Chapter 18 is new to The Internet for Commodore C64/128 Users 3rd Edition. Posting it here allows those who have purchased the first and second editions the opportunity to read this chapter.

It also shows the style that The Internet for Commodore C64/128 Users is written in, and demonstrates (as closely as possible) the formatting used throughout the book. This includes bolding, references to other chapters as required and other similar touches.

All rights reserved. Reproduction in whole or in part in any form or medium without express written permission of VideoCam Services is prohibited.

The majority of this book covers Internet use through a terminal program and UNIX shell account. TCP/IP connections let you hook your computer directly to the Internet as an Internet computer. Since many Internet providers only offer direct access to the Internet and are disallowing shell accounts, we need this type of connection. Another reason we need this type of access is that it'll give us more variety in the ways we can use our Commodores on the Net. At the time of this writing (July, 1998), there are only a handful of Commodore programs that allow us to place our computers directly on the Net, and once connected, there isn't much we can do online - yet. We've witnessed many "firsts" for the Commodore, and one thing we can be certain of is you can't keep a good programmer away from a challenge.

This chapter will discuss directly connecting your computer to the Internet, what it will involve, and the things we'll need to know when the software becomes available. You should be aware that it's entirely possible for you to have a shell account and a TCP/IP connection with the same Internet provider (if they offer both), and you can use both, with the use of telnet (covered in Chapter 10).

TCP/IP

As briefly touched on in Chapter 6, the Internet is made up of many networks, and there is an inter-connectivity between many of the different Internet features - such as when we send Email while viewing a Web page. For two computers to talk to each other a common means of communication is required. For instance, to use our shell accounts, we must use VT100 or ANSI terminal emulation, and this must match what the Internet provider is using. If we try to use a different terminal emulation than what the other computer expects, they won't talk to each other correctly. Another example is when we transfer files using Xmodem, Ymodem or Zmodem. If we try to use a protocol different from what the other computer is using, the file transfer will fail. Both computers need to be 'speaking in the same language'.

As the Internet is many computers connected with each other, a common set of protocols is required to allow communication between many differing types of computers. TCP (Transmission Control Protocol) and IP (Internet Protocol) are the two most fundamental protocols of the set, and are referred to as TCP/IP. These two protocols work together and allow computers to be networked with each other.

The TCP/IP protocols take information and send it to another computer on the network in small pieces called packets. The packets arrive at the other computer and are pieced back together again in proper order. The packets can arrive out of sequence, and some travel different routes to arrive at the same destination. It reminds me a little of when someone arrives on the transporter pad in a Star Trek episode.

As users, we really don't need to know how TCP/IP works any more than we need to understand how a microwave warms up pizza. It's based on very simple principles, but it's very difficult to explain.

What's a TCP/IP Stack?

The protocols that make up a TCP/IP suite are referred to as a "stack", because they work in stacked layers, one protocol calling an adjacent protocol. To continue the Star Trek analogy, the TCP/IP Stack would equate to the transporter room control panel. In Commodore use and experience, a TCP/IP Stack is a program that controls the flow of data across a network.

Why is a standard TCP/IP Stack for the Commodore Important?

There are many different ways to write a program, as witnessed by the many different word processors that we have for the Commodore. How things are implemented differs depending on the author. We need a standard TCP/IP stack that other programmers can make use of in writing their contributions. One programmer could probably write many different types of client or server programs (described below) for use on the Internet, but to ask one programmer to carry the full load of creating, testing, and perfecting all the utilities we want to use on the Net isn't fair to them, and will never do justice to the suite of utilities as a whole. A team approach and a common stack program will allow programmers to work on areas they're interested in and will give way to a variety of utilities.

Without a common TCP/IP stack that can be built upon, we could end up with many separate Internet clients that will only work with certain "TCP/IP stacks".

SLIP and PPP Connections

When we dial our service provider, TCP/IP needs an additional protocol to handle the task of communicating via modem to another TCP/IP computer. SLIP (Serial Line Internet Protocol), or PPP (Point-to-Point Protocol) are the two connection protocols that are used to hook up your computer to another computer on the Internet via a phone line, and we use either one or the other, not both. Whether we will ever have a PPP type connection is up in the air, but SLIP has been implemented by a few different Commodore programs, and we will likely see SLIP access first. From the perspective of the user, once online you wouldn't be able to tell the difference between a SLIP or PPP type of connection as each work basically the same, with a few differences in the initial set-up.

