Internet Gopher User's Guide Editor Paul Lindner DISCLAIMER OF WARRANTY COPYRIGHT NOTICE Copyright (C) 1993 by the Regents of the University of Minnesota. All rights reserved. Permission to use, copy, modify, and distribute the Gopher software for non commercial purposes and without fee is hereby granted, provided that the University of Minnesota copyright notices and this permission notice appear in all copies, and that the name University of Minnesota not be used in advertising or publicity pertaining to this software and documentation without specific, written prior permission. The University of Minnesota makes no representations about the suitability of this software and documentation for any purpose. It is provided 'as is' without express or implied warranty. Commercial use of Gopher requires specific permission from the University of Minnesota; contact the internet gopher development team at for further information. Printed in the U.S.A. TRADEMARKS Internet Gopher is a trademark of the University of Minnesota. UNIX is a registered trademark of UNIX system Laboratories, Inc. All other products or services mentioned in this manual are covered by the trademarks, service marks, or product names as designated by the companies who market those products. ISNB 0-000-000000-0 ABCDEFGHIJ-DO-89 CHAPTER 1 Introducing Gopher Introduction What is the Internet Gopher? Types of Information Finding Information Some Example Information Using this Manual Platforms Supported Conventions used in this Manual Support Questions? CHAPTER 2 Getting the Internet Gopher Software Getting the Software Subsequent Retrievals CHAPTER 3 TurboGopher Overview Requirements Getting Help Navigating Gopherspace Alternate Views Connecting to Terminal-based Services Advanced Navigation Features Authenticated Servers Bookmarks and Bookmark files Item Attributes Canceling Slow Network Operations Starting and Configuring TurboGopher TurboGopher Options Miscellaneous For Experts CHAPTER 4 PC Gopher III System Requirements How to Obtain the Software Installation Installing the Gopher Software Configuring Gopher Application Configuration Network Configuration Advanced Network Configuration Options Gopher and Command Line Switches The Gopher User Interface Using Gopher Help! Acknowledgments CHAPTER 5 The UNIX Gopher Client Overview Requirements Starting the UNIX Gopher Client Navigating Gopherspace Special Items Advanced Navigation Features Bookmarks Saving and Downloading CHAPTER 6 Installing the Unix Gopher Distribution Overview Requirements Compatibility NeXT Indexing WAIS Indexing Preparing to Compile the Distribution Other Optional Features Special Modifications for Compatibility Compiling and Installing the Distribution Testing CHAPTER 7 Configuring the Unix Server Overview Requirements What is gopherd? First Steps Adding Information to the Server Starting the Gopher Server Optional Features for the Gopher Server CHAPTER 8 Go4gw - The Gopher Gateway Overview Requirements Installation Writing New go4gw Gateways CHAPTER 9 Common Questions Overview CHAPTER 1 Introducing Gopher This chapter describes how you can use the Internet Gopher system to easily pub lish and retrieve information on a network. Introduction The internet is a wide vast place with many resources available. Searching, finding and retrieving these resources has been difficult in the past. The Internet Gopher was developed to let an average user access these resources quickly and simply. What is the Internet Gopher? All over the world, data is stored on computers, many of which are connected by the Internet, a confederation of computer networks. With Internet Gopher you can easily access publicly available information stored on many of these connected computers. Gopher combines features of electronic bulletin board services and databases, allowing you to either browse a hierarchy of documents, or search for documents that contain certain words or phrases. The Internet Gopher software was conceived at the Computer and Information Ser vices department of the University of Minnesota. Software developed at the Uni versity of Minnesota is freely distributable for non-commercial puposes. Types of Information Gopher supports a diverse range of data, all of which can be accessed by a simple keystroke or click of the mouse. Here are some of the most popular data types: Directories The most basic information type in gopher is a directory. A directory is a list of documents. Directories allow easy browsing of information. Items can be organized into specific areas, making it easy to find the information you need. In addition, a special type of directory called a "link" allows gopher to reference directories on a different computer. This allows gopher to traverse a hierarchy of information residing on multiple machines. These links are transparent, you won't notice that you're connecting to another machine. Text Files Most of the information in gopher is stored in ASCII text files. These files can be used with most popular computer applications Search Items This is a special kind of directory. You can specify any number of keywords to a search item. Only those documents that match the given criteria will show up in the resulting list. Telnet