}
-\newcommand\@tocrmarg {2.55em}
-\newcommand\@dotsep{4.5}
-\setcounter{tocdepth}{2}
-%
-\newcommand\l@section      {\@dottedtocline{1}{1.5em}{2.3em}}
-\newcommand\l@subsection   {\@dottedtocline{2}{3.8em}{3.2em}}
-\newcommand\l@subsubsection{\@dottedtocline{3}{7.0em}{4.1em}}
-\newcommand\l@paragraph    {\@dottedtocline{4}{10em}{5em}}
-\newcommand\l@subparagraph {\@dottedtocline{5}{12em}{6em}}
-\let\l@table\l@figure
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%  Appendix  %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-% command to add a section to the appendix
-\newcommand{\appendixsection}[1]{\addtocounter{section}{1}%
-   \setcounter{table}{0}
-   \setcounter{figure}{0}
-   \setcounter{equation}{0}
-  \section*{Appendix \Alph{section}: #1}%
-}
-%
-\newcommand\appendix{%
-   \setcounter{section}{0}
-   \renewcommand{\theequation}{\Alph{section}.\arabic{equation}}
-}
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% End Appendix %%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%% Document & End Document %%%%%%%%%%%%%%%%%%%%%
-%
-%\def\@watermark{\offinterlineskip\vbox to 0pt{\setlength\overfullrule{0pt}\vskip12pc\hskip2pc\begin{turn}{45}\hbox to \textheight{\hss\grayColor{{\sffamily\fontsize{107.5}{107.5pt}\selectfont\bfseries spi}\hskip16pt\raise10pt\hbox{\sffamily\fontsize{50}{50}\selectfont\bfseries publisher services}}\hss}\end{turn}}\vskip0pt}
-%
-\def\@questionmark{\vbox to 0pt{\vskip13pc\hskip-5pc
-                  \hbox to \textwidth{\fontsize{570}{570}\selectfont ?\hss}}}
-\def\questionmark#1{\xdef\@questionmark{#1}}%
-%
-%\def\draftnote{\vbox to 0pt{\offinterlineskip%
-      %\hbox to \trimwidth{\hfill\footnotesize\jobname\vphantom{q}\qquad\today\qquad\currenttime\hfill}\par\@questionmark}}%
-%
-\def\document{\endgroup
-  \ifx\@unusedoptionlist\@empty\else
-    \@latex@warning@no@line{Unused global option(s):^^J%
-            \@spaces[\@unusedoptionlist]}%
-  \fi
-  \@colht\textheight
-  \@colroom\textheight \vsize\textheight
-  \columnwidth\textwidth
-  \@clubpenalty\clubpenalty
-  \if@twocolumn
-    \advance\columnwidth -\columnsep
-    \divide\columnwidth\tw@ \hsize\columnwidth \@firstcolumntrue
-  \fi
-  \hsize\columnwidth \linewidth\hsize
-  \begingroup\@floatplacement\@dblfloatplacement
-    \makeatletter\let\@writefile\@gobbletwo
-    \global \let \@multiplelabels \relax
-    \@input{\jobname.aux}%
-  \endgroup
-  \if@filesw
-    \immediate\openout\@mainaux\jobname.aux
-    \immediate\openout\@mainqry=\jobname.qry
-    \immediate\write\@mainaux{\relax}%
-  \fi
-  \process@table
-  \let\glb@currsize\@empty  %% Force math initialisation.
-  \normalsize
-  \everypar{}%
-  \@noskipsecfalse
-  \let \@refundefined \relax
-  \let\AtBeginDocument\@firstofone
-  \@begindocumenthook
-  \ifdim\topskip<1sp\global\topskip 1sp\relax\fi
-  \global\@maxdepth\maxdepth
-  \global\let\@begindocumenthook\@undefined
-  \ifx\@listfiles\@undefined
-    \global\let\@filelist\relax
-    \global\let\@addtofilelist\@gobble
-  \fi
-  \gdef\do##1{\global\let ##1\@notprerr}%
-  \@preamblecmds
-  \global\let \@nodocument \relax
-  \global\let\do\noexpand
-  \ignorespaces}
-%
-\def\enddocument{%
-   \ifx\@biography\@empty\else{\par\ifbrief\vskip10pt\fi\biofont\noindent\@biography\par}\fi
-   \@enddocumenthook
-   \@checkend{document}%
-   \immediate\closeout\@mainqry
-   %\ifquery
-   %   \process@queries\clearpage
-   %\else
-      \ifodd\c@page\clearpage\thispagestyle{empty}\null\clearpage\else\clearpage\fi
-   %\fi
-%   \ifquery\clearpage\else\ifodd\c@page\clearpage\thispagestyle{empty}\null\clearpage\else\clearpage\fi\fi
-   \begingroup
-     \if@filesw
-       \immediate\write\@mainaux{\string\questionmark{}}%
-       \immediate\closeout\@mainaux
-       \let\@setckpt\@gobbletwo
-       \let\@newl@bel\@testdef
-       \@tempswafalse
-       \makeatletter \input\jobname.aux
-     \fi
-     \@dofilelist
-     \ifdim \font@submax >\fontsubfuzz\relax
-       \@font@warning{Size substitutions with differences\MessageBreak
-                  up to \font@submax\space have occured.\@gobbletwo}%
-     \fi
-     \@defaultsubs
-     \@refundefined
-     \if@filesw
-       \ifx \@multiplelabels \relax
-         \if@tempswa
-           \@latex@warning@no@line{Label(s) may have changed.
