Since Gopher is the oldest and most decentralized of the services, it provides the most extensive access to useful services. Online library catalogues, the Project Gutenberg library, FTP archives, and numerous exits to telnet.
The telnet access to demonstration Gopher systems (and through them to Gopher servers on *any* campus) is through:
x.x.x.x.x.x.x.x.x
Students will of course want to use Gopher directly from their school"s system. Try typing "gopher" at any prompt.
Wide Area Information Services (WAIS) WAIS was developed in a very different environment from Gopher. It was developed by a joint collaboration of Thinking Machines, Inc.
(Artificial Intellegence technology), Dow Jones News Retrieval (Information systems), and Apple Computer (User interface). Its ability to find information given a plain English description of what you want ("hey, find everthing on Personal Computers and Health) is truely mind- boggling. It returns a list of "hits" together with a likelihood that it contains what you wanted. It can also look for doc.u.ments that are "something like" a sample doc.u.ment.
The user interface is a pleasure to use--but requires a direct or SLIP connection to the Internet. The line-oriented version that is publically available is a pale imitation of the real thing.
WAIS strength is its ability to retrieve information from almost any source, not just FTP sites. A list of all WAIS sources is maintained in a directory-of-directories (available at think.com). You import a set of instructions on how to access a given information server to create your own personalized list of sources.
This service is probably the one of greatest interest to business (i.e.
non-academic and non-library) users. To try out WAIS, telnet think.com and log in as SWAIS.
World Wide Web (WWW or W3) The newest of the three services is the World Wide Web. It was created at the European European Center for High Energy Physics (CERN). It is based on yet another technology (besides FTP and client-server)-- hypertext. The World Wide Web views the entire complex of FTP sites as a single "doc.u.ment" with cross-references.
A WWW server lets you read that doc.u.ment and jump to any cross-reference that you find--hence the term "hypertext". The result is rather like a menu driven system but (at least in the graphical interface versions) you stay inside the familiar context of a text editor. If you can imagine clicking on a cross-reference and having your text editor fetch the doc.u.ment from an FTP site you will get the idea.
The doc.u.ments that can be viewed by WWW are ASCII text with special "tags" that give a addresses of the "hypertext links." The tags use the syntax of the Standard Generalized Markup Language (SGML). SGML is a language used by scholars to mark text for academic research. The WWW vision of the library of the future is a collection of doc.u.ments spread all over the world, the whole of which can be looked at starting from any one of them. Sort of like having the whole world on your desktop.
There is not "top" node to the Web, but you can find points of entry at:
telnet info.cern.ch (European Center for High Energy Physics in Geneva, Switzerland, the "home" of WWW).
Summary of Navigation Tools To summarize, here are the three systems discussed, together with their underlying technology and "const.i.tuency":
Gopher : Simple FTP and Telnet : Campus Info
WAIS : Artificially Intellegent searches : Business
World Wide Web : Hypertext and SGML markup : Ac. Research
Research Methods II: Usenet Newsgroups [This chapter is under construction]
Research Methods III: Advanced Techniques The previous two chapters covered the more or less standard techniques for finding your way around the net. This chapter covers more
There are three basic "problem situation" that every researcher using the Internet will eventually face:
(1) You know who has the information but you don"t know their "address".
(2) The information is on the net, only you don"t know where.
(3) The desired information is not on the net, but their *is*
information on how to get it from a non-network source.
The methods described in this chapter are more tentative than in the preceeding two chapters. They don"t always work.
Finding Persons and Computers There are a couple of standard methods for checking and verifying E-mail addresses.
% ping rtfm.mit.edu
(Remember that "%" is the prompt the computer gives you. Your system prompt may look different). You should get back a message saying "rtfm.mit.edu is alive" or something like that.
Many machines support a command called "nslookup" that will return the dotted decimal address given the name of the machine
% nslookup rtfm.mit.edu
returns "x.x.x" as the dotted decimal address.
If you can guess the name of person or inst.i.tution--this is not hard-- then you can try to see if you have a valid address by "fingering" your intended victim:
% finger [email protected]
If the system supports the "finger" feature (and many don"t), you can try any number of guesses or permutations. If you succeed, you can find out lots of information about the person: their telephone number, when they last logged on, when they last read their mail, what department they work in, etc. Many systems allow you to leave a file called ".plan"--note the initial dot--that contains further contact information.
Good guesses for names:
o last name (bach),
o last name with one or more initials (pbach, pdqbach),
o three (or more!) initials (pdqb)
o nicknames, cute handles, etc. (fluffy,aragorn)