Sessions This item type is a reference to a public telnet connection. Selecting one of these items will connect you to a system using terminal emulation. Phone Books This document type is a special version of the search item based upon the Ph/Qi server developed at the University of Illinois. The phone book search allows you to search on certain fields of a database, such as name, phone number, or address. Multimedia Gopher supports a number of multimedia file formats including images, audio and video file formats. Images such as weather maps are available. Audio data, including the presidential debates is available. Movies in Quicktime and MPEG are avail able. Formatted Text Some Gopher servers will allow you to view documents in formats other than text. Postscript is one of the most popular formats for this "Rich Text." Finding Information When using Gopher, looking for information located in other continents is as easy as looking for information residing on a computer in the next room. When using Gopher, you begin at the first or root level of your home Gopher Server. From there you can choose between two methods or locating information: browsing and searching. Browsing You may want to follow the paths in Gopher from level to level until you find the data you're looking for. From the first level, you can choose a topic, which leads to another level, and another, until you finally come to an item that looks interesting. Some Example Information For instance at the University of Minnesota, you might look for a salmon recipe by looking in the Fun & Games directory, then in the Recipes directory, and then in the Seafood directory until you see an item with salmon in its title. Searching Alternatively, you may want Gopher to do the work for you by using a search item. For instance, at the University of Minnesota you could select a search item called Search Recipes. A message prompts you to type in the words you're looking for; you type salmon. The server searches the text of a collection of items and lists the ones that have the word salmon in them. You can then examine these items until you find one that contains a recipe that strikes your fancy. Additionally, some search items will let you specify a complex search expression for your query. If you only had sugar, eggs and chocolate chips, you could search for items that contained all of these words by specifying sugar and eggs and "choc olate chip" to the Search Recipes item. Some Example Information We think the easiest way to get a grasp of the information inside of GopherSpace is to try it. However, for your listing pleasure we have a lisitng of some of the more popular information: * Weather Forecasts and current conditions for the United States and Canada. * Recipes * Movie Reviews. * Computer Questions and Answers. * Weather Maps. * Movies of Chemical reactions. * E-mail addresses and phone books for most major institutions * Newspapers and USENET news. * Full electronic versions of many of the major classics, Shakespeare, Moby Dick, etc. * Library Catalogs from around the world. * And many many more! Using this Manual You do not need to read this entire manual to use the Internet Gopher. Most likely, you will only need to read the chapters about the Internet Gopher client for your software. If you wish to publish information you will want to pay attention to the chapters referring to the Internet Gopher servers for your system. The First Steps See "Getting the Internet Gopher Software" on page 15 for information about retrieving and installing the necessary software. Then read the chapter appropriate for your system. Common Questions It's a good idea to read through the"Common Questions" on page 93. This chapter contains answers to the most frequently asked questions about the Internet Gopher - from installing it to serving data. As Needed Chapters x and x may be consulted when you blah blah blah. For an interesting history discussion refer to Chapter X - Growing Gopher. Platforms Supported The Internet Gopher works on a number of platforms and operating systems. The operating system for all of the following must have TCP/IP networking support and a connection to a network. Gopher Clients Clients to access data are available for the following platforms: * Macintosh * DOS * Microsoft Windows * Unix, full-screen, emacs and Xwindow * VMS full screen. * NeXTstep * OS/2 * VM/CMS * MVS Gopher Servers Servers to publish information are available for the following platforms: * Unix * VMS * Macintosh * VM/CMS * DOS * MVS Conventions used in this Manual [not applicable for ASCII version] This manual uses these conventions: * The Courier font is used to to show sample output of textual clients. Bold Courier is used whenever you have to enter text or commands. * In reference to the Unix Server, gophertop refers to the directory where the gopher source code is installed. * Figure boxes are used to show usage. Support Questions? We hope you enjoy using Gopher and that it helps you in you to access and publish data quickly and easily. If you have questions or comments, please send us electonic mail at gopher@boombox.micro.umn.edu. Or send paper mail to the following address: Internet Gopher Development Team 100 