-               Rerun to get cross-references right}%
-         \fi
-       \else
-         \@multiplelabels
-       \fi
-     \fi
-   \endgroup
-   \deadcycles\z@\@@end}
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Trimmarks  %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-\def\trimmarks{%
-  \vbox to 0pt{\offinterlineskip%
-    \vskip-25pt\parindent0pt
-    %\draftnote
-    \offinterlineskip}%
-    \vbox to 0pt{\hsize\trimwidth\offinterlineskip
-    \parindent0pt\leftskip0pt\rightskip0pt
-    \vbox to \trimheight{\offinterlineskip\parindent0pt
-    \hbox to \trimwidth{\vbox to 2pc{\vskip-3.5pc\hbox{\vrule height2pc width\trimrule}}\raisebox{2pc}{\hbox{\hskip-3.5pc\vrule width2pc height\trimrule}}\hfill
-      \raisebox{2pc}{\vrule width2pc height\trimrule\hskip-3.75pc}
-      \vbox to 2pc{\vskip-3.5pc\hbox{\vrule height2pc width\trimrule}}%
-    }\vfill
-    \hbox to \trimwidth{\hbox{\hskip-3.5pc\vrule height\trimrule width2pc}\vbox to 3pc{\vspace*{4.5pc}\hbox{\hskip1.5pc\vrule width\trimrule height2pc}}\hfill
-    \vbox to 3.5pc{\vskip5pc\hbox{\vrule height2pc width\trimrule}}\rlap{\hskip1.5pc\vrule width2pc height\trimrule}}}}
-\insidedraftrules}
-%
-\def\insidedraftrules{\setlength\overfullrule{0pt}\vbox to 0pt{%
-  \offinterlineskip\parindent0pt
-  \vskip \topmargin
-  \tempdimen\normaltextheight
-  \advance\tempdimen\headheight
-  \advance\tempdimen\headsep
-  \moveright\@themargin
-  \vbox{\vbox to 0pt{\vskip\headheight\vskip\headsep
-   \vrule height\draftrule width\textwidth}
-   \hbox{\fboxsep0pt\fboxrule\draftrule
-        \fbox{\vbox to \tempdimen
-                    {\hsize\textwidth\hskip\textwidth}}}}}}%
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Output Routine  %%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-\def\@outputpage{%
-\begingroup%
-  \let \protect \noexpand
-  \@resetactivechars
-  \@parboxrestore
-  \shipout \vbox{%
-    \set@typeset@protect
-    \aftergroup \endgroup
-    \aftergroup \set@typeset@protect
-  \if@specialpage
-    \global\@specialpagefalse\@nameuse{ps@\@specialstyle}%
-  \fi
-  \if@twoside
-    \ifodd\count\z@ \let\@thehead\@oddhead \let\@thefoot\@oddfoot
-         \let\@themargin\oddsidemargin
-    \else \let\@thehead\@evenhead
-       \let\@thefoot\@evenfoot \let\@themargin\evensidemargin
-    \fi
-  \fi
-  \reset@font
-  \normalsize
-  \baselineskip\z@skip \lineskip\z@skip \lineskiplimit\z@
-    \@begindvi\trimmarks
-    \vskip \topmargin
-    \moveright\@themargin \vbox {%
-      \setbox\@tempboxa \vbox to\headheight{%
-        \vfil
-        \color@hbox
-          \normalcolor
-          \hb@xt@\textwidth {%
-            \let \label \@gobble
-            \let \index \@gobble
-            \let \glossary \@gobble
-            \@thehead
-            }%
-        \color@endbox
-        }%
-      \dp\@tempboxa \z@
-      \box\@tempboxa
-      \vskip \headsep
-      \box\@outputbox
-      \baselineskip \footskip
-      \color@hbox
-        \normalcolor
-        \hb@xt@\textwidth{%
-          \let \label \@gobble
-          \let \index \@gobble
-          \let \glossary \@gobble
-          \@thefoot
-          }%
-      \color@endbox
-      }%
-    }%
-\global \@colht \textheight
-\stepcounter{page}%
-\let\firstmark\botmark
-  \ifodd\c@page
-     \ifspreadlong\global\spreadlongfalse
-     \enlargethispage{\@spreadlong}\fi
-  \fi
-}
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Miscelleneous %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-\newcommand\callout[1]{\ifdraftrules\marginpar{\bf#1}\fi}
-%
-\def\comment{\@ifnextchar[\@comment{\@comment[\relax]}}
-\def\@comment[#1]#2{\marginpar{\ifx#1\relax\else\vspace*{#1}\fi\bf\raggedright#2}}
-%
-\def\pos@of@dbl@text{0pt}
-\def\movetext{\vrule height\z@ depth\pos@of@dbl@text width\z@}
-%
-\def\acknowledgments{\par\gdef\withintwocol{true}%\starttwocolumn
-   %\subsection
-   \xhead*{Acknowledgments}%\vspace*{2pt}
-   \ackfont}
-\def\endacknowledgments{\par}
-%
-\newenvironment{displaytext}{\par\addvspace{14pt plus2pt}\bgroup\hangindent18pt\parindent\z@\let\sc\scshape\let\it\itshape}{\par\egroup\addvspace{12pt plus2pt}\@endparenv}
-%
-\newenvironment{algorithm}{\par\list{}{\leftmargin\z@\labelwidth\z@\labelsep\z@\itemsep6pt\topsep12pt plus2pt
-                        \let\makelabel\algorithmlabel}}{\endlist}
-\def\algorithmlabel#1{{\bfseries #1: }}
-%
-\newenvironment{dialogue}{\par\addvspace{12pt plus2pt}\normalsize\hangindent18pt\parindent\z@}{\par\addvspace{12pt plus2pt}\@endparenv}
-%
-\newenvironment{deflist}{\par\list{}{\leftmargin18pt\rightmargin18pt\itemindent-\leftmargin\labelwidth\z@\labelsep\z@\itemsep6pt\topsep12pt plus2pt
-                        \let\makelabel\definitionlabel}\raggedright}{\endlist}
-\def\definitionlabel#1{{\bfseries #1:} }
-%
-\newcounter{rules}
-\newenvironment{rules}{\par\addvspace{12pt plus2pt}
-   \global\addtocounter{rules}{1}\noindent{\bfseries Rule \therules: }\noindent\ignorespaces%
-   }{\par\@endparenv}
-%
-\@namedef{rules*}{\par\addvspace{12pt plus2pt}\noindent{\bfseries Rule: }\noindent\ignorespaces}
-\@namedef{endrules*}{\par\@endparenv}
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-\def\str@right{right}
-\def\str@left{left}
-%
-\newwrite\@mainqry
-%
-\newcounter{qcount}
-\newdimen\qcountdim\qcountdim0pt
-\def\qtoafont{\reset@font\normalsize\bfseries\raggedright}
-%\def\queryfont{\fontsize{9}{10.5}\selectfont}
-\def\qlist@headfont{\fontsize{15}{15}\selectfont\centering}
-\def\qlist@subheadfont{\fontsize{12}{14}\selectfont\bfseries}
-\def\qlist@font{\fontsize{10}{12}\selectfont}
-%
-\def\qtoa{\@ifnextchar[{\@qtoa}{\@qtoa[\@empty]}}
-\def\@qtoa[#1]#2{\@ifnextchar[{\@@qtoa[#1]#2}{\@@qtoa[#1]#2[0pt]}}
-\def\@@qtoa[#1]#2[#3]{\global\addtocounter{qcount}{1}%
-   \protected@write{\@mainqry}{}{\string\item{} #2\par}%
-   \@question{#1}{Q\theqcount}{#3}\ignorespaces}
-%
-\newenvironment{qlist}{\par\list{}{\usecounter{enumi}\topsep30pt\labelsep5pt\settowidth{\labelwidth}{Q\theqcount:}\leftmargin\labelwidth\labelsep5pt\advance\leftmargin\labelsep\itemsep\baselineskip\rightmargin\z@\def\makelabel##1{\hbox
-      to\labelwidth{\hss Q\theenumi.}}}}{\endlist}
-%
-%
-\def\qtom{\@ifnextchar[{\@qtom}{\@qtom[\@empty]}}
-\def\@qtom[#1]#2{\@ifnextchar[{\@@qtom[#1]#2}{\@@qtom[#1]#2[0pt]}}
-\def\@@qtom[#1]#2[#3]{\@question{#1}{#2}{#3}\ignorespaces}
