Common Questions and Answers about veronica, a title search and retrieval system for use with the Internet Gopher. The current version of this FAQ can be retrieved through gopher at veronica.scs.unr