Please allow me to introduce myself. There is only one of me, so I guess I'm unique. My parents gave me the name “Mark”. Most of the time I'm the only “Mark” around. If you say my name, I will think you are referring to me. But what if there is more than one “Mark” in the room? Easy. I just introduce myself by my full name “Mark Leicester”. So far so good, but the world is a big place and I'm not the only Mark Leicester in the world. What happens when I have to register with the tax department? The tax officers will ask me for all sorts of extra information to uniquely identify me: my middle name, my date of birth, etc. To avoid any confusion, possibly resulting in someone else paying my tax, they will issue me with an identifier. From then on I am just a number.
When it comes to the world wide web, your browser works a little bit like the tax department. Your browser identifies every page, image or movie by a single identifier. The Internet equivalent of your tax number is the URI1, or Universal Resource Identifier. You've seen them, probably even memorised a few. The URI for my blog looks like this: http://www.atomicmaestro.com/. You might have heard these strings of characters described as URLs; for our purposes they're the same thing.
So what exactly is a URI? A URI is a resource in name only. My tax number identifies me uniquely, but says nothing about where I live, what I look like or how old I am. A URI identifies a resource, but doesn't tell your browser how to find it, what the resource actually is, or how old it is. You may be thinking that a URI often contains useful information, maybe a company's name or a brief description of a page's content. Don't be fooled2. These niceties are for humans only, a browser doesn't care. A browser just sees a sequence of letters or numbers. Thus, an eagerly typed URI such as http://www.coolnewstuff.com/whatsonnow may result in a document that is nothing of the sort. There are no rules to prevent a content provider using this URI to identify a page that hasn't been updated since October 1996!
A URI guarantees nothing. Why is this? Why can't a URI be more like a map reference that leads us to the resource? Why can't the lords of the Internet pass a law that demands a URI tell us something about a resource? The reason is simple. Resources change, and resources move. Imagine if a URI guaranteed something about the content of a document. You would have to change the URI every time you changed the document! Imaginge if the URI told the browser where to find a document. You could only ever have one copy of the document! Too bad if your server crashed and had to be replaced by a new one! The URI is simply a contract between a content provider and the browser. No one trusts a content provider who breaks this contract regularly.
What's the upshot of all this? Here a couple of ideas:
- Making a URI public is like signing a contract. Remember that Google crawlers will publish your URIs without asking you first. If we take down the page we are breaking our contract.
- For any URI, keep the content it identifies true to the original intention. If you choose http://www.coolnewstuff.com/whatsonnow to identify a page showing current events, stick to it!If we re-purpose a URI we are bending the rules of our contract.
1While URIs are part of the magic of the Internet, they are not to be confused with the spoon-bending paranormalist of the same name.
2Seriously. Please don't be fooled. The human-readable aspect of a URI is being exploited by phishing scams. They rely on you ignoring the difference between “www.mybankswebsite.com” and “www.mybankswebsile.com”. The scamming site looks exactly like the real one. It asks you for your credit card number. You are scammed. Be wary.