Union Street SE #152 Shepherd Labs - University of Minnesota Minneapolis, MN 55455 Or fax us at +1 (612) 625-6817. CHAPTER 2 Getting the Internet Gopher Software This chapter will show you how to get the Internet Gopher software. You should only need to read the first section "Getting the Software" on page 15 and the sec tions appropriate to your platform. Getting the Software You will need to retrieve the Internet Gopher Software from the software repository at the University of Minnesota. Follow these instructions to do so. The First Retrieval Most of the software for the Internet Gopher is available on the machine boombox.micro.umn.edu. This machine is on the Internet network. To retrieve the software for the first time you'll need to use an FTP client on a local machine con nected to the internet. FTP, the file transfer protocol, is a quick, reliable and widely popular means of transferring files across the Internet. FTP clients come in many varieties. On a Unix or VMS system the command is called ftp. On the Macintosh there are two ftp applications, Fetch, and Xferit. NCSA has developed a version of ftp for the IBM-PC called `ftp'. Refer to your local network administrator if you don't have an ftp client. Once you connect to the boombox server you will want to change to the directory /pub/gopher The different portions for gopher are in this directory. The file 00README con tains descriptions of each directory. Refer to Table 1, "Available Gopher Software," on page 16 for a breakdown of the contents of each directory. +----------------------------------------------------------------+ TABLE 1. Available Gopher Software Directory Description: Mac_server Gopher server for the Macintosh Macintosh-TurboGopher A graphical Gopher client for the Macintosh NeXT A graphical Gopher client for NeXTstep PC_client A graphical Turbovision based client for PCs running DOS and using the Clarkson/Crynwr packet drivers PC_server Two PC Gopher server implementations, one based on Phil Karn's NOS, another using Clarkson/Crynwr packet drivers. Rice_CMS A Gopher server and client for VM/CMS systems written at Rice University. Unix Gopher for Unix. Includes a Gopher server, full screen client, Xwindow client, and an emacs client. VMS Gopher for VMS. Includes a Gopher server and full screen client VieGopher A Gopher Server and client for VM/CMS written at Vienna. gopher_protocol Descriptions of the Gopher network Protocol are stored here. incoming Submitted untested gopher software. mvs A Gopher server and client for MVS. os2 A Gopher client for OS/2. +----------------------------------------------------------------+ Getting the Software Each directory contains gopher for a specific platform. You should retrieve the soft ware from the specific directories that are compatible with your hardware and oper ating system. There are README files inside of each directory that will help you decide which files to transfer. Once you find a file that interests you.you should use the GET function of your FTP client. This usually means typing "get filename", or clicking on the appropriate file with your mouse.. Some clients, notably Unix, VMS, and IBM-PC will reqire you to specify a binary "transfer mode." You can set this mode by typing binary at the ftp> prompt. Refer to Figure 1, "Using FTP with Unix," for an example ftp session to boombox.micro.umn.edu +----------------------------------------------------------------+ FIGURE 1. Using FTP with Unix % ftp boombox.micro.umn.edu Connected to boombox.micro.umn.edu 220 boombox FTP server (Version 4.1 Tue Apr 10 05:15:32 PDT 1990)ready. Name (boombox.micro.umn.edu:lindner): anonymous 331 Guest login ok, send ident as password. Password:(TM) 230 Guest login ok, access restrictions apply. ftp> cd /pub/gopher/Unix 250 CWD command successful. ftp> binary 200 Type set to I. ftp> get gopher1.12.tar.Z 200 PORT command successful. 150 Opening BINARY mode data connection for gopher1.12.tar.Z (306512 bytes) 226 Transfer complete. 306512 bytes received in 2.4 seconds (1.3e+02 Kbytes/s) ftp> quit % +----------------------------------------------------------------+ Subsequent Retrievals Things get easier after the first retrieval. You don't have to use the arcane ftp command to get the Gopher software, you can use Gopher itself. You will want to familiarize yourself with your gopher client before doing this though. See the chapters that follow to find out about clients. You can use your Gopher client to connect to the software repository at boombox.micro.umn.edu And choose the files, just as you would with ftp. However, now you can just select the software you want from the menu instead of typing commands to retrieve the file. CHAPTER 3 TurboGopher TurboGopher is a Gopher client for the Macintosh. The following chapter contains information on using, configuring and administrating the TurboGopher application. A Websters definition is: TurboGopher n. 1. A small rodent with a turbocharger strapped on its back to increase its speed and ferocity. 2. (Amer. colloq.) Native or inhabitant of Minnesota after consuming three double espressos. 3. (Amer. colloq.) An Olympic sprinter who runs errands, does odd-jobs, fetches or delivers documents for office staff. 