Diving into the Internet by Joel Snyder for Internet World Mar/April 1994 [ With this issue, Internet World welcomes another columnist to its roster: Joel M. Snyder. Joel's both a writer and a doer; when he's not working on networking Russia or X.25 standards in Switzerland, he's at home in Tucson, Arizona, writing all about TCP/IP, Macintoshes, DEC, and the Internet. - dpd ] ---------- Diving into the Internet: The Trouble With Gopher by Joel Snyder "The Internet is the librarian's full employment act of the 1990s" - Daniel Dern Ever hear the expression "Better late than never?" When people say that to me, I have an answer: "who says?" I'm a believer in doing things right. When it's time to help fix a half-done job, that bothers me. So when I started to explore the convoluted and confusing world of Gopherspace, I got excited and upset. Gopher, for those of you who have been living in a cave for the past two years, is the brainchild of a group of support folks at the University of Minnesota. Their original design goals were modest: replace the cumbersome anonymous FTP system for information retrieval with something that was friendly, understandable, and usable. Build a campus-wide information system (CWIS) to help distribute information. That's the exciting part. At the University of Minnesota, Gopher was an incredible success. People could actually find information about things that mattered to them. More importantly, they could find information without knowing anything about the conventions of the Internet, Unix, or any other esoteric bit of computer arcana. That's the exciting part. Gopher's ideas aren't new. CompuServe, the grand dame of information services, has had a menu-oriented view of its on-line world for a decade. And the most powerful concept of Gopher--- its ability to jump from one Gopher server to another without showing the user any seams in the web of information resources--- has been part of CompuServe from the beginning. What's lacking in Gopherspace is something that CompuServe has: an information architect. Someone who's responsible for making sure that it all makes sense. That there are overall indices. That users can jump into the system with some hope that they'll find resources waiting to be discovered. That the user isn't plunged into total anarchy. Unfortunately, that's what Gopherspace is today. Anybody can install and advertise a Gopher. And, you'll discover, lots of geniuses and morons have. Can we blame the good folks at Minnesota for that? For making Gopher too easy to install and use? For making it so simple to link the Gophers of the world together? I don't think so. Think of the Gopher pioneers as true research scientists. The results of the Minnesota Gopher team are something that can be used or abused. Gopher is a tool. Using tools like Gopher to build information systems takes practice. Right now, there are a lot of amateurs running Gopher servers. That's a problem. Look at the typical new Internet user, someone who got lured into it and uses one of the dozen new "Internet User Guides" as a bible. These aren't computer folks. They're confused. How to find anything in the million-plus nodes of the network? Suddenly, they find Gopher. There it is---a menu-oriented interface to the Internet. Finally. While drowning in a sea of information and resources, Gopher offers an anchor. Try and take Gopher away from a new user, and you'll see how important any anchor, even a bad one, is. Gopherspace isn't organized. It's assembled. Gopher is a true client server application. The software you run when you type "gopher" (or click on the Gopher icon, for you GUI folks) doesn't have any data. It points at a server, one of hundreds, maybe thousands, on the Internet. You may have to pick one out yourself, or your system administrator may have selected one for you as a default. The top level menu at that first server forms your window on the Internet. If the maintainer of the server you pointed your client at has been diligent and thoughtful, you may have a well designed view of some small chunk of the Internet. Or, you may be pointing at an archive that was up-to-date when it was created, but hasn't been maintained since. Most likely, though, you'll find yourself pointing at a server which holds a small repository of local information and a menu item pointing at "All Gophers in the World." That menu item is one place where your server ends and another begins. And it's the most common way for someone to link their Gopher server into the wild world of Gopherspace. Certain Gophers specialize in collecting pointers to a particular kind of information, such as legal information or religion. How can you find the Gopher maintained by someone with your same interests? Good luck. "All Gophers in the World" is helpfully organized by the physical location of the server: states, countries, and continents. That's the same as sorting books in a library by the publisher. And just as useful. Some Gopher enthusiasts have jumped in to fight the inevitable entropic disintegration that the second law of thermodynamics predicts. David Riggins, for example, has created a Gopher server list organized by subject. His "Gopher Jewels" list is available via Gopher to cwis.usc.edu (or anonymous ftp to usc.edu or ftp.einet.net). Other Gopher server operators (Gopher masters) have made similar efforts. At the University of Saskatchewan, gopher.usask.ca is a collection of well-organized pointers to the world of Gopher space. University of Saskatchewan is also the home of Hytelnet, the world's most ambitious (and complete!) collection of telnet-able Internet resources. If you like Gopher, you should also get a copy of Hytelnet. At Rice University, Prentiss Riddle tries to keep on top of things with a Gopher that merges directories from other Gophers. Sort of a Gopher of Gophers of Gophers, all kept at riceinfo.rice.edu. You could also turn to Veronica and Jughead, tools which let you do a full-text search of some popular Gopher server menus. These fine examples are counterbalanced by the truly sublime part of Gopher space. Check out the legislative information system at Gopher.cc.utah.edu, where last year's legislation is organized by bill number or full text search, and nothing in between. And don't miss the Salt Lake City Police Department's daily reports: "October 17, 1993, 10 PM. Officers were dispatched to the 7-11 after reported gang members had threatened the clerk claiming to have a gun." Or slip on over to The Management Archive at chimera.sph.umn.edu (don't these Minnesota people talk to each other?), which is advertised as "an innovative idea with considerable potential to improve the flow of ideas, knowledge, and substantive learning between researchers and practitioners and across the management academy." It will take a long while for The Management Archive to achieve its potential. I don't mean to be the grouch of the Internet. The Minnesota Gopher implementers have given us a great product. Lots of information organizers out there are trying to make sense of the world of the Internet. They're shooting at a fast moving target. Keep that in mind as you navigate in Gopherspace. Joel Snyder jms@Opus1.COM - Joel Snyder is a senior partner at Opus One, specializing in telecommunications and information technology. He spends most of his time running around the world helping companies with their networking problems. ---------- Snyder's Guidelines For Good Gophering 1. Remember that Gophers are microscopic views on the overstuffed world of the Internet. If you think you're seeing it all, you're not even close. To find something, you may have to take many twisty passages, all seemingly alike. 2. Even in the world of Gopher, cross-system index tools like Veronica, Jughead, and Archie are quite incomplete. There is no lazy way out of hard digging with your Gopher client. 3. When you find a well-organized place in Gopherspace, make yourself a note using Gopher's "bookmark" facility so you can come back. Your personal list of Gopher jewels is something you'll come back to again and again. 4. Information without organization is a waste of time. If you think being a Gopher master is for you, make sure you've got the commitment to keeping your Gopher up to date and accurate. Otherwise, don't even bother. If you're a Gopher user, remember that there is no quality assurance on the data you are getting, and there is no guarantee that what you have is the latest and greatest. 5. When you do find a particularly tasty resource, make sure that others know about it. Some wonderful resouces are well hidden on the Internet. If you know about one, share the wealth. 6. Be flexible. Gopher searches are only as good as where you start. If your strategy isn't working, start somewhere else. 7. Gopherspace is not hierarchical. There is no information architect. There is no "top" below which everything falls into neat pigeonholes. There is no guarantee that you can get there from here, or that you're not going around in circles. 8. When you take information from the Internet, remember that you owe a debt to those who made it available. Repay your debt by making your own contribution. Ask your local Gopher master; he or she will probably be happy to accept a donation of time to help organize, update, index, and improve the system.