-%
-\def\@question#1#2#3{%
-   \ifvmode%
-      \@@question{#1}{#2}{#3}%
-   \else%
-      \vadjust{\vbox to 0pt{%
-      \vskip-7.5pt\@@question{#1}{\qtoafont#2}{#3}\vskip7.5pt}}%
-   \fi}%
-%
-\def\@@question#1#2#3{\edef\@argone{#1}\hbox to \hsize{%
-   \if@twocolumn%
-      \if@firstcolumn
-         \ifx\@argone\str@right
-            \hfill\rlap{\hskip\marginparsep%
-               \vbox to 0pt{\hsize\marginparwidth\vspace*{#3}{#2}\endgraf\vss}}%
-         \else%
-            \tempdimen\columnwidth\advance\tempdimen-\hsize%
-            \ifdim\columnwidth>\hsize\hskip-\tempdimen\fi%
-            \hskip-\marginparsep\llap{\hskip\columnwidth%
-               \vbox to 0pt{\hsize\marginparwidth\vspace*{#3}{#2}\endgraf\vss}}%
-         \fi%
-      \else
-         \ifx\@argone\str@left
-            \tempdimen\columnwidth\advance\tempdimen-\hsize%
-            \ifdim\columnwidth>\hsize\hskip-\tempdimen\fi%
-            \hskip-\marginparsep\llap{\hskip\columnwidth%
-               \vbox to 0pt{\hsize\marginparwidth\vspace*{#3}{#2}\endgraf\vss}}%
-         \else
-            \hfill\rlap{\hskip\marginparsep%
-               \vbox to 0pt{\hsize\marginparwidth\vspace*{#3}{#2}\endgraf\vss}}%
-         \fi%
-      \fi%
-   \else%
-      \ifx\@argone\str@left%
-         \tempdimen\columnwidth\advance\tempdimen-\hsize%
-         \ifdim\columnwidth>\hsize\hskip-\tempdimen\fi%
-         \hskip-\marginparsep\llap{\hskip\columnwidth%
-            \vbox to 0pt{\hsize\marginparwidth\vspace*{#3}{#2}\endgraf\vss}}%
-      \else%
-         \hfill\rlap{\hskip\marginparsep%
-         \vbox to 0pt{\hsize\marginparwidth\vspace*{#3}{#2}\endgraf\vss}}%
-      \fi%
-   \fi}\ignorespaces}%
-%
-\newif\ifspreadlong
-\def\spreadlong#1{\ifodd\c@page\wlog{Ignoring spreadlong}
-                  \else
-                     \spreadlongtrue\gdef\@spreadlong{#1}%
-                     \enlargethispage{#1}%
-                  \fi}
-%
-\def\leaflong#1{\enlargethispage{#1}}
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-\let\dochead\sptitle
-\let\quote\extract
-\let\endquote\endextract
-\let\tcaption\tbl
-\let\unenumerate\unnumlist
-\let\endunenumerate\endunnumlist
-\let\definition\numdef
-\let\enddefinition\endnumdef
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-%\let\starttwocolumn\relax
-\newdimen\@twocolsep\@twocolsep=16pt
-\def\twocolsep#1{\global\advance\@twocolsep by #1\relax}
-\newbox\partialpage
-\def\starttwocolumn{%
-  {\output={\global\setbox\partialpage=\vbox{\unvbox255}}\newpage}%
-  \twocolumn[\unvbox\partialpage\vspace{\@twocolsep}]%
-}
-%
-\def\cleardoublepage{\clearpage\if@twoside \ifodd\c@page\else
-    \hbox{}\newpage\thispagestyle{empty}\if@twocolumn\hbox{}\newpage\thispagestyle{empty}\fi\fi\fi}
-%
-\def\onecolumnnew{%
-  %\clearpage
-  \global\columnwidth\textwidth
-  \global\hsize\columnwidth
-  \global\linewidth\columnwidth
-  \global\@twocolumnfalse
-  \col@number \@ne
-  \@floatplacement}
-%
-\def \twocolumnnew {%
-  %\clearpage
-  \global\columnwidth\textwidth
-  \global\advance\columnwidth-\columnsep
-  \global\divide\columnwidth\tw@
-  \global\hsize\columnwidth
-  \global\linewidth\columnwidth
-  \global\@twocolumntrue
-  \global\@firstcolumntrue
-  \col@number \tw@
-  \@ifnextchar [\@topnewpage\@floatplacement
-}
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-\newcount\tblcolcount
-\newskip\tcsepbefore
-%
-\def\@addamp{\global\advance\tblcolcount\@ne\relax%
-  \if@firstamp
-    \@firstampfalse
-  \else
-    \edef\@preamble{\@preamble &}%
-  \fi}
-%
-\def\CLINE#1{\expandafter\@CLINE#1\@nil}
-\def\@CLINE#1-#2\@nil{%
-  \omit%
-  \@multicnt#1%
-  \advance\@multispan\m@ne%
-  \ifnum\@multicnt=\@ne\@firstofone{&\omit}\fi%
-  \@multicnt#2%
-  \advance\@multicnt-#1%
-  \advance\@multispan\@ne%
-  \ifthenelse{#1=1}{\tcsepbefore\z@}{\tcsepbefore\tabcolsep}%
-  \ifthenelse{\the\tblcolcount=#2}{%
-     {\kern\tcsepbefore\leaders\hrule\@height.5pt\hfill\kern\tabcolsep}%
-   }{%
-     \kern\tcsepbefore\leaders\hrule\@height.5pt\hfill\kern\tabcolsep}%
-  \cr
-  \noalign{\vskip-\arrayrulewidth}}%
-%
-\def\fulltabular{\global\tblcolcount\z@\def\@halignto{to \textwidth}\@tabular}
-\def\endfulltabular{\endtabular\global\tblcolcount\z@}
-%
-%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
-%
-\usepackage{amsfonts,amssymb}
-\usepackage{multicol}
-\multicolsep = 14pt plus 4pt% minus 3pt
-%
-%
-\usepackage{dcolumn}
-\newcolumntype{D}[3]{>{\DC@{#1}{#2}{#3}}l<{\DC@end}}
-%
-\AtBeginDocument{%
-\@ifpackageloaded{lingmacros}{%
-\long\def\@enumsentence[#1]#2{\begin{list}{}{\topsep12pt%
-\labelsep6pt\leftmargin30pt\labelwidth30pt\advance\leftmargin by\labelsep\advance\leftmargin by\widelabel \advance\labelwidth by \widelabel}
-\item[#1] #2
-\end{list}}
-}{}
-}
-%
-\AtBeginDocument{
-
-\DeclareFontFamily{OML}{pns}{}
-\DeclareFontShape{OML}{pns}{m}{n}{<->  zplmr7m }{}
-\DeclareSymbolFont{itgreek}{OML}{zplmr7m}{m}{n}
-\SetSymbolFont{itgreek}{normal}{OML}{pns}{m}{n}
-
-%%%%%%%%%%%%%%%%%%% Uppercase Greek Italic  %%%%%%%%%%%%%%%%%%%%%%%%%%
-  \DeclareMathSymbol{\slGamma}  {\mathalpha}{itgreek}{"00}
-  \DeclareMathSymbol{\slDelta}  {\mathalpha}{itgreek}{"01}
-  \DeclareMathSymbol{\slTheta}  {\mathalpha}{itgreek}{"02}
-  \DeclareMathSymbol{\slLambda} {\mathalpha}{itgreek}{"03}
-  \DeclareMathSymbol{\slXi}     {\mathalpha}{itgreek}{"04}
-  \DeclareMathSymbol{\slPi}     {\mathalpha}{itgreek}{"05}
-  \DeclareMathSymbol{\slSigma}  {\mathalpha}{itgreek}{"06}
-  \DeclareMathSymbol{\slUpsilon}{\mathalpha}{itgreek}{"07}
-  \DeclareMathSymbol{\slPhi}    {\mathalpha}{itgreek}{"08}
-  \DeclareMathSymbol{\slPsi}    {\mathalpha}{itgreek}{"09}
-  \DeclareMathSymbol{\slOmega}  {\mathalpha}{itgreek}{"0A}
-  \DeclareMathSymbol{\slupDelta}{\mathalpha}{itgreek}{"01}
-  \DeclareMathSymbol{\slupOmega}{\mathalpha}{itgreek}{"0A}
-
-%%%%%%%%%%%%%%%%%%% Lowercase Greek Italic  %%%%%%%%%%%%%%%%%%%%%%%%%%
-\DeclareMathSymbol{\slalpha}{\mathalpha}{itgreek}{"0B}
-\DeclareMathSymbol{\slbeta}{\mathalpha}{itgreek}{"0C}
-\DeclareMathSymbol{\slgamma}{\mathalpha}{itgreek}{"0D}
-\DeclareMathSymbol{\sldelta}{\mathalpha}{itgreek}{"0E}
-\DeclareMathSymbol{\slepsilon}{\mathalpha}{itgreek}{"0F}
-\DeclareMathSymbol{\slzeta}{\mathalpha}{itgreek}{"10}
-\DeclareMathSymbol{\sleta}{\mathalpha}{itgreek}{"11}
-\DeclareMathSymbol{\sltheta}{\mathalpha}{itgreek}{"12}
