The amount of information available on the Internet is staggering and continues to grow. The Internet's size, as well as the lack of a centralized, governing body to issue standards for information retrieval, may cause even the bravest "navigator" frustration and confusion when attempting resource location. The compilation and maintenance of traditional paper directories is now obsolete because currency of information is all important. Any serious attempt to compile resource directories must now take place on the medium being indexed.
Today, an Internet user has access to a number of resource-discovery tools, including Gopher, the World Wide Web, Veronica, Jughead, and WAIS. But this was not the case when Hytelnet was developed. In fact, Hytelnet was the first such tool to be made freely available to the Internet community. Its original purpose, which still holds true today, was to make access to resources on the Internet as painless as possible for both the beginner and the experienced user.
This chapter explains how and why the program was developed, how to use it, and how it has been adapted by various Internet users.
Hytelnet was developed in late 1990. It is an acronym for HYpertext browser for TELnet-accessible sites on the InterNET. These sites include hundreds of online public-access library catalogs, bulletin boards, campus-wide information systems, Free-Nets, network information centers, and full-text databases. Also included are an Internet glossary, a help file, and a directory of cataloging software. The initial design was aimed at the IBM-PC user, and that design still forms the basis for other versions running on different platforms.
When I first learned how to use Telnet to connect to remote sites, the only available information regarding login procedures was in paper directories. These directories were written by Art St. George (1990) and Billy Barron (1991). At that time, there was no mention of an electronic directory of such resources. This struck me as odd, because the lists were carrying information about an electronic procedure. I downloaded the directories, tested the information, and made connections to most of the sites listed. Immediately, a number of questions arose. Paper lists lose currency very quickly. So how often would they be updated to retain their usefulness? Obviously, more sites would be added on a continual basis; so would a user have to keep downloading new paper copies each time a change was made to the information?
I had already created a hypertext utility, called HYPERVAX, for helping people understand all the commands associated with VAX/VMS mail. This utility was renamed HYMAIL; it was designed to be driven by HyperRez (1989), written by Neil Larson from MaxThink.
Barron's list, UNTs Accessing On-Line Bibliographic Databases, contained entries listing full Telnet addresses, login and logout procedures, and (when available) an indication of the software running the catalog.
The list was loaded into a very fast and powerful text editor called QEdit (1991), which is ideal for marking chunks of text and saving the text to discrete files. Each location, therefore, was housed in a discrete file. The naming of the files corresponded with the domain of the country in which the site is located (for example, FI for Finland and BR for Brazil).
The next step was to determine the arrangement of the information so that files with sensible hypertext links could be created. The HyperRez driver enables the linking of pure ASCII files with hypertext jumps. (A jump is text surrounded by angle brackets.) This was accomplished by creating files that listed the names of the sites belonging to a particular country. These files then had to be linked to a file naming all the countries, which is eventually linked back to the START.TXT file.
If hypertext links lead somewhere, they can be inserted into any file. It seemed like a logical step to create files for the various cataloging systems listed in the descriptions of the files. These are simply small help files that enable a user to understand how a given catalog is to be searched.
Depending on the document being formatted, hypertext creation can be frustrating and time consuming. Fortunately, Barron had created an ideal document as the basis for this purpose. My job was to design a system that users would find intuitive and not distracting. I also felt that I should create other files (such as a file explaining the functions of the arrow keys, a READ.ME file explaining what the program does, and a CUSTOM file that explains how a user can update files to keep current on all new information).
The PC version of Hytelnet is designed to run as a terminate-and-stay-resident (TSR) program. That is, the program is loaded into the computer's memory and invoked when needed. It runs as a perfect complement to communications software. While connected to a mainframe with a PC, a user can invoke the Hytelnet program by pressing Ctrl+Backspace to browse and return the program to memory when the information has been found. The arrow keys enable quick maneuverability, and the program can be reinvoked at will.
