WHAT'S THE BIG DEAL ABOUT TCP/IP? If you're new to amateur packet radio, you may think checking into packet bulletin board systems (PBBSs), keyboard-to-keyboard ragchewing and hanging around the local DX PacketCluster is about all there is to this facet of the hobby. You may have seen or heard the term "TCP/IP" here and there, but it sounds like some goony computer programmer blabber that you'd never be able to understand. Here's where that changes! As a packeteer, you may sometimes wish the equipment you've put together to operate your packet station could do more than read and write a bunch of messages. How about being able to transfer complete files--ASCII text, executable programs (.EXE and .COM) and even compressed files (.ZIP, .ARJ, .ARC)? How about sending mail directly from your computer to someone else's without either of you having to check into a PBBS on a congested channel? How about being able to log onto another hams computer, poke around in a few subdirectories, and upload or download interesting data? HOW ABOUT BEING ABLE TO DO ALL OF THESE AT THE SAME TIME??! And simultaneously hold a keyboard-to-keyboard chat? And operate your own private PBBS? And get in on the networking at 9600 bauds on 440 without first being designated an "official" full-service PBBS? That's the kind of stuff you can do with TCP/IP! So how does it work? It's not extremely simple to explain in a message like this, but the necessary elements of a TCP/IP station include any TNC that can be operated in KISS mode, a VHF or UHF FM transceiver (and antenna), a personal computer (PC) and an appropriate software package. With TCP/IP, you turn off the limited semi-intelligent internal "brain" in your TNC and let your PC do the "thinking." Because your PC is much faster and smarter than the TNC--and can hold lots more information on its disk drives--just imagine how much more you can do on packet when your PC is the core of your packet setup! As mentioned in the previous part of this text, TCP/IP (which stands for Transmission Control Protocol/Internet Protocol) doesn't require any special equipment to try. All you need is the proper software. There are plenty of programs available to run TCP/IP on your PC. Assuming you have an IBM- compatible MS-DOS machine, you can use programs called NET, NOS and variations thereof. An interesting aspect of TCP/IP is that the program isn't "hardwired" on your TNC's EPROM chips--because it's disk-based, you can change it any time. In fact, the TCP/IP source code is publicly distributed, so anyone who's a good C programmer can (and is encouraged to) experiment with adding new code and tweaking the basic program. New features are added all the time by the various people who are good at that sort of stuff. And so, there are lots of "versions" of TCP/IP software floating around. You can select from a neat variety of packages...and if you get one you don't like, erase the disk and run a new one! There's another compelling reason to "switch over" from plain AX.25 packet (like you're using now) and operate TCP/IP packet... Now you know that the world of TCP/IP packet radio operating is a much more flexible, useful way to put your TNC and radio to good use. But here's one other point that may convince you to give it a try. Almost every packet operator has been frustrated by the constant hassle of channel congestion that makes you wait and wait and wait for your packets to get through...and often result in a "retry out" condition where the connection drops and you're dead in the water. NetROM and TheNet nodes help a bit, but it's still terribly clumsy. TCP/IP takes most of the waiting out of packet. To explain it simply, the software all stations use contains a built in technique by which all stations continuously adjust their retry rates based on how quickly their packets are acknowledged. If the channel is busy, everybody slows down a notch until the packets are getting through. As the channel frees up, each station transmits more often until the upper limit is reached. Your PC monitors the situation all the time and the retry rate is always optimized. So most of the time, you get your packets through more efficiently and you virtually never "retry out." One more note for people who "hate to wait": While you're sending out a pile of messages, downloading a couple of files and uploading a few others, you can also be holding keyboard-to-keyboard ragchews and check into PBBSs! The TCP/IP software features built-in "multitasking," so you can do more than just sit and wait while you're transferring a file or replying to your mail. To receive personal mail and read general bulletins, you check into a local packet bulletin board system (PBBS). You may also have a personal mailbox built into your TNC's firmware. To send or receive mail, you log onto the other station, tell it you want to send a message, address it to a call sign and that station's "home" PBBS (or to 4SALE@NEBBS or whatever) and then type your text. From there, it's forwarded to the other end where, you hope, the addressee happens to log on and read the message. To read your mail, you log onto a PBBS, list a bazillion messages (usually) and read the ones that appeal to you. You can still do all that with TCP/IP if you like, but the best part is that you don't have to! In TCP/IP, every station is assigned its very own unique identifier (called an IP Address). This is a numerical string that tells every other TCP/IP station how to find you. So, if I want to send a message to you, I type it up on my computer, edit it, maybe import a text file as prt of it, or whatever, then tell my PC to send it on its way. Within a few minutes the message is transmitted to the nearby relay "hub" (or "switch," in TCP/IP jargon) and sent TO YOUR PC. When it's safely stored on your disk, my PC deletes it from the outgoing mail queue automatically and I know you got it! You don't have to log onto a PBBS or go node hopping to find me--it's on your machine and you can read it, save it, reply to it or whatever whenever you like. If TCP/IP is so great, why aren't we all using it? There are a couple of minor limitations to it: First of all, it requires you to have your PC booted up and running at least a few hours a day (ideally it should be up all the time, although that's not strictly necessary). Some people object to this, however, most PCs draw minimal electricity when they're just idling and it usually does less harm to components to leave them running than to subject them to the surges of power caused by switching them on and off several times a day. Another thing is that the TCP/IP network isn't yet fully integrated with the "normal" AX.25 packet network in many areas. You may still have to switch to the local PBBS frequency and log on if you want to read the generic bulletins you're used to seeing. (In many places, gateway stations are able to forward general-interest bulletins to hams who want them and you're can use your TCP/IP software to connect to a "plain" AX.25 station anytime. A final hitch: Most of the existing documentation for TCP/IP software and operation is sort of technical and confusing (although we're getting a fairly simple "manual" together). Unfortunately, computer programmers aren't always the best documentation authors, so newcomers can find the "directions" a bit incoherent! Fortunately, there are plenty of us who speak English and will be happy to help. To try out TCP/IP, it's easy to get free software... If you think you'd like to give TCP/IP a try, contact the New England TCP Association for information: If you're ready to try it now, send a blank, MS-DOS-formatted diskette (any size) with a self-addressed, stamped return mailer, to: Brian Battles, WS1O Jerry Walker, N1HAT 77 Arbutus St or 49 Inverness Ln Middletown, CT 06457 Middletown, CT 06457 and we'll make you a free copy of the simplest TCP/IP program and some documentation that tells you how to get started.