-\DeclareMathSymbol{\sliota}{\mathalpha}{itgreek}{"13}
-\DeclareMathSymbol{\slkappa}{\mathalpha}{itgreek}{"14}
-\DeclareMathSymbol{\sllambda}{\mathalpha}{itgreek}{"15}
-\DeclareMathSymbol{\slmu}{\mathalpha}{itgreek}{"16}
-\DeclareMathSymbol{\slnu}{\mathalpha}{itgreek}{"17}
-\DeclareMathSymbol{\slxi}{\mathalpha}{itgreek}{"18}
-\DeclareMathSymbol{\slpi}{\mathalpha}{itgreek}{"19}
-\DeclareMathSymbol{\slrho}{\mathalpha}{itgreek}{"1A}
-\DeclareMathSymbol{\slsigma}{\mathalpha}{itgreek}{"1B}
-\DeclareMathSymbol{\sltau}{\mathalpha}{itgreek}{"1C}
-\DeclareMathSymbol{\slupsilon}{\mathalpha}{itgreek}{"1D}
-\DeclareMathSymbol{\slphi}{\mathalpha}{itgreek}{"1E}
-\DeclareMathSymbol{\slchi}{\mathalpha}{itgreek}{"1F}
-\DeclareMathSymbol{\slpsi}{\mathalpha}{itgreek}{"20}
-\DeclareMathSymbol{\slomega}{\mathalpha}{itgreek}{"21}
-\DeclareMathSymbol{\slvarepsilon}{\mathalpha}{itgreek}{"22}
-\DeclareMathSymbol{\slvartheta}{\mathalpha}{itgreek}{"23}
-\DeclareMathSymbol{\slvarpi}{\mathalpha}{itgreek}{"24}
-\DeclareMathSymbol{\slvarrho}{\mathalpha}{itgreek}{"25}
-\DeclareMathSymbol{\slvarsigma}{\mathalpha}{itgreek}{"26}
-\DeclareMathSymbol{\slvarphi}{\mathalpha}{itgreek}{"27}
-}
-
-%
-\medmuskip=4mu 
-\thickmuskip=5mu 
-%
-\pagestyle{headings}
-\pagenumbering{arabic}
-%
-%\makeindex
-\frenchspacing
-\sloppy
-%
-%\ifprinter
-%   \voffset-6.03pc
-%   \hoffset-6.03pc
-%\else
-   \voffset-1pc
-   \hoffset-1.5pc
-%\fi
-%
-\endinput
diff --git a/Publication/tex_manuscript/manuscript.tex b/Publication/tex_manuscript/manuscript.tex
deleted file mode 100644
index 8ad5b3338322a9401a15633d9084875d911bacf2..0000000000000000000000000000000000000000
--- a/Publication/tex_manuscript/manuscript.tex
+++ /dev/null
@@ -1,548 +0,0 @@
-\documentclass{clv3}
-
-\usepackage{hyperref}
-\usepackage{xcolor}
-\definecolor{darkblue}{rgb}{0, 0, 0.5}
-\hypersetup{colorlinks=true,citecolor=darkblue, linkcolor=darkblue, urlcolor=darkblue}
-
-\bibliographystyle{compling}
-
-% test compatibility with algorithmic.sty
-%\usepackage{algorithmic}
-
-\issue{1}{1}{2016}
-
-%Document Head
-\dochead{Journal of Computational Linguistics}
-
-
-\runningtitle{Science Accessibility Project}
-
-\runningauthor{Russell J. Jarvis}
-
-\begin{document}
-
-\title{Science Accessibility Project}
-
-\author{Russell J. Jarvis\thanks{Tempe, ASU. E-mail: rjjarvis@asu.edu.}}
-\affil{School of Life Sciences ASU}
-
-\author{Another Author\thanks{PITC Building}}
-\affil{Publishing / SPi}
-
-\author{And Another Author}
-\affil{Publishing / SPi}
-
-\author{And Yet Another}
-\affil{Publishing / SPi}
-
-\maketitle
-
-\begin{abstract}
-This article describes how we scrapped a wide variety of written word from the internet, and then analysed scrapped samples, in order to better charecterize the readability of scientific and non scientific writings.
-\end{abstract}
-
-\section{The issue of Irreducible Complexity}
-It has been observed that the readability of scientific texts is declining with increased time[]. Decreasing readability, is usually due to increased writting complexity. One could argue that scientific writing is increasing in complexity, as this reflects, the obvious advances in science methods and analysis, which increase the complexity in the act of simply doing science. Afterall manys cientific advances add details to existing complex models and imaging techniques.\\
-\\
-Given that the act of science is becoming more complex, is it reasonable to argue that all scientific writing needs to be complex in order to capture irreducible complexity? We argue that there are some great counter examples to the idea of irreducible complexity.\\
-\\
-
-\subsubsection{}keywords} 
-Bag of words analysis, naive Bayes.
-
-\section{Introduction}
-Non-scientific writing typically exceeds genuine scientific writing in one important criteria: in contrast to genuine science non-science ideas are often expressed with a more accessible writing style. We believe non-science writing occupies a more accessible style niche, that academic science writing should also occupy. We show that we can use machine learning to predict the status of writing: popular culture writing, opinionative writing, and traditional science, by first scrapping a large variety of web documents, and then classifying among the different writing types. By predicting which of the several different writing types any writing piece occupies, we are able to characterize among different writing niches.\\
-\\
-Multiple stake holders can benefit when science is communicated with lower complexity expression of ideas. With lower complexity science writing, knowledge would be more readily transferred into public awareness, additionally, digital organization of facts derived from journal articles would occur more readily, as successful machine comprehension of documented science would likely occur with less human intervention.\\
-\\
-Objectively describing the different character of the different writing styles will allow us to prescribe how, to shift academic science writing into a more accessible niche, where science can more aggressively compete with pseudo-science, and blogs, facilitating greater knowledge transference, at a moment in history when public awareness is critically at stake.\\
-\\
-The accessibility of written word can be approximated by a computer program that reads over the text, and guesses the mental difficulty, associated with comprehending a written document. The computer program maps reading difficult onto a quantity that represents the number of years of schooling needed to decode the language in the document. For convenience, we can refer to the difficulty associated with the text as the 'complexity' of the document. 
-
-
-\section{Introduction Old}
-
-Non-scientific writing typically exceeds genuine scientific writing in one important criteria: in contrast to genuine science non-science ideas are often expressed using a less complex more engaging writing style.  Yet, multiple stake holders could benefit if science was communicated using a less complex expression of ideas. Using less complicated science writing, knowledge could be more readily transferred into public awareness, also, and machine reading machine organization, and machine action on factual information derived from journal articles could occur more readily.
