There is a very small amount of networking background required for Project 2. In Project 2, we want our computer (or our "host") to request information over the Internet from a remote webserver (another Internet "host"). On our computer, from within our Project 2 ap-plication, we will open a "socket", which on many operating systems acts like a ?le which you can read or write. Moving down the network stack into the operating system, a socket uses a transport protocol to communicate with the remote host. The application-layer HTTP protocol which webservers use is built on top of the transport-layer TCP protocol, which provides reliable, in-order, stream-based delivery between your socket and the remote socket. A socket also has a "port" number associated with it. When we open our socket, we will request that it opens port 80 on the remote host, since this is the port number on which almost all webservers listen for re-quests. Finally, when opening the socket, we need to specify which remote webserver we want to communicate with, such as www.cnn.com, etc. Moving down one more layer, every host (laptop, server, cell phone, etc) which is directly accessible on the Internet has a unique network-layer IP address, such as 128.119.240.19, and most of those have hostnames, such as www.cs.umass.edu; the Domain Name Service (DNS) translates from hostnames to IP addresses.
Note that the sock.read() method is in a while-loop, since reading from a socket will return between one byte and the maximum number of bytes requested ((MAX READ-1-size) in the code above). You'll need to keep repeatedly reading from the socket again until you get all the bytes you want. sock.read() will block until it has at least one byte to return, unless its request times out (set by sock.setTimeout(5) to be 5 seconds), or unless the connection is closed.