Add tweet example

rtd_docs/source/index.rst

@@ -19,7 +19,7 @@ Contents
 
    use_cases/commits
    use_cases/documentation
-   .. use_cases/tweets
+   use_cases/tweets
 
 .. toctree::
    :maxdepth: 2

rtd_docs/source/use_cases/commits.rst

@@ -54,7 +54,7 @@ single character is changed.
 
 
 Mapping Binaries to ``F2`` event
-^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
 For this event, the SHA256 digest of the binary, will be simple and good to
 detect changes.

rtd_docs/source/use_cases/documentation.rst

@@ -77,7 +77,7 @@ the content of the file.
 
     # Create the document event
     document_hash=$(sha256sum <document> | cut -d' ' -f1 )
-    cat > docuent_event.json <<EOF
+    cat > document_event.json <<EOF
     {
         "document_hash": "${document_hash}",
     }

rtd_docs/source/use_cases/tweets.rst

@@ -1,8 +1,6 @@
 Lie Detector in Tweeter feeds
 =============================
 
-.. image:: /_static/images/Uc3.png
-
 Can we know instantly (to some extent) if a tweet is real or not?
 
 Day after day we faced the necessity to detect inconsistencies between what was
@@ -12,8 +10,85 @@ tries us to sell.
 In this **Use case** we will discuss the high throughput that is required in
 order to allow to detect inconsistencies in real time.
 
-Logarithmic long tail
----------------------
 
-Since QED is an cryptographic append-only storage, it has some design
-capabilities and limitations. One of those it that the `Merkle Tree`_ has
+Theory and Operation
+--------------------
+
+In Order to push to QED the tweets, we will need a stream publisher which can be
+created with any third-party library that has `streaming-api` capabilities like
+golang `go-twitter <https://github.com/dghubble/go-twitter/blob/master/examples/streaming.go>`_ module,
+python's `tweeepy <http://docs.tweepy.org/en/v3.4.0/streaming_how_to.html>`_ library or,
+npm `twitter <https://www.npmjs.com/package/twitter#streaming-api>`_ package.
+
+.. image:: /_static/images/Uc3.png
+
+Event Source
+++++++++++++
+
+Since the **Streaming Publisher** will act as the event source it will need to
+process all the tweets and creating the events that identifies individual tweets
+univocally. And since tweets are small in size, there's no need to digest it 
+beforehand.
+
+.. note::
+
+    Mapping example of event ``F(TWEET)``:
+
+    .. code:: json
+
+        {
+            "user_screen_name": "TwitterDev",
+            "text": "Today's new update means that you can finally add Pizza Cat to your Retweet with comments! Learn more about this ne… https://t.co/Rbc9TF2s5X",
+        }
+
+
+Auditor
++++++++
+
+The **LIE DETECTOR** service will act as the auditor giving transparency to 
+any tweet source. 
+
+Untrusted Sources
++++++++++++++++++
+
+Any source where you can fetch a tweet are susceptible to be used as untrusted
+source to give transparency by auditing the event tweets. 
+
+Creating transparency in tweets
+-------------------------------
+
+.. warning::
+
+    The following snippets are atop :ref:`Quick start`. please visit it to
+    configure the required code.
+
+We will process a tweet and creating his event ``F(TWEET)`` to send it to QED.
+
+.. code:: shell
+
+    # Create the tweet event
+    cat > tweet_event.json <<EOF
+    {
+        "user_screen_name": "TwitterDev",
+        "text": "Today's new update means that you can finally add Pizza Cat to your Retweet with comments! Learn more about this ne… https://t.co/Rbc9TF2s5X",
+    }
+    EOF
+
+Push the tweet event to QED.
+
+.. code:: shell
+
+    # pushing the tweet event to QED server
+    qed_client \
+        add \
+        --event "$(cat tweet_event.json)"
+
+And Finally retrieve and verify the proof.
+
+.. code:: shell
+
+    # Verify the proof
+    qed_client \
+        membership \
+        --event "$(cat tweet_event.json)" \
+        --auto-verify