limitation that messages can’t uses the reserved values for other purposes. If the messages contain text characters which can include characters for numerals , network addresses can use any values that don’t represent characters. For example, if the messages use only US ASCII codes 0—127 , addresses can use the values 128—255. If the messages contain binary data, one solution is to send the data in ASCII Hex format as described in Chapter 2. Because ASCII Hex can represent any binary value using just 16 codes, plenty of codes remain for use as addresses. Defining a Message Format Many network protocols define a message format with the address and other information in assigned locations in the message. For example, an 8-byte message might consist of an address byte followed by seven message bytes. On receiving a byte, a node examines the value to see if it matches the node’s address. If it’s a match, the node reads and acts on the seven bytes that follow, then waits for another address byte to examine. If the address doesn’t match, the node counts but otherwise ignores the seven bytes that follow. With this method, every node has to detect every byte sent, if only to know when the message is finished. A node that misses a byte for any reason won’t detect the correct address bytes in future messages. Using Dedicated Codes for Start and End To make it easier to detect addresses, a protocol can define values that indicate Start of Transmission and End of Transmission. A node that gets lost can then recover on the next Start of Transmission code. Conventional values are 02h Control+B for Start of Transmission and 03h Control+C for End of Transmission. Some networks use other characters such as “:” or “$”. These values are then unavailable for other uses, so using dedicated codes is useful mainly for communications that send data as plain text. 273 Chapter 12 A message format can also define a field that indicates the length of the data that follows, and receiving n prom-electric.ru