4. (computer tech.) Speed-optimized Macintosh software following a simple protocol for tunneling through a TCP/IP internet; network speed is achieved by using turbo charged software; incoming bits spin the turbine that pumps out the outgoing bits. Overview TurboGopher is a Macintosh application that we believe is (still!) the fastest Macin tosh Gopher client available. Beyond optimizing TurboGopher for raw speed while fetching documents and directories, we turbocharged the user interface by displaying information as soon as possible... you can read the first part of a document or directory while the rest is being fetched. This version of TurboGopher also supports many of the Gopher+ extensions to the original Internet Gopher protocol. In spite of the design goal to run fast as possible, TurboGopher is a good Mac citizen: it shares time with other applications. You can put TurboGopher in the back ground to fetch lengthy items in the background while you work in another application in the foreground. Requirements You will need a copy of the TurboGopher application, if you haven't retrieved this yet refer to"Getting the Internet Gopher Software" on page 15. You will also want to retrieve the "helpler applications." These applications will allow you to connect to telnet sites, view graphics, etc. TurboGopher requires a Macintosh running a system 6.0.7 or later. We highly reccomend using System 7 or higher. Without it you won't be able to automatically launch the "helper applications." You should have at least one megabyte of memory, more if you're using Multi finder or System 7. You will need a correctly configured copy of MacTCP and a network connection. If you don't have MacTCP you can get it from the Apple Programmers and Developers Association (APDA). MacTCP is a licensed product, it is not free. Getting Help The information in this chapter can also be found by picking the menu item "Help" from the "Gopher" menu of TurboGopher. In addition to this information you are reading right now, if you are using System 7 (and we strongly recommend that you do), some Balloon Help is available in Tur boGopher. See your Macintosh System 7 documentation for information on using Balloon Help. +----------------------------------------------------------------+ FIGURE 2. TurboGopher-Top Level not available in ASCII version +----------------------------------------------------------------+ Navigating Gopherspace To navigate through gopherspace, double-click on any interesting items to open (fetch) them. If you double-click on a file, TurboGopher will fetch and display the file. Opening a folder will let you view its contents. Double-clicking on the ques tion-mark (Search) icons will let you search a database. On most gopher servers these databases are full-text indexes of a collection of information. Full-text index means that every word in every document is considered a keyword. The best way to search a gopher full-text database is specify the words for which to search when TurboGopher presents you with a dialog box. The results of the search are returned as a list of documents containing those words. Phone Books If you double-click on the phone-book icons you can search electronic phone books. Phone book databases are structured in the sense that they have fields (i.e. a name field, an address field, and a phone number field). The simple phone-book lookup dialog assumes that you want to search for someone by name. Click on the `More Choices' button if you wish to specify exactly which fields to search in, when you formulate a query. The result of the query is displayed in a document. +----------------------------------------------------------------+ FIGURE 3. TurboGopher- Phone Book Search not available in ASCII version +----------------------------------------------------------------+ Software Archives Disk icons represent archived Mac software or documents that you can fetch; these items will be saved to your hard disk. Such items are often stored in compressed form to save space. You need to have certain Mac utilities to `uncompress' such items. The most common such utilities are StuffIt and Compactor. Both may be obtained from the Info-Mac archives with gopher. If you are running System 7 and TurboGopher retrieves an item that needs to be `unstuffed' or `uncompacted', it will ask you if it should open them via Finder and the appropriate decompress util ity. We recommend that you use CptExpand and SitExpand for your decompress needs; both of these are available with the TurboGopher software distribution as well as from the Info-Mac archives. Items that are specific to MS-DOS are seen as documents with a tiny `PC' embla zoned on them; similarly items specific to UNIX show up as documents with a tiny `UX' on them. You can retrieve either of these kinds of items if you like. TurboGo pher will ask you if you wish to save them to your disk. You may not `view' these items as text. In fact it will probably make very little sense to fetch such items unless you intend to further transfer them later from your Mac to another kind of computer, or if you have some special tools on your Mac that will allow you to use these kinds of items. Pictures Items appearing as documents with a starburst