diff --git a/docs/DesignDoc.rst b/docs/DesignDoc.rst
deleted file mode 100644
index 6fd0acf..0000000
--- a/docs/DesignDoc.rst
+++ /dev/null
@@ -1,2 +0,0 @@
-Design Document
-===============
diff --git a/docs/design.rst b/docs/design.rst
new file mode 100644
index 0000000..736f67a
--- /dev/null
+++ b/docs/design.rst
@@ -0,0 +1,206 @@
+Design Document
+===============
+
+Chatroom Discovery
+-------------------
+
+Function
+~~~~~~~~~
+
+While running the ChronoChat application, a list of ongoing chatrooms
+is maintained with the basic chatroom information which can be
+presented to the user whenever he requests. The user can select to
+join a chatroom. The application will check if the user is allowed to
+join the chatroom. For chatrooms with hierarchical trust model, the
+user will join immediately. For chatroom with web of trust
+model, if one of the user's chatroom contact is in the chatroom, the
+user will send a request for invitation. Otherwise, the application
+will tell the user that he is disallowed to join in the chatroom.
+
+Detailed Design
+~~~~~~~~~~~~~~~~
+
+When a user starts the ChronoChat application, the application will
+begin to discover ongoing chatrooms. Each of the participants in a
+chatroom needs to maintain a list of all the participants in the
+chatroom for other users to query.
+
+Two kinds of work need to be done during the chatroom discovery
+process.
+
+Discovery
++++++++++++++
+
+A user sends a broadcast interest to query for all the ongoing
+chatrooms' information with all the existing chatroom in the exclude
+filter to enumerate all the existing chatrooms and avoid retrieving
+the same data packet from in-network caches. (There can be problems
+when there are so many chatrooms and the exclude filter exceeds the
+practical limit of NDN packet size. It is an open issue. We assume
+in this version that there are not so many chatrooms and we will try
+to solve the problem in later version.) No chatroom
+needs to be added when generating the list for the first time .
+
+The name of the interest can be:
+
+``/ndn/broadcast/chronochat/chatroom-list``
+
+Every chatroom participant maintains an interest filter under the
+prefix:
+
+``/ndn/broadcast/chronochat/chatroom-list``
+
+When a participant in a chatroom receives the interest,
+the participant will check the exclude filter and send a data packet
+with all the information of the chatroom if the chatroom name is not
+in the exclude filter. There might be problems when there are too many
+participants in one chatroom which causes the Data packet to exceed
+the practical limit of NDN packet size. However, the participants'
+information is only used for a user to decide whether to join a
+chatroom. Therefore, in this case, we can only send the user
+information of a subset of participants.
+
+The name of the data packet can be:
+
+``/ndn/broadcast/chronochat/chatroom-list/chatroom-name``
+
+Because users can obtain the chatroom name through the
+data name, the data packet that the information provider sends back
+only need to contains: the name of every participant and the
+chatroom's trust model.
+
+For every participant, the data structure can be defined as:
+::
+
+   Participant := PARTICIPANT-TYPE TLV-LENGTH
+                         Name
+
+For every chatroom,
+::
+
+   Chatroom := CHATROOM-TYPE TLV-LENGTH
+                        TrustModel
+                        Participant+
+
+TrustModel:
+::
+
+   TrustModel := TRUST-MODEL-TYPE TLV-LENGTH
+                          nonNegativeInteger
+
+TrustModel value:
+
++------------------+----------------+
+| TrustModel       | Assigned value |
++==================+================+
+| Hierarchical     | 0              |
++------------------+----------------+
+| Web of Trust     | 1              |
++------------------+----------------+
+
+
+TLV assignments:
+
++------------------+----------------+----------------------+
+| Type             | Assigned value | Assigned value (hex) |
++==================+================+======================+
+| PARTICIPANT-TYPE |  128           |  0x80                |
++------------------+----------------+----------------------+
+| CHATROOM-TYPE    |  129           |  0x81                |
++------------------+----------------+----------------------+
+| TRUSTMODEL-TYPE  |  130           |  0x82                |
++------------------+----------------+----------------------+
+
+
+After receiving a data packet, the user puts the information of the chatroom into
+a list, while adding the data name into the interest's exclude
+filter and resend the interest. And he will receive more data
+packets with more chatrooms' information. When the interest expired
+which means that there is no other chatrooms, the list of chatrooms
+is completely generated.
+
+After a specific time(discovery time), the discovery process will
+start again. Discovery time can be set as a default time(e.g. 10
+minutes) or set by the user.
+
+
+Refreshing
++++++++++++++
+
+To check the existence of an old chatroom, the user sends an interest
+with the name:
+
+``/ndn/broadcast/chronochat/chatroom-list/chatroom-name``
+
+and with MustBeFresh selector to be true which is used querying for
+sync data. If a participant of the certain chatroom receives the
+interest, he will send a data packet with the chatroom's
+information. Otherwise, if the chatroom
+no longer exists, there is no data packet sent back. Therefore, if
+receiving a data packet, it means that the specific chatroom still
+exists. If the interest expires, the chatroom no longer exists.
+
+Security Consideration
+~~~~~~~~~~~~~~~~~~~~~~~
+
+In the design above, it is very hard for the user to identify if the
+information he received is true. Therefore, although the data
+packet can be verified by hierarchical trust model or web of trust,
+it can only verify that the data is the person from whom the user
+intends to receive. If the data-sending-person intends to send some
+fake content in the data packet, there is no way the user can tell if
+the content is correct or not. Therefore, an attacker can attack the
+design by sending a great number of fake chatroom information to the
+network which may cause waste in traffic and even prevent a user to
+receive real chatrooms' information. However, there are not many people
+having the motivation to do the attack. Therefore, some assumptions
+are made below.
+
+For every chatroom, it is assumed that every participant of a
+chatroom will not intend to send fake information to other users.
+
+Under this assumption, all the data can be verified by the public key
+generated according to the corresponding trust model and the security can
+be guaranteed in this level.
+
+Chatroom Discovery Protocol
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
++ **Timing**
+
+Two duration need to be defined when maintaining the list of
+chatroom:
+
+- **Refreshing time:** For every chatroom there is a refreshing time
+  which measures the frequency of checking the existence. The
+  refreshing time can be defined in the data packet as FreshnessPeriod
+  and after the time expired, the user will send a new interest to
+  check the existence of the specific chatroom. And finally, the
+  chatroom list changes according to the query result.
+
+- **Discovery time:** Discovery time is defined by the user. When
+    the timer expired, the user sends an interest to query for new
+    chatrooms.
+
++ **Chatroom List Data Structure**
+
+For every chatroom:
+
+::
+
+   enum TrustModel { TRUST_MODEL_HIERARCHICAL, TRUST_MODEL_WEBOFTRUST };
+
+   struct Chatroom
+   {
+     str::string name;
+     vector<Name> participants;
+     TrustModel trustModel;
+     vector<Name> contacts;
+   };
+
+
+The whole chatroom list:
+
+::
+
+   typedef vector<Chatroom> chatroomList;