How are web resources identified?

So HTTP is the rule book for communicating on the web.....but what about identifying, disovering and locating web servers in the first place? Only slightly more complex is the notion of addressing on the web.

Some analogies to think about: Again truck drivers and taxi-cabs use a form of identifying, disovering, and addressing that can be useful to help us understand the web. A cab driver might have a unique driver number and probably a driver name (a handle in CB speak). His or her car would have a unique cab number and be associated with a certain cab company (and that cab company too would have a unique company name and use a unique radio frequency). Likewise for a truckers, their rigs, and their transportation company. In fact this sort of scheme appears in lots of places in our society. Even, our postal addressing system isn't very different.

Well the web uses a few similar techniques. Let's start with web addresses. You've all used these and many of you will know them as URLs (uniform resource locators) as in:

Just as the name implies this system is intended to uniquely identify web sites and the resources (i.e. files) they manage.

One thing to remember about addressing on the web, though, is that all content fragments (a picture, a post, a song, a program, a video) can be univerally identified separately - not just by the web page that presents them. So a web page is better understood as a collection of many identifiable objects. And what's more is that a URL can serve as an address for all kinds of different communication protocols (email, file transfer, http, and so on).

The long and short of this is that the web is well understood as a vast collection of micro or atomic resources that can be pulled together by anyone in any fashion using various presentation techniques. This design is why the web is so efficient and universal.

Is there a resemblance here to the way nature makes universal order out of chaos? to matter?

Though it's beyond the scope of this course URL's are really a nuance of a broader concept called URI's. Here's a peak

There is a somewhat related concept that the Internet uses to make the physical connection between computers but we already have more than enough to go on for our purposes.