-
-We believe non-science writing occupies a style niche, that academic science writing should also occupy. We show that we can blindly predict the status of writing: popular culture writing, opinionative writing, and traditional science, by using machine learning to classify the different writing types. By predicting which of the several different writing types any writing piece occupies, we are able to characterize among different writing niches.
-
-Objectively describing the different character of the different writing styles will allow us to prescribe how, to shift academic science writing into a more accessible niche, where science can more aggressively compete with pseudo-science, and blogs, facilitating greater knowledge transference, at a moment in history when public awareness is critically at stake.
-
-
-\section{Methods}
-
-We built a web-scraping and written text analysis infrastructure by extending many existing Free and Open Source (FOS) tools. The Web scrapping interface employs several common python modules, chief among those was: Google Scrape, Beautiful Soup and Selenium. The Text analysis infrastructure was based on the two substantial code bases Text-stat, which contained measures of text reading level (complexity), and NLTK (the Natural Language Processing Tool Kit), which contained measures of text subjectivity, and sentiment type.
-
-The scraping, and analysis work we performed, rested on top of a large hierarchy of software dependencies. However, it is increasingly well understood, that, dependency heavy software stacks act as a significant impediment to investigating or reproducing any product of digital, scholary research. In order to address this problem and to enhance the reproducibility of our approach, we created the necessary web-scrapping, and analysis infrastructure inside a dedicated Docker Container.
-
-Reproducibility is burdened by the technical task of satisfying each software dependencies necessary to recreate a digital scientific experiment. Our position, of starting with from a cloned software environment, will mitigate, the burden of duplicating our digital research environment. That is why we have used a Docker file, and associated Docker container together, as they act as self-documenting and extremely portable software environment clone.
-
-Initially we created two different, unrelated and broad ranging lists of scientific queries. The first type of query was predominantly cultural in nature, or world view related. The second set of queries represented gains in knowledge about physical entities or physical processes in the world. We were interested in scientific, and pseudo-scientific writing.\\
-
-There were two types of writing that we actively excluded from our analysis, those were websites expressed in a non-English language, and also websites, that were highly commercial in nature. These were websites advertisements, of consumer goods, and online shopping generally. These websites, utilizing wording, that significantly biases text stat metrics. Webpages of less than 400 words, were most often advertisements, and websites of a comercial nature such as Amazon shopping.\\
-
-Non English, websites were excluded for the simple reason that they are not amenable to Textstat, and NLTK tools, however, even if this was not the case, it is also known, that there are  significant differences in per-word information entropy between different natural languages.
-
-\subsection{Search Engine Queries:}
-The first two lists of queries were chosen to be belong to an overt set of exclusively scientific or cultural search terms. A third list of terms was designed to be deliberately ambiguous.
-
-\subsubsection{Science Queries:}
-The three lists of search engine queries were as implemented as follows: science engine queries: 'evolution', 'photo-sysnthesis' ,'Transgenic', 'GMO', 'climate change', 'cancer', 'Vaccines', 'Genetically Modified Organism', ‘differential equation’,"psycho-physics","soma”
-
-\subsubsection{Cultural queries were as follows:} 'reality TV', 'prancercise philosophy', 'play dough delicious deserts', 'unicorn versus brumby', 'football soccer' , god fearing.
-
-As discussed we also designed ambiguous queries which were equally likely return content that was either scientific, or non scientific in nature. The reason for doing this was to provide a challenge for the classification algorithm.
-
-\subsubsection{Ambiguous Queries, were as follows:}
-
-"the singularity","skynet","","killer robot","franken-science","Frankenstein,”the God Delusion","god does not play dice", "the selfish gene","political science", ", "requim for a spike"
-
-
-After scraping across the two different lists were performed, the resulting queries were filtered, according to specific sets of criteria. As stated previously, we discarded from our analysis, web pages that were not written in English, since we did not have the necessary tools, to analyze them.
-
-\subsection{Text Metrics:}
-A list of metrics applied to downloaded corpus include: TextStat, which was used to measure word complexity (an average of several important word complexity metrics, such as the Gunning Fog measure of reading level), LZW compression-ratio, de-compression ratio, sentiment analysis, subjectivity analysis, and page rank.
-
-Compression ratios were used to investigate the notion, that well written scientific writing, might simply be lower in information entropy, and an information theoretic analysis, can be used to both to better characterize, and corroborate the findings of other reading word complexity metrics
-
-\subsection{Reference Texts:}
-
-Some reference texts were used, as a means of providing contrast, and context, to data points, among our web scraping derived corpus. The Upgoer5 is a library, of scientific texts, written with the aid of a text editor, which imposes, that output documents are exclusively comprised by, only the 10,000 most commonly occurring English words.
-
-The Post Modern Essay Generator (PMEG), embodies an artificial English synthesis technique. Documents that are output from the website, consist of sentences that obey the rules of written English, however there are no restraints governing the semantic conceptual references in the sentences. If any particular sentence in a PMEG document, embodies an objective meaning, it is only by chance. Output from PMEG reads as highly coded, and vague.
-
-The reference data points in some ways provide further validation to the existing text metric tools, as we needed to verify that word readability metrics provided results that were consistant with prior assumptions about known texts. For instance, the corpus derived from upgoer editor, should require a very low reading grade level to understand. Texts, from the PMEG should require a very high reading level to understand, and cumulative entropy of such texts should be high.
-
-
-\verb|\documentclass[bookreview,manuscript]{clv3}|
-
-%\subsection{Default Option}
-
-\begin{deflist}
-\item[bookreview] Sets the article layout for Book Review.
-\item[brief] Sets the article layout for Briefly Noted.
-\item[discussion] Sets the article layout for Squibs and Discussions.
-\item[pubrec] Sets the article layout for Publication Received.
-\item[shortpaper] Sets the article layout for Short Paper.
-\item[manuscript] Sets the baseline spacing to double space. This
-option can be used in combination with other options.
-\end{deflist}
-
-By not declaring any option in the \verb|\documentclass| command the class file
-will automatically set to standard article layout.
-
-\section{Results}
-
-We created a total list all of the different queries obtained, from combining both the list of cultural queries, and scientific queries, and then applied such queries exclusively to the Wikipedia, search interface. We take the result of evaluating this pool of queries, and then plot the resulting pool of queries versus page rank. The Wikipedia, actually showed a small but consistant preference for web pages of higher complexity.
-
-%\begin{figure}
-%	\centering
-%	\includegraphics[width=0.7\linewidth]{screenshot001}
-%	\caption{}
-%	\label{fig:figure1}
-%\end{figure}
-
-%
-
-The plots below may appear to look a bit unprofessional, as not all data points have error bars. The reason for this, is pandas+seaborne allows you to plot on the same axis multiple sample data points, and single sampled data points. Only of the five mentioned search engines was made to sample beyond page rank 10, but all five sampled under page rank 10.
-
-%\begin{figure}
-%	\centering
-%	\includegraphics[width=0.7\linewidth, height=0.7\textheight]{figure2}
-%	\caption{}
-%	\label{fig:figure2}
-%\end{figure}
-%
-
-The same plot but with mean and std deviation plots when multiple samples per page rank are available. When we plotted the Wikipedia queries, where reading (text-stat standard) level is instead plotted against page rank, we again see that there is a slight trend towards increasing text complexity with decreasing page rank. 