SLIP (or PPP) Set Up

In Chapter 4, we talked about a few items we would need to decide on before calling the Internet Provider, such as a UserID and Password. To use SLIP or PPP, you will still need these (or similar), but you'll also need to contact the provider to arrange your account and find out a few things the connection program may need in advance.

SLIP (and/or PPP) Name

Some Internet providers use the name given during the login procedure to determine what type of connection is required. For instance, my Internet provider uses the letters S or P (upper case) before the user name to determine whether to initialise a SLIP or PPP connection. The user name can also vary from the UserID used with your shell account.

SLIP (and/or PPP) Password

This password may be the same as what you use for your shell account (if you have one with the provider), or it may be totally different, depending on the provider.

IP Address

Every computer on the Internet has a unique address assigned to it that consists of a series of numbers separated by dots, known as the IP address (an example IP address would be: 123.123.123.123) . This address is used to identify your computer. Internet providers usually have a range of IP numbers for their customers to use, and assign them in two different ways. The first method is with a static (or permanent) address that you use whenever you access the provider, or they may allocate a different number each time your computer connects to it.

Daniel Dallmann's SLIP Demo requires that you enter the IP number each time you use the program, but other programs may deal with this in other ways, such as in the form of a menu that you fill out before getting online (if you have a static address). The program may even save this information, similar to how you save phone book entries in a term program. Another method may be with script programs for various Internet providers, which would work for those systems that use a different number each time you connect, which may work best for Internet providers that use a different IP address each time.

Dynamic Name System

To make it easier for you, and for any Internet utilities that you use, the connection program may ask you to provide the IP address for your Internet provider's Dynamic Name System, or DNS. The Dynamic Name System does exactly what it's title implies, allowing us to type words instead of numbers. For instance, the IP address for videocam.net.au is 203.23.139.10 - without access to a DNS Server, to reach our web site, you would have to type: "http://203.23.139.10/".

Some Internet providers have more than one name server, and refer to one as the primary DNS, and the other as the secondary. When in doubt, if there's only one choice within the program, type in the primary DNS number.

Web Proxy

This is usually something along the lines of your Internet providers name, and port 8080, such as "any.isp.com, port 8080". This item won't affect anything relating to logging in to use your Internet provider, but it's information you may be given when you sign up. When a Commodore web browser becomes available, it may use the proxy to access popular web pages and graphics that the Internet provider stores in a cache for users of their system. This means web pages load faster because the data is retrieved from the local host computer rather than across the web. Shell account users can use this information when configuring Lynx (Chapter 16), too.

Connecting Your C= to the Internet

Instead of guessing about what future programs will use, I'll make use of Novaterm 9.6 and Daniel Dallmann's SLIP Demo to demonstrate the sequence of events when we first use a SLIP account. The demo program requires a 9600 baud or faster modem (it won't work at lower speeds), and will make use of the Turbo-232 or SwiftLink Cartridge port interface or the Enhanced 9600bps User port interface.

1. The SLIP demo doesn't have a built in phone dialer, so we need to use Novaterm for this. Use VT100 or ANSI emulation, and have the modem speed set for 9600bps. Dial the telephone number for your Internet provider.

2. Once connected, the provider will request your user name and then your password. Remember to use the format for your user name that the provider has instructed you to use for SLIP connections.

3. A. If your Internet provider uses a static IP address system, the SLIP connection will automatically be established (usually you see text on the screen to tell you it's starting the connection).
   

   B. If the Internet provider uses a dynamic IP address system, your IP address will appear on the screen. Write it down.

4. Without resetting your computer, exit Novaterm (use the menu option labeled "Exit Terminal"), and boot the SLIP demo.

5. The Demo will ask for your IP address. It has an easy to understand user interface for typing this in. It also asks for the IP address of the IRC server, but this is already filled in with a working example for you to use, so you can just press return at this option.

6. A menu will appear on screen:
   
   1. telnet
      
   2. passive telnet (telnetd)
      
   3. IRC
      
   4. TCP/IP setup
      
   5. TCP/IP statistics

The first three items are client applications built into the demonstration. "TCP/IP setup" offers the same questions as when the program was booted. The last option tells you the number of packets that have been received and sent.

Clients and Servers

When we access the Internet with a shell account, we make use of the programs that the Internet provider has made available for us to use. This "luxury" disappears when we connect our computer to the Internet directly. Every Internet 'utility' discussed in this book is a client application, which has a corresponding server. The server accepts and acts on the commands sent by the client program. For each activity we want to do on the Internet, we will need a client application available.