on them are probably graphics or picture files (typically in GIF, JPEG, or PICT format). TurboGopher will transfer such files, save them on your disk, and optionally launch a picture-viewing applica tion if you wish to view the picture. Alternate Views Gopher+ server may store more than one representation of a document (an image, text, sound, video, etc. ). If more than one view of a document is available, you can select between the views by choosing the `Get Alternate Views' item from the Gopher menu. When alternate views are available, an alternate view window will be displayed with descriptions of the views. For many alternate views of documents, TurboGopher will save a copy of the item to your Macintosh's disk, and then (under System 7) ask the Finder to launch a helper application to actually display the item. You can configure which applica tions should be used as TurboGopher helpers; to do this see the section on configu ration options. Connecting to Terminal-based Services Opening the `terminal' icons will make TurboGopher launch NCSA Telnet (or TN3270 if appropriate) and start a terminal session to a terminal-based information system. Typically these are library catalogs or other such services. Note that for this to work, you need to have Telnet or TN3270 installed on your Macintosh. If you are still running System 6, TurboGopher will only save a Telnet session file rather than launch Telnet; this is another good reason to run System 7 on your Mac. Following the Macintosh guidelines, TurboGopher starts out displaying text in the standard application font. The Font and Size menu items let you change font appearance for documents and lists. TurboGopher will attempt to display phone book lookup results in a monospaced font (Monaco), effectively ignoring your font selection. It IS possible to change the font of a phonebook results lookup window after it has been opened; however things will not line up correctly and generally look icky. Advanced Navigation Features If you double-click on a folder while holding down the Option key, TurboGopher will re-use the current window instead of using a new one to display the folder con tents. Some folks like to do this to prevent screen clutter and keep from having too many open windows. You may use the up and down arrow keys on your keyboard to move up and down lists in windows. The Enter or Return key opens an item (and is equivalent to dou ble-clicking on an item). Use of the arrow keys is not supported on the Mac Plus.... (sorry). You may also type the first letter (or first few letters) of an item name in a list and TurboGopher will highlight the first item it finds that matches (sort of like the Finder does in list views). The Find menu item is available to locate an item in a list: just type a string in the Find dialog. Find can also be used to search for text in open document windows, by the way. To delete any item in any list window, click once on the item to select it. Then choose `Delete Item' from the Gopher menu. TurboGopher remembers the last full-text index search (item with question-mark icon) you used. If now you hold down the Option key and double-click on a word in a document window, TurboGopher will query that search service for all documents containing the word you selected. You must use a search service before this pseudo hypertext capability is accessible (TurboGopher needs to know which search ser vice to use in your hypertext query). The Recent menu lets you go back to any directory window you have viewed during your Gopher session. The left and right arrow keys may be used to go to previ ous or next windows; they are equivalent to traversing the recent menu upwards or downwards. Note that the Recent menu only remembers directory windows you visit. Document or phone-book lookup windows are not listed. If you close a window by clicking in its close box while holding down the Option key, then (like Finder) TurboGopher will close all open windows. If you click in the close box while holding down the Shift key, then the cache of information about that window will be released. So the next time you open this window from the recent menu, TurboGopher will have to fetch the contents over the net once more. Of course, if you hold down both Option and Shift keys then all windows are closed and all caches released. Caches are explained below. For the terminally curious, if you hold down the Shift key while single clicking on an item in a list, TurboGopher will display the item's Gopher selector string, host name, port number and Gopher+ baggage in the status pane. This is a quick peek. If you want to copy-paste this information, use the Get Attribute Info menu item found under the Gopher menu; the information you want is in the +INFO attribute (the first line). If wish to fetch an item that is a Macintosh file (binhexed file, shows up with the icon of a disk), but don't want TurboGopher to automatically dehex it as it fetches, you may force TurboGopher to fetch the document as a raw, unprocessed file. To do this, hold down the Control key while you double-click on the item. For folks using DownLine or other applications for de-binhexing and de-archival, this is one way to do what you want. Authenticated