Web Server (HTTP 1.0)

In the OSI reference model, there is a thing called an Application layer which allows computer processes to comunicate with each other.

Of course, there is more to it, I mean, there are 7 layers that act on behalf of the user to inter comunicate (In the networking sense) through different chanels (The Physical layer): copper wires, fiber optic, wireless, etc.

This abstraction allows to build very robust services on top of very simple, but complete, protocols.

Ok, HTTP is simple, but it is only simple due to the fact of the ingeniosness (may not be a word) of it's creator, Tim Berners Lee. And when I say that it is simple, I mean to implement.

The server that I've written, only implements the HTTP GET request and allows for directory listings. It follows the version HTTP 1.0, in the sense that it makes no use of something they call "pipelining." No pipelining means that every object requested uses one connection. With pipelining, a page composed of many objects may use only one connection to download the whole web page (Great for performance, TCP is rather expensive in regards to connection settings and such).

This server makes use of Java threads, so that each request is served on its own. If you are interested, the source code is here.

BTW, the J in the file name, JWebServer.java, stands for Jose, not Java (TM).