-
-
-If instead we consider how resistant Wikipedia pages are to compression, we see that low page rank pages, are more resistant to compression. This finding recapitulates the same result as the above figure, where increasing Wikipedia page rank slightly decreases text complexity.
-
-When aggregated search results between all possible search engines, and then plotting page rank versus complexity, for particular types of search terms, there was a strong positive correlation between page rank, and reading level.
-
-In the case of GMO, and Genetically Modified Organism, positive increasing trends where observed
-
-The title page is created using the standard \LaTeX\ command \verb|\maketitle|.
-Before this command is declared, the author must declare all the data which are
-to appear in the title area.\footnote{$\backslash$maketitle is the command to execute all the title page information.}
-
-\subsection{Volume, Number and Year}
-
-The command \verb|\issue{vv}{nn}{yyyy}| is used in declaring the volume, number
-and year of the article. The first argument is for the volume, the second argument
-is for the issue number. Volume and Issue number will appear on the even page
-running head opposite the journal name. The third argument is for the Year which
-will appear in the copyright line at the bottom of the title page.
-
-\subsection{Document Head}
-
-Document head is produced with the command \verb|\dochead{Document Head}|. Doc head
-will output differently, or may not appear at all, depending on the option used in the
-documentclass.
-
-\subsection{Paper Title}
-
-The paper title is declared like: \verb|\title{Computer Linguistic Article}|
-in the usual \LaTeX manner. Line breaks may be inserted with (\verb|\\|) to equalize
-the length of the title lines.
-
-\subsection{Authors}
-The name and related information for authors is declared with the \verb|\author{}| command.
-
-The \verb|\thanks{}| command produces the ``first footnotes.''. \LaTeX\ \verb|\thanks|
-cannot accommodate multiple paragraphs, author will have to use a separate \verb|\thanks|
-for each paragraph.
-
-The \verb|\affil{}| command produces the author affiliations that appears right under
-the author's name.
-
-\subsection{Running Headers}
-The running heads are declared with the \verb|\runningtitle{Running Title}| for the
-journal name and \verb|\runningauthor{Author's Surname}| for author. These information
-will appear on the odd pages. For {\tt bookreview} option, odd page running head is
-automatically set to "Book Reviews". Even page running head is default to Computational
-Linguistics opposite volume and issue number.
-
-\subsection{History Dates}
-
-History dates are declared with \verb|\historydates{Submission received:...}|. This data
-should contain Submission, Revised and Accepted date of the article. History dates appear
-at the footnote area of title age.
-
-
-\section{Abstract}
-
-Abstract is the first part of a paper after \verb|\maketitle|. Abstract text is
-placed within the abstract environment:
-
-\begin{verbatim}
-\begin{abstract}
-This is the abstract text . . .
-\end{abstract}
-\end{verbatim}
-
-\section{Section Headings}
-
-Section headings are declared in the usual \LaTeX\ way via \verb|\section{}|,
-\verb|\subsection{}|, \verb|\subsubsection{}|, and \verb|\paragraph{}|. The
-first 3 levels of section head will have Arabic numbering separated
-by period. The \verb|\paragraph{}| section will have the title head in Italics
-and at the same line with the first line of succeeding paragraph.
-
-\section{Citations}
-Citations in parentheses are declared using the \verb|\cite{}|
-command, and appear in the text as follows:
-This technique is widely used \cite{woods}.
-The command \verb|\citep{}| (cite parenthetical) is a synonym of \verb|\cite{}|.
-
-Citations used in the sentence are declared using the \verb|\namecite{}|
-commands, and appear in the text as follows:
-\namecite{woods} first described this technique.
-The command \verb|\citet{}| (cite textual) is a synonym of \verb|\namecite{}|.
-
-This style file is designed to be used with the BibTeX
-style file \verb|compling.bst|.  Include the command
-\verb|\bibliographystyle{compling}| in your source file.
-
-Citation commands are based on the \verb|natbib| package;
-for details on options and further variants of the commands,
-see the \verb|natbib| documentation.  In particular, options
-exist to add extra text and page numbers.  For example,
-\verb|\cite[cf.][ch.\ 1]{winograd}| yields: \cite[cf.][ch.\ 1]{winograd}.
-
-The following examples illustrate how citations appear both in the text
-and in the references section at the end of this document.
-\begin{enumerate}
-\item Article in journal:
- \namecite{akmajian};
- \namecite{woods}.
-\item Book:
-  \namecite{altenberg};
-  \namecite{winograd}.
-\item Article in edited collection/Chapter in book:
-  \namecite{cutler};
-  \namecite{sgall};
-  \namecite{jurafsky}.
-\item Technical report:
-  \namecite{appelt};
-  \namecite{robinson}.
-\item Thesis or dissertation:
-  \namecite{baart};
-  \namecite{spaerckjones};
-  \namecite{cahn}.
-\item Unpublished item:
-  \namecite{ayers}.
-\item Conference proceedings:
-  \namecite{benoit}.
-\item Paper published in conference proceedings:
-  \namecite{krahmer};
-  \namecite{Copestake2001}.
-\end{enumerate}
-
-
-\section{Definition with Head}
-
-Definition with head is declared by using the environment:
-\\
-\begin{verbatim}
-\begin{definition}
-Definition text. . .
-\end{definition}
-\end{verbatim}
-
-This environment will generate the word {\bf ``Definition 1''} in bold on separate
-line. The sequence number is generated for every definition environment. Definition
-data will have no indention on the first line while succeeding lines will have hang
-indention.
-
-\section{Lists}
-
-The usual \LaTeX\ itemize, enumerate and definition list environments are used
-in CLV3 style.
-
-To produce Numbered List use the environment:
-
-\begin{verbatim}
-\begin{enumerate}
-\item First numbered list item
-\item Second numbered list item
-\item Third numbered list item
-\end{enumerate}
-\end{verbatim}
-
-To produce Bulleted List use the environment:
-
-\begin{verbatim}
-\begin{itemize}
-\item First bulleted list item
-\item Second bulleted list item
-\item Third bulleted list item
-\end{itemize}
-\end{verbatim}
-
-To produce Definition List use the environment:
-
-\begin{verbatim}
-\begin{deflist}
-\item[First]  Definition list item. . .
-\item[Second] Definition list item. . .
-\item[Third]  Definition list item. . .
-\end{deflist}
-\end{verbatim}
-
-Additional list environment were also defined such as Unnumbered, Arabic and Alpha lists.
-
-Unnumbered List is the list where item labels are not generated. To produce Unnumbered List use the environment:
-
-\begin{verbatim}
-\begin{unenumerate}
-\item First list item
-\item Second list item
-\item Third list item
-\end{unenumerate}
-\end{verbatim}
-
-To produce Arabic List use the environment:
-
-\begin{verbatim}
-\begin{arabiclist}
-\item First arabic list item
-\item Second arabic list item
-\item Third arabic list item
-\end{arabiclist}
-\end{verbatim}
-
-To produce Alpha List use the environment:
-
-\begin{verbatim}
-\begin{alphalist}
-\item First alpha list item
-\item Second alpha list item
-\item Third alpha list item
-\end{alphalist}
-\end{verbatim}
-
-All the list environments mentioned above can be nested with each other.
-
-\subsection{Other List Types}
-
-\subsubsection{Outline List or Example List}
-
-\begin{verbatim}
-\begin{exlist}
-\item First outline list item. . .
-\item Second outline list item. . .