Servers TurboGopher now supports AdmitOne Authentication for Gopher+ servers that restrict access to information on a per-user basis. When you attempt to open a direc tory that has such access restrictions, TurboGopher will prompt you for your user name and password on the restricted-access server. TurboGopher and the server negotiate to establish a valid ticket, and subsequent requests to the server use a new ticket for each request. This means that your password is never sent over the net work, and that you only have to enter your password to get the initial authorization. Bookmarks and Bookmark files If you formulate a search or find a folder you would like to come back to quickly later, you can use the Set Bookmark menu item to save your place. Set Bookmark saves a reference to a folder, file, search, terminal session... in fact any Gopher item. If nothing is selected in the front list window, a bookmark is made for a folder representing the window. Try it, it's easier to do than to say. Bookmarks are placed in a special Bookmarks window. Use the Show Bookmarks menu item to open the Bookmarks window. Bookmarks are remembered even after you quit TurboGopher; contents of the Recent menu are however forgotten. It might be useful to consider the Bookmarks window to be a special kind of `work sheet' or `construction area' for you to build your own scratch-pad of useful gopher resources. You can save your Bookmarks window or ANY list window or ANY selected item as a Bookmark file (`Save as Bookmark FileI' menu item). Bookmark files are small and may be exchanged with other Gopher users if you wish. The `Import Bookmarks' menu item lets you read a Bookmark file right into your Bookmarks window. The `Open Gopher Bookmark File' menu item on the other hand, places the contents of a bookmark file in a window of its own. The latter is equivalent to double-clicking on bookmark files from the Finder. If you launch Tur boGopher by double-clicking on a bookmark file, it will not immediately connect to the Home Gopher server, but will open the Bookmark file instead. So you may use bookmark files as custom Home Gopher servers. Item Attributes More sophisticated Gopher servers (referred to as Gopher+ servers) can provide you with information about any selected item, much like the Finder's Get Info menu lets you get information about a Mac item. Use the `Get Attribute Info' menu item of the Gopher menu for this. Canceling Slow Network Operations TurboGopher executes most requests in a few seconds. A document or directory is displayed as it is received. However, if a server is especially slow or busy or if you have started to fetch what appears to be a very long and uninteresting document or directory, you may cancel the fetching process by closing the document or directory window. Once you have closed the window, TurboGopher assumes that it does not need to fetch the remainder of the item. Note that when you fetch a file that is saved to your Mac disk, TurboGopher displays a window with the status of the fetch. Again, to cancel the fetch operation you can simply close the window. A more extreme method for canceling a network operation is to choose the Cancel All Requests item from the Gopher menu (or by holding down the command key and typing a period). If you open a file or folder by double-clicking while holding down the Option key (to conserve screen space as described earlier), TurboGopher will also cancel the running network transaction if one is outstanding; and this is in fact exactly what you want over low speed connections such as SLIP. Both these features make TurboGopher seem faster and more responsive over SLIP links. TurboGopher runs happily in the background (if you wish to fetch something very large or from a very slow server). You may also fetch an item (file or directory) while another one is still being fetched, since TurboGopher supports multiple con current streams. Command-Period will cancel all active streams. Starting and Configuring TurboGopher The `Start Gopher' item under the File menu opens a window containing the initial directory fetched by connecting to your home Gopher server. TurboGopher is pre configured to connect on startup to one of the University of Minnesota's Gopher servers. You may change your home Gopher server by using the `Configure' item under the Setup menu. To do this you need the full Internet domain name and port of the desired primary server. You can configure TurboGopher to use one of two alternate primary servers. This distributes the load for a campus over two different (but equivalent) primary servers and it prevents a single point of failure. TurboGopher will randomly try one of the two and then try the other if the first is down. If you don't understand this, you don't have to bother with it or with the Another Gopher item under the File menu. TurboGopher attempts to speak Gopher+ to all Home Gopher servers, and then backs off if it sees that the server does not have Gopher+ style items. This is a bit presumptuous, and it confuses some servers. These servers should be revised in the future so they don't confuse so easily. If configured to connect to a Home Gopher server and nothing is displayed, you may want to