Ok, forgive my ignorance here, but:
In the old days (early 1990’s), I used to dial up by modem into a database for various purposes. It would take repeated tries to finally establish a connection due to a very limited number of “ports” (for lack of a better word). 90% of the time, you could count on getting a busy signal.
Ok, so what quantum leap in technology allows us to have the internet? Or, did it just magically become economical for every company to set up a huge server with an abundance of “ports”…just so you can have a website to visit? Maybe my image is incorrect.
Please explain how we went from an electronic backwoods country road to an information superhighway? (Keep it simple.)
I think the basic answer is yes, it just magically became a lot cheaper for companies to set up an abundance of “ports.” A couple of factors entered into this. The first is Moore’s law (not actually a law), which says that the number of circuits that can be put on a silicon chip will double every 18 months or so, therefore doubling the speed/power of computers at the same rate (more or less). This has allowed systems such as modems (ports) to be simplified and put on fewer chips, costing less money.
The second thing is that as more people purchased computer equipment to go online, the proudction runs of the equipment got longer, and economies of scale increased so, like most electronic equipment, the price went way down.
Along with deregulation, cellular phones, and faxes, additional dial-up ports are one reason for the proliferation of new area codes as phone lines get used up.
The Internet is basically a computer network. A friggin massive one but a network of computers is all. The concept of networked computing, supposedly dreamt up by bods at Arpanet, was, in the 60s thought to be a “Good Thing”. Skip forward 10 years or so and Universities are on the scene, using educational networks for academic purposes. Hence networks sprung up all over the place wherever there was funding. No-one had yet recognised that there was a need for these networks to talk to one another and they were, for the most part, incompatible. It was the introduction of protocols and standards which made wide area networking and, crucially the Internet, possible. Now, back to your original point…the reason your dial-up connection to work sucked is nothing to do with the telecommunications architecture being particularly bad in 1990, the simple fact is that this is simply a connection between two computers, the computing hardware probably being the problem. The short answer is: the Internet was made possible by the widespread adoption of protocols and standards and the prevolence of distributed network computing. Of course I could be talking rubbish…anyone?
More to the point, in the old days each user would tie up a physical asset on your computer - you want 100 users connected, you need 100 modems connected to 100 serial ports. 99 of those people may be sitting there reading a page you just displayed for them, so the port and all that hardware is sitting idle, but no one else can use it.
The internet uses a packet protocol that allows resources to be shared. One physical piece of hardware can be accessed by hundreds, thousands, or millions of people, with the software sorting out who is who.
Sam’s explanation is essentially correct. Part of your confusion may also stem from the usage of the term “port”.
In your dail up connotation, the concept of port is physical. Port = Modem or other connectivity device.
You do, however, continue to encounter the term in reference to the Internet, but the meaning is different. A “port” is part of the IP addressing scheme, and is used as a reference for the protocol being transmitted. As an example, FTP normaly uses Port 21, SMTP 125, Telnet 23, etc. However, unlike the physical modem “ports”, these are parts of the logical addressing scheme. When you connect to an ftp server, you aren’t actually tying (damn can’t spell today), up the port, the use of port 21 simply insures the packet gets to the right application.