The first version of Hytelnet was released in January 1991. Many Internet users responded favorably to the program. The comments received were positive and encouraging, and many felt that such a directory was long overdue. One user, Richard Duggan of the University of Delaware, immediately adapted the files to run under Windows as a stand-alone program called CATALIST (1991).
As my interest in the Internet grew, I began to discover other types of Telnet-accessible resources, including bulletin boards, Free-Nets, and campus-wide information systems. What began essentially as a guide to library catalogs quickly became a complete index of Internet resources. It wasn't long before a second version of the program was released, containing hundreds of other sites.
Certain Usenet newsgroups also became a mine of information about new sites (such as alt.bbs.internet and alt.internet.services) that were becoming available for Telnet login. I decided to create my own mailing list for informing users of new sites. Now, thanks to Diane Kovacs at Kent State University, a Listserv list called HYTEL-L is now in operation. It keeps users up to date on new versions of Hytelnet and all new, updated, and deleted files.
NOTE The Hytelnet updates distribution list, HYTEL-L, is a mailing list that sends information about new versions of the popular Hytelnet program, which gives users access to all known Telnet-accessible sites on the Internet. List members also receive announcements of new, changed, and defunct sites—announced between full versions of the program. The list is moderated, which means that only the list owner posts messages. Any messages you send to HYTEL-L are forwarded to the list owner. Feel free to contribute any of the following:
To better show how Hytelnet operates, this section presents a sample session. Suppose that you are interested in discovering the Internet address and login procedures for the Medical College of Wisconsin library in the United States. On the PC, the command to invoke Hytel-net is entered at the DOS prompt:
C:\> cd hytelnet
C:\Hytelnet> HR
To load Hytelnet into memory, press the Ctrl+Backspace combination; the START.TXT file starts and displays the following information:
Welcome to HYTELNET version 6.8
November 27 1994
What is HYTELNET? <WHATIS>
Library catalogs <SITES1>
Other resources <SITES2>
Help files for catalogs <OP000>
Catalog interfaces <SYS000>
Internet Glossary <GLOSSARY>
Telnet tips <TELNET>
Telnet/TN3270 escape keys <ESCAPE.KEY>
Keystroke commands <HELP.TXT>
Up/Down arrows MOVE Left/Right arrows SELECT F1 for HELP anytime
CONTROL/HOME returns here ALTT quits
.............................................................
HYTELNET 6.8 was written by Peter Scott
Northern Lights Internet Solutions, Saskatoon, Sask, Canada
(aa375@freenet.carleton.ca)
In this example session, the <SITES1> link is selected to access library catalogs:
OnLine Library Catalogs
<SITES1A> The Americas
<SITES1B> Europe/Scandinavia
<SITES1C> Asia/Pacific/South Africa
Then, <SITES1A> is selected to access information about the Americas:
The Americas
<BR000> Brazil
<CA000> Canada
<CL000> Chile
<MX000> Mexico
<US000> United States
<VE000> Venezuela
Then the <US000> link, the top-level United States directory, is selected. Because there are so many catalogs accessible in the United States, a submenu of catalogs was created, dividing the catalogs into types; from this submenu, <US000MED> is selected:
United States Medical Libraries
<US376> Albert Einstein College of Medicine
<US482> Arizona Health Sciences Center
<US011> Association of Operating Room Nurses
<US098> Audie L. Murphy Memorial Veterans' Administration Hospital
<US520> Bowman Gray School of Medicine, Wake Forest University
<US381> Cornell University Medical College
<US293> Creighton University Health Sciences Library
<US011> Denver Medical Library
<US145> Georgetown University Medical Center
<US214> HSLC HealthNET (Health Sciences Information Network)
<US408> Massachusetts College of Pharmacy
<US362> Medical College of Ohio
<US242> Medical College of Wisconsin
(Only a portion of the file is being shown for this sample session.) From this list, select the <US242> link to see information about the Medical College of Wisconsin:
|—————————————————————————|
| |
| Medical College of Wisconsin |
| |
| TELNET ILS.LIB.MCW.EDU or 141.106.32.19 |
| login: library |
| |
| OPAC = INNOPAC <OP009> |
| |
| To exit, type Q |
