There’s a brief segment of the Today show from 1994 which has recently gone somewhat viral.  Bryant Gumbel, Katie Couric and Elizabeth Vargas are doing a segment on Violence in America and they briefly get sidetracked by a discussion of the internet.  They apparently presented an email address for viewers to contact about the subject, but none of them knew what it meant.  In fact, they weren’t certain what the now ubiquitous “@” symbol was.  Katie thought it meant “about.”  Somewhere during the discussion, the question is asked, “What is internet, anyway?”

One of the crew for the show gave a decent answer to the question, stating it was a number of computers connected together.  The Today hosts grasped this, but couldn’t understand what you would do with it.  Bryant asked if you could send it mail, and Elizabeth stated you didn’t need a phone line to connect to it.  This last statement was mostly false in 1994.  Sure, universities and government installations had “broadband” connections, but the average Joe had to use a dial-up connection.  I put broadband in quotes there because connection speeds then were pretty low, nowhere near the broadband speeds of today.

The internet, or “interwebs” as some places jokingly call it, started in the late 1950’s.  The launch of Sputnik in 1957 ignited the space race between the U.S. and Russia, and with it a demand for a better ground-based computer network.  A government group called DARPA was formed and this network was one of their first objectives.  I will skip a bit of the history that followed; suffice to say they went through a few generations before the first two nodes of ARPANET were connected in 1969.  Within two years later, the fledgling network had fifteen nodes connected together.

In 1974, the word “internet” was used for the first time to describe a network connected using the TCP/IP protocol used today.  ARPANET switched over to TCP/IP in 1983.  This protocol was later used for NSFNET (National Science Foundation Network, used to connect universities)  This network soon connected to commercial networks.  The flexibility of the TCP/IP protocol meant the network could grow fast, and it did.  The World Wide Web project was started in 1989 and, well, that went everywhere fast.

So that’s a little history, but what does it mean?  What is TCP/IP?  First, it stands for Transmission Control Protocol / Internet Protocol.  It is a definition of how two or more computer systems will talk to each other.  Not the language they will speak, exactly, it’s more basic than that.  When two people meet, they have a few ways they can communicate.  They can speak if they both know a common language, they can use hand signals, they can write notes, etc.  This is what TCP/IP is about.  It defines how packets–little parcels of data–will be exchanged between two systems.  If I hand you a written note but you don’t have a clue what writing is, communication fails.  If you give me the finger, but I don’t even know people can communicate with hand gestures, again communication fails.

The TCP/IP protocol is independent of what is being communicated.  It simply allows the data to move along.  If I pass you a note written in English but you don’t know the language, communication is still going to fail, but if you understand the concept of written language, you can at least recognize the attempt to communicate.  The receiving system may not know what to do with the data handed to it in a TCP/IP communication, but it can at least acknowledge reception.

This is where the world wide web steps in.  The WWW is another protocol, one more akin to individual languages, although not quite.  To put it in the human communication analogy, it would be like using English to talk about math but German to discuss physics and French to chat about cooking.  The TCP/IP protocol can be used for a wide variety of kinds of information to be passed along.  In the early days of the internet, quite a few were used.  These protocols are still around, but many are not in common use.  Mostly we use the HyperText Transfer Protocol (HTTP), File Transfer Protocol (FTP) and Simple Mail Transfer Protocol (SMTP).  There are others in use today, but those are the most popular.  Each is used for a different purpose–websites, transferring files and sending email.  Each relies on TCP/IP to work.

Every website address is translated into an IP address.  My blog, for example, is not really http://blog.shadowkatmandu.net, it’s really 74.208.89.232.  You can put that into a browser address bar instead of blog.shadowkatmandu.net and it should work just as well if not better.  This brings us to the problem with the TCP/IP protocol.  Its final implementation on the internet involved using an address composed of four numbers called octets.  Every device on the internet has one of these devices.  The problem is there’s a lot of devices on the internet asking for an IP address, a lot more than what the original implementation was designed for.  There’s a new implementation, but it’s not being implemented fast enough.

This is the internet, then.  An immense collection of computers connected to one another.  They communicate using TCP/IP in conjunction with some other protocol.  Once they are in communication with each other, they pass data back and forth.  If necessary, the receiving end may interpret the data using, for example, a web browser like Firefox or Internet Explorer or Safari, etc.  I’m sorry I couldn’t answer the Today crew’s question in 1994, but hopefully I have provided some enlightenment for today.  See you tomorrow.