-\item Third outline list item. . .
-\end{exlist}
-\end{verbatim}
-
-\subsubsection{Output Formula or Algorithm}
-
-\begin{verbatim}
-\begin{algorithm}
-\item[Step 1] First item. . .
-\item[Step 2] Second item. . .
-\end{algorithm}
-\end{verbatim}
-
-% test compatibility with algorithmic.sty
-%\begin{algorithmic}
-%\STATE i
-%\end{algorithmic}
-
-See sample on the {\tt COLI-template.pdf}.
-
-\section{Word Formula or Displayed Text}
-
-Word formula and displayed text are text that should be displayed in a
-separate line without indention. This are achieved by using the environment:
-
-\begin{verbatim}
-\begin{displaytext}
-This is a sample of displayed text . . .
-\end{displaytext}
-\end{verbatim}
-
-\section{Dialogue}
-
-Dialogue text are presentation of people's conversation. These will be presented
-on a separate line where each dialogue starts with the name of speaker, followed by
-colon. Succeeding lines will be hang indented. To produce Dialogue use the environment:
-\\
-\begin{verbatim}
-\begin{dialogue}
-Speaker 1: dialogue. . .
-
-Speaker 2: dialogue. . .
-\end{dialogue}
-\end{verbatim}
-
-
-\noindent Please make sure to insert an empty line between dialogues.
-
-\section{Extracts}
-
-Extract text acts like quote, where left and right margins are indented.
-To produce Extract use the environment:
-
-\begin{verbatim}
-\begin{extract}
-This is an example of Extract text. . .
-\end{extract}
-\end{verbatim}
-
-\noindent See sample on the {\tt COLI-template.pdf}.
-
-\section{Theorem-like Environments}
-
-There are several theorem-like environments defined in CLV3 class file. Theorem-like
-environments generate the name of the theorem as label, and counter number in bold.
-
-\subsection{Example}
-
-To produce Example use the environment:
-
-\begin{verbatim}
-\begin{example}
-This is Example text. . .
-\end{example}
-\end{verbatim}
-
-\subsection{Lemma}
-
-To produce Lemma use the environment:
-
-\begin{verbatim}
-\begin{lemma}
-Lemma text. . .
-\end{lemma}
-\end{verbatim}
-
-This produces the following output:
-\begin{lemma}\label{lem}
-Lemma text.
-\end{lemma}
-A small vertical space separates the end of the lemma
-from the following text.
-
-\subsection{Theorem}
-
-To produce Theorem use the environment:
-
-\begin{verbatim}
-\begin{theorem}
-Theorem text. . .
-\end{theorem}
-\end{verbatim}
-
-This produces the following output:
-\begin{theorem}\label{thm}
-Theorem text.
-\end{theorem}
-\noindent
-A small vertical space separates the end of the theorem
-from the following text.
-
-\subsection{Proof}
-
-The proof environment produces a square at the end of the text. To produce Proof
-use the environment:
-
-\begin{verbatim}
-\begin{proof}
-Proof text. . .
-\end{proof}
-\end{verbatim}
-
-This produces the following output:
-\begin{proof}\label{proof}
-Proof text.
-\end{proof}
-A small vertical space separates the end of the lemma
-from the following text.
-
-\subsection{Unnumbered Theorem-like Environments}
-
-There are also unnumbered version of some of the theorem-like environments.
-These are declared by using its asterisked version. Here are the three
-unnumbered version of theorem-like environments:
-
-\begin{verbatim}
-\begin{theorem*}
-Unnumbered theorem text. . .
-\end{theorem*}
-\end{verbatim}
-
-\section{Appendix}
-
-Appendix is declared by issuing the command \verb|\appendix|. This will set
-the necessary labels to appendix's rule (i.e. (A.1) for equation number).
-
-Sections inside Appendix are declared using \verb|\appendixsection{}|, which
-will produce {\bf Appendix A: Section Title} for first section.
-
-Equation numbers are automatically set to (A.1), (A.2) and (A.3). Where the letters
-follow the current level of Appendix section. So equations on {\bf Appendix B}
-will have equation numbers as follow: (B.1), (B.2) and (B.3).
-
-\section{Acknowledgments}
-
-Acknowledgments are produce by using the environment:
-\\
-\begin{verbatim}
-\begin{acknowledgments}
-Acknowledgments text. . .
-\end{acknowledgments}
-\end{verbatim}
-
-\section{Others}
-
-Other items such as Equations, Figures, Tables and References are produced in
-the standard \LaTeX\ typesetting.
-
-\starttwocolumn
-\bibliography{compling_style}
-
-\end{document}
diff --git a/README.md b/README.md
index bf4a1661e46eb3e5d99a9a862de9ff43f5636283..5ab226ea04b8de05d0c7d15fd26f6e4e6a25c71f 100644
--- a/README.md
+++ b/README.md
@@ -1,9 +1,8 @@
 [![Build Status](https://travis-ci.com/russelljjarvis/ScienceAccessibility.png)](https://travis-ci.com/russelljjarvis/ScienceAccessibility) 
 
-[![Binder](https://mybinder.org/badge_logo.svg)](https://mybinder.org/v2/gh/russelljjarvis/ScienceAccessibility/dev)
+[![Binder](https://mybinder.org/badge_logo.svg)](https://mybinder.org/v2/gh/russelljjarvis/simple_science_access.git/master)
 
-[Also see this short publication document](
-https://github.com/russelljjarvis/ScienceAccessibility/blob/dev/comparing_the_written_language_of_scientific_and_non_scientific_sources.md)
+[Also see this short publication document](https://github.com/russelljjarvis/ScienceAccessibility/blob/master/manuscript.md)
 
 ## Overview
 Understanding a big word is hard, so when big ideas are written down with lots of big words, the large pile of big words is also hard to understand. 
diff --git a/compete.png b/compete.png
new file mode 100644
index 0000000000000000000000000000000000000000..2af439fcc09571ab0e1761878af8939420c3da4e
Binary files /dev/null and b/compete.png differ
diff --git a/for_joss_standard_dev.png b/for_joss_standard_dev.png
new file mode 100644
index 0000000000000000000000000000000000000000..291b688f5c7f9d1f93096c66a26f384bb65ff157
Binary files /dev/null and b/for_joss_standard_dev.png differ
diff --git a/paper.bib b/paper.bib
new file mode 100644
index 0000000000000000000000000000000000000000..4826d26ba5f104a1b79d586c7e8cba9a809acee0
--- /dev/null
+++ b/paper.bib
@@ -0,0 +1,29 @@
+## References
+
+[1] Kutner, Mark, Elizabeth Greenberg, and Justin Baer. "A First Look at the Literacy of America's Adults in the 21st Century. NCES 2006-470." The National Center for Education Statistics. (2006).
+
+[2] Plavén-Sigray, Pontus, Granville James Matheson, Björn Christian Schiffler, and William Hedley Thompson. "The readability of scientific texts is decreasing over time." Elife. (2017).
+
+[3] Kincaid JP, Fishburne RP Jr, Rogers RL, Chissom BS. "Derivationof new readability formulas (Automated Readability Index, FogCount and Flesch Reading Ease Formula) for Navy enlistedpersonnel".The Institue for Simulation and Training, (1975): 8–75.
+
+[4] Soldatova, Larisa, and Maria Liakata. "An ontology methodology and cisp-the proposed core information about scientific papers." JISC Project Report (2007).