|—————————————————————————|
This screen shows the Telnet address you need to access the site along with the login and exit requirements. Notice also that there is another link embedded in the file; if you select that link, you see the help for searching an INNOPAC catalog:
|——————————————————————————————————|
| |
| Using INNOPAC |
| |
| INNOPAC is very easy to use. Just press the letter |
| or number next to the item that you want. There is no need |
| to press the ENTER or RETURN key when choosing one of the |
| menu options. For example: |
| |
| Title searches: To search for a particular title, select |
| T on the main menu. |
| |
| Author searches: To search for a particular author, |
| select A on the main menu. |
| |
| Subject searches: To search for a particular subject, |
| select S on the main menu. |
| |
| Keyword searches: Either "k" or "w" (varies from system to |
| system) as listed on the menu. |
| |
| Other search options: Different INNOPAC libraries have added |
| additional search options, such as Medical Subject Headings, |
| Call Number, SuDocs numbers, Reserve Lists, etc. |
| |
| Popular options available when looking at any one record |
| include: |
| |
| S > Show items with the SAME SUBJECT |
| Z > Show items nearby on the shelf |
|——————————————————————————————————|
The PC version of the program doesn't make automatic connections to the remote sites. However, other versions do; they are discussed in the following section.
In the early days of Hytelnet, many Internet users had machines other than IBM-compatible PCs and were unable to gain access to the Hytelnet files. That situation was first rectified in 1992 by Earl Fogel of the University of Saskatchewan. He designed versions of the program that run on VAX/VMS and UNIX machines. One attractive new feature of these versions is that they connect automatically to remote sites. If we had used the Fogel versions of Hytelnet in the example session shown in the preceding section, Hytelnet would have asked whether we wanted to connect. If so, the program would automatically run a Telnet session. Of course, we would still have to log in. Because there are now so many accessible sites throughout the Internet listed in Hytelnet, browsing through the program can easily become a chore. Fogel has created a search feature that enables a user to type a term and be presented with a new index; in effect, you can create customized indexes.
NOTE The UNIX/VMS version of Hytelnet is also available for browsing through the Web at http://access.usask.ca/cgi-bin/hytelnet. For more information on this version, contact Earl Fogel at
Macintosh users can thank Charles Burchill of the University of Manitoba for creating a HyperCard reader for the original Hytelnet database. It reads the specially formatted Macintosh or IBM files over a mixed network, such as Novell or AppleShare. It also can automatically launch the NCSA Telnet or TN3270 software and connect a user to a chosen site. There is also limited support for serial and dial-up connections. Burchill has designed the HyperCard reader to read letters from HYTEL-L Listserv messages or saved files from the bit.listserv.hytel-l newsgroup and automatically update the database. This version of Hytelnet runs only with HyperCard 2.1 and higher. The preferred operating system is System 7.x, but it also runs with System 6.0.5 or later.
John Doyle of Washington and Lee University has linked the information contained in the original Hytelnet database with the power of Gopher. To see this action, point your Gopher client to liberty.uc.wlu.edu 70 and move to Explore Internet Resources->Hytelnet. The URL is gopher://liberty.uc.wlu.edu:70/11/internet/hytelnet.
Doyle's design maps Hytelnet's hypertext structure into a Gopher structure by creating a directory for each of Hytelnet's hypertext menu links and each Telnet or TN3270 link. The directories also contain copies of the original Hytelnet file that created the directory. For example, follow this menu hierarchy:
Library Catalogs
Asia/Pacific/South
Australia
Macquarie University
You will find this Gopher screen:
1. Macquarie University <TEL>
2. OPAC = DYNIX.
3. Macquarie University.
Item 1 is the Telnet link to the library at Macquarie University. Item 2 lists the catalog software in effect (other embedded links would also appear as menu choices). Item 3 is the original Hytelnet file that contains full login instructions and information about the site.