For those who enjoy an interesting challenge, the documentation for the SLIP demo gives some interesting examples of how you can use the telnet client to read and send Email or browse a web page. This isn't as easy as it sounds, because you must access the server programs directly and the mail and web pages stream across the screen without any paging, or saving the data to your disk drive. This gives a really good example of the commands that our future client applications will use to send and retrieve data for us.

What client application programs do we need?

The list below are some of the client applications we'll need once we have SLIP or PPP access, along with the chapter they correspond to. One of the benefits of understanding how Internet features work with a shell account is that their basic function doesn't change, and much of the information we've already learned can be put into place when using client applications that do the same (or similar) job.

Email             Chapter 7
Newsgroup Reader  Chapter 9
Telnet            Chapter 10
Ping              Chapter 10
FTP               Chapter 11
Web Browser       Chapter 12
IRC               Chapter 14

With time, we will probably see many utility type applications as well, such as programs designed to do a specific Internet related task for us, like one that converts our C= text files to ASCII and uploads them to our Web directory, or BBS related utilities so C= sysops can transfer BBS messages easily. Time will tell.

Will I need new hardware?

With a TCP/IP connection to the Internet, our computer will need to be able to handle the job of running the TCP/IP stack plus the client or server applications, PLUS any data that these programs retrieve or generate for us. Instead of reading our Email online, it will be retrieved for us and saved to disk. A web browser program (graphic or text based) will need to store the web pages it retrieves. A graphic Web browser will need even more disk space.

It's hard to speculate on what will be required without the software to go with it, but a UART Cartridge Port Interface would probably top the list, as many of the programmers working on TCP/IP projects are working exclusively with the routines that work with these, and the interfaces allow the use of higher speed modems. A graphical web browser may need a faster computer to display GIF files at a faster pace. This would mean a SuperCPU from CMD may be required for such a program. Another possible requirement is an REU, as any applications written to use Maurice Randall's Wheels (an update to GEOS v2) will have to work within the constraints of the operating system, and Wheels requires an REU. At the time of this writing, there are no known Wheels TCP/IP programs in, but the possibilities of a GEOS based TCP/IP stack has been speculated about for years.

What's available right now?

Daniel Dallmann's SLIP Demo
http://hld.c64.org/poldi/lunix/lunix.html* - requires either a UART cartridge interface or his 9600 user port interface (the schematics can be found on his Web site). Daniel can be contacted via Email at: dallmann@heilbronn.netsurf.de

LUnix (Little Unix) and LNG (LUnix Next Generation)
http://www.heilbronn.netsurf.de/~dallmann/lunix/lng.html* - LUnix is a Commodore UNIX operating system written by Daniel Dallmann and other programmers. It has a working TCP/IP stack but at the time of this writing, not many client applications that are in common use. It's generally best suited for experienced users, and has the same requirements as the SLIP Demo above.

*The URL in the book is listed as http://rpool1.rus.uni-stuttgart.de/~etk10217/c64.html. This address is updated.

What's in the works?

A few things are in development, but I can't write about "vapour ware" (something promised but never delivered), and it's human nature that when we hear of something, we tend to try to send encouragement to the people working on them, and in some cases, this has turned into rumours and speculation that's slowed development of the software as the programmers spend time and effort debunking the rumours.

How will you know that a standard TCP/IP connection package has been developed for the Commodore? Good news travels through our Commodore communities like wildfire. Information about it will be available from the same places where you learned about this book, such as Commodore World, Loadstar, User Group newsletters, on Bulletin Boards and on the Internet too, such as in the TIFCU mail list or comp.sys.cbm newsgroup. It is truly hoped that this chapter is made "out of date" as soon as it's published as many Commodore users will be happy to connect their C64/128's directly to the Internet.

Summary:

In this chapter we have learned:

• TCP/IP connections let you hook your computer directly to the Internet as an Internet computer.
  
• TCP/IP are two protocols that allow Internet computers to communicate with each other.
  
• What a TCP/IP Stack is, and why we need one.
  
• That SLIP and PPP are protocols for connecting to the Internet.
  
• Things we will need to know when setting up to use a SLIP or PPP connection.
  
• What clients and servers are, and why we need them.
  
• About hardware that may be required for a SLIP or PPP connection.
  
• What's available now and where to find out about new Commodore TCP/IP software.