+
+[5] Kuhn, Tobias. "The controlled natural language of randall munroe’s thing explainer." International Workshop on Controlled Natural Language. Springer, Cham, (2016).
+
+[6] Bulhak, Andrew C. "On the simulation of postmodernism and mental debility using recursive transition networks." Monash University Department of Computer Science (1996).  
+
+[7] Gopen, George D., and Judith A. Swan. "The science of scientific writing." American Scientist 78, no. 6 (1990): 550-558.
+
+
+@article{Pearson:2017,
+  	Adsnote = {Provided by the SAO/NASA Astrophysics Data System},
+  	Adsurl = {http://adsabs.harvard.edu/abs/2017arXiv170304627P},
+  	Archiveprefix = {arXiv},
+  	Author = {{Pearson}, S. and {Price-Whelan}, A.~M. and {Johnston}, K.~V.},
+  	Eprint = {1703.04627},
+  	Journal = {ArXiv e-prints},
+  	Keywords = {Astrophysics - Astrophysics of Galaxies},
+  	Month = mar,
+  	Title = {{Gaps in Globular Cluster Streams: Pal 5 and the Galactic Bar}},
+  	Year = 2017
+}
diff --git a/paper.md b/paper.md
new file mode 100644
index 0000000000000000000000000000000000000000..046ab01fd69422c8984282bf042abd2bbdd8a368
--- /dev/null
+++ b/paper.md
@@ -0,0 +1,98 @@
+title: 'Exploring the Readability of Scientific and Non-scientific Sources'
+		
+tags:
+  - readability
+  - science communication
+  - science writing
+
+authors:
+  - name: Russell Jarvis
+    affiliation: PhD Candidate Neuroscience, Arizona State University
+  - name: Patrick McGurrin
+    affiliation: National Institute of Neurological Disorders and Stroke, National Institutes of Health
+  - name: Shivam Bansal
+    affiliation: Senior Data Scientist, H2O.ai
+  - name: Bradley G Lusk
+    affiliation: Science The Earth; Mesa, AZ 85201, USA
+    
+date: 20 October 2019
+
+bibliography: paper.bib
+
+## Summary
+To ensure that writing is accessible to the general population, wauthors must consider the length of written text, as well as sentence structure, vocabulary, and other language features [@Kutner:2006]. While popular magazines, newspapers, and other outlets purposefully cater language for a wide audience, there is a tendency for academic writing to use more complex, jargon-heavy language [@Plavén-Sigray:2017]. 
+
+In the age of growing science communication, this tendency for scientists to use more complex language can carry over when writing in more mainstream media, such as blogs and social media. This can make public-facing material difficult to comprehend, undermining efforts to communicate scientific topics to the general public.
+
+To address this, we created a tool to analyze complexity of a given scientist’s work relative to other writing sources. The tool first quantifies existing text repositories with varying complexity, and subsequently uses this output as a reference to contextualize the readability of user-selected written work. 
+
+While other readability tools currently exist to report the complexity of a single document, this tool uses a more data-driven approach to provide authors with insights into the readability of their published work with regard to other text repositories. This will enable monitoring of the relative complexity of their writing, guiding readability improvements to online material. We hope it will help scientists interested in science writing make their published work more accessible to a broad audience, and lead to an improved global communication and understanding of complex topics.
+
+## Methods
+
+### Text Analysis Metrics
+We built a web-scraping and text analysis infrastructure by extending many existing Free and Open Source (FOS) tools, including Google Scrape, Beautiful Soup, and Selenium.
+
+We first query a number of available text repositories with varying complexity:
+
+| Text Source | Mean Complexity | Description |
+|----------|----------|:-------------:|
+| Upgoer 5                            | 6   | library using only the 10,000 most commonly occurring English words |
+| Wikipedia                               | 14.9 | free, popular, crowdsourced encyclopedia   |
+| Post-Modern Essay Generator (PMEG)  | 16.5 | generates output consisting of sentences that obey the rules of written English, but without restraints on the semantic conceptual references   |
+| Art Corpus                       | 18.68  | library of scientific papers published in The Royal Society of Chemistry |
+
+The author's name entered by the user is then queried through Google Scholar, returning the results from articles containing the author's name. 
+
+The Flesch-Kincaid readability score [@Kincaid:1975] - the most commonly used metric to assess readability - is then used to quantify the complexity of all items.
+
+### Plot Information 
+The resulting plot is a histogram binned by readability score, initially populated exclusively by the ART corpus [@Soldatova:2007] data. We use this data because it is a pre-established library of scientific papers. 
+
+The mean readability scores of Upgoer5 [@Kuhn:2016], Wikipedia, and PMEG [@Bulhak:1996] libraries are labeled on the plot to contextualize the complexity of the ART corpus data with other text repositories of known complexity. 
+
+We also include mean readability scores from two scholarly reference papers, Science Declining Over Time [@Kutner:2006] and Science of Writing [@Gopen:1990], which discuss writing to a broad audience in an academic context. We use these to demonstrate the feasibility of discussing complex content using more accessible language.
+
+Lastly, the mean reading level of the entered author's work, as well as the maximum and minimum scores, are displayed on the plot. Thus, the resulting graph displays the mean writing complexity of the entered author against a distribution of ART corpus content as these other text repositories of known complexity.
+
+### Reproducibility
+A Docker file and associated container together serve as a self-documenting and portable software environment clone to ensure reproducibility given the hierarchy of software dependencies.
+
+## Results
+Data are available here: [Open Science Framework data repository](https://osf.io/dashboard).
+
+### Setting Up the Environment (Developer)
+A docker container can be downloaded from Docker hub or built locally.
+```BASH
+docker login your_user_name@dockerhub.com
+docker pull russelljarvis/science_accessibility:slc
+mkdir $HOME/data_words
+docker run -it -v $HOME/data_words russelljarvis/science_accessibility:slc
+```
+### Running a Simple Example (User)
+After Docker installation on your Operating System, run the following commands in a BASH terminal.
+```BASH
+docker pull russelljarvis/science_accessibility_user:latest
+```
+Here is a python example to search for results from academic author Richard Gerkin. When inside the docker container, issue the command:
+```BASH
+mkdir $HOME/data_words
+docker run -v $HOME/data_words russelljarvis/science_accessibility_user "R Gerkin"
+```
+![Specific Author Relative to Distribution](for_joss_standard_dev.png)
+
+This tool also allows the entry of two author names to view whose text has the lowest average reading grade level. Public competitions and leader boards often incentivize good practices, and may also help to improve readability scores over time.
+
+![Specific Author Relative to Distribution](compete.png)
+
+
+## Future Work
+We have created a command line interface (CLI) for using this tool. However, we aim to expand this to a web application that is more user friendly to those less familiar with coding. 
+
+The readability of ART Corpus is comparable to that of other scientific journals [2], but incorporating a larger repository of journal articles of various topics, and perhaps even overlaying them on the plot, would nonetheless be beneficial. In addition, adding search engine queries of different, broad-ranging lists of search would also help to further contextualize the readability of published scientific work with regard to topics engaged by the public on a more daily basis.
+
+While the Flesch-Kincaid readability score is the most common readability metric, including other metrics, such as information entropy, word length, and compression ratios, subjectivity, and reading ease scores, will serve to provide more robust feedback to the user with regard to the complexity and structure of their written text.
+
+Finally, we believe that the idea public competition could be a fun and interactive way for scientists to improve their science communication skills, and believe there is room for expansion here as well.
+
+## References