One particularly useful feature of the Gopher version of Hytelnet is a directory of new or revised entries. These are automatically sorted in reverse chronological order and cover postings to the Hytelnet mailing list made in the previous month. Another feature is the searchable database of the names of all the remote destinations. If you enter a term, the program retrieves a Gopher menu of links, as well as the original Hytelnet file containing the term. The URL is e gopher://liberty.uc.wlu.edu:3004/77/.
To join the HYTEL-L mailing list, send the following e-mail message to LISTSERV@KENTVM:
SUBSCRIBE HYTEL-L First Name Last Name
The list server also has an Internet address:
LISTSERV@KENTVM.KENT.EDU
To remove your name from the HYTEL-L mailing list, send the following e-mail message to LISTSERV@KENTVM:
UNSUBSCRIBE HYTEL-L
To stop HYTEL-L mail when you go on vacation, send the following e-mail message to LISTSERV@KENTVM:
SET HYTEL-L NOMAIL
To resume HYTEL-L mail delivery, send the following e-mail message to LISTSERV@KENTVM:
SET HYTEL-L MAIL
To determine what your HYTEL-L distribution settings are, send the following message to LISTSERV@KENTVM:
QUERY HYTEL-L
The QUERY command is useful to try if you stop receiving HYTEL-L mail. The moderators may have had to set you to NOMAIL because of problems with your e-mail account. To send a message to HYTEL-L, send your e-mail message to HYTEL-L@KENTVM.
To obtain a list of HYTEL-L users, send the following e-mail message to LISTSERV@KENTVM:
REVIEW HYTEL-L F=MAIL
To receive full documentation about searching the message database, send the following e-mail message to LISTSERV@KENTVM:
INFO DATABASE
A brief tutorial for Hytelnet is also available. Send the following e-mail message to LISTSERV@KENTVM:
GET SEARCH DOC F=MAIL
To see what files are available, send the following e-mail message to LISTSERV@KENTVM:
INDEX HYTEL-L F=MAIL
To retrieve a file, send the following e-mail message to LISTSERV@KENTVM:
GET File Name File Type
All messages to the conference are automatically archived for one month only. Users can search the message database for specific information using complex Boolean queries. The most reliable method of searching the database is to submit batch-search jobs to the list server in e-mail messages. Search features include nested Boolean expressions, search limitation by date and time, and SOUNDEX searches.
For more information about HYTEL-L, contact Peter Scott at
aa375@freenet.carleton.ca
His postal address is also given:
324 8th Street East
Saskatoon
Saskatchewan
Canada S7H0P5
This section gives the procedure for retrieving Version 6.6 of the PC version of Hytelnet with File Transfer Protocol (FTP).
To retrieve Hytelnet, follow these steps:
The Hytelnet program is archived using PKZIP.EXE. To unarchive the program, you must be able to unzip the file. If you have the file PKUNZIP.EXE, use it to unarchive the hyteln68.zip file you just downloaded. If you do not have PKUNZIP.EXE, you can retrieve it by following these instructions:
You also can unarchive the hyteln68.zip file with UNZIP.EXE. To retrieve it, follow these instructions:
NOTE Because of the plethora of PC communications programs, I can't give more specific instructions here. Check your software instructions for help about downloading a binary file from your Internet account to your PC.
Make a new directory on your hard disk (for example, type mkdir hytelnet). Copy PKUNZIP.EXE or UNZIP.EXE and HYTELN68.ZIP into the new directory. Make sure that you are in that directory and then enter pkunzip hyteln68.zip. PKUNZIP unarchives HYTELN68.ZIP, which |contains the following files:
HYTELNET.ZIP
READNOW
The READNOW file gives full instructions for unarchiving the HYTELNET.ZIP file. (Simply put, if you are using PKUNZIP.EXE, you must unzip the HYTELNET.ZIP file with the -d parameter so that all the subdirectories are recursed. If you are using UNZIP.EXE, no parameters are required.)
To use Hytelnet, refer to the README file included with the package.
REFERENCE NAMES