try using the `Another Gopher' item under the file menu and making sure that the gopher+ check box is NOT checked. TurboGopher Options You can use the Options item under the Setup menu to customize some behaviors of TurboGopher. The Single Directory Window checkbox makes TurboGopher recycle its windows whenever you open a new directory (and not just when you hold down the Option key). Use the ISO Latin-1 checkbox if you need to use Latin/Romance language charac ters. Many sites in Europe have a need for this to display their national characters. If you visit European gopher servers and the accented characters don't display cor rectly, you may want to choose this option. The Extended Directory listings checkbox is disabled; it's for a future feature. If you visit Gopher servers with Japanese or Chinese documents, and you use a Jap anese or Chinese font while running System 7.1's WorldScript, Gopher will display Japanese SJIS (MS Kanji) or traditional Chinese characters (BIG-5). When you save documents, TurboGopher normally saves them as MacWrite II TEXT documents (we like MacWrite II). If you'd rather save them as TEXT docu ments of your favorite word processor or editor, use the appropriate `Set' button in the Options dialog to select the word processor of your choice. With Gopher+ servers, items may be available in a number of alternate views. When fetched, these alternate views are typically saved as a file to be viewed by a Mac application. If you are a network/macintosh cowboy, you can change the Mac application and Mac filetype associated with a Gopher+ alternate view.... this changes the helper application that TurboGopher asks the finder to launch to view the document. Double-click the view in the list and you will be presented with a dialog in which you can select the application to be used as a TurboGopher helper. By the way, you won't be able to do this if you're running System 6... Miscellaneous TurboGopher saves or caches the item lists for all visited directories (even when windows are closed). Since these lists don't ever have to be fetched from the server again, this dramatically increases speed when choosing any item from the Recent menu. If TurboGopher runs out of memory it will first inform you and then auto matically release the caches for closed windows. As mentioned earlier, you may selectively release a directory's cache by closing the window while holding down the Shift key if you wish more control over memory. TurboGopher stores what it needs in a file called TurboGopher Settings; this lives in the Preferences folder, which is found in your System Folder. If you drag this file out from the Preferences folder and place it in the same folder as TurboGopher, it will still be found and used. For Experts This section describes features of TurboGopher that either network administrators or Gopher weenies would want to know. Waking up TurboGopher. When TurboGopher starts up, it looks for the TurboGopher Settings file first in its own folder, and failing that in the Preferences folder. If it finds the TurboGopher Settings file, it uses the settings and resources (more on this later) it finds inside. If it cannot find the file, it creates a new one in the Preferences folder based on its internal defaults (more on this later too). If there are any bookmarks the user had set in the Bookmarks worksheet window, the Bookmark worksheet window is opened. If the user had deleted all bookmarks, the Bookmarks window is not opened (the contents of the Bookmark worksheet window are remembered via the Settings file). Next TurboGopher looks to see if the user has launched it by double-clicking on a Bookmark file (or by dropping a Bookmark file on it under System 7). If this is so, the contents of the Bookmark file (more on format later) are read and placed in a list (directory) window; in this case TurboGopher will not make a connection immedi ately to the Home Gopher Server. If TurboGopher was launched directly (not via a Bookmarks document) then it con nects to the Home Gopher Server, retrieves the contents of its top directory, and places this in a list (directory) window. At this point, TurboGopher is fully up and running. Next we need to describe the various parts alluded to above. The Home Gopher Servers Out of the box, TurboGopher comes preconfigured to connect AT RANDOM to one of the two primary gopher clones (gopher.tc.umn.edu or gopher2.tc.umn.edu; the former is also known to old Gopher folks as gopher.micro.umn.edu) at the Uni versity of Minnesota. It must be emphasized that these are NOT a primary and sec ondary server: they contain exactly the same information and changes made to one are automatically propagated to the other. If the first connection attempt fails, Tur boGopher will try the next. This has two purposes. First, initial connections are very reliable for users and since we serve as the world's Gophermeisters, this is good. More importantly for us, the randomization ensures that users will spread the load over our two small, economical, primary servers (Mac IIci's). If your campus' primary Gopher server experiences heavy use, we suggest you make a clone too. The user may of course reconfigure for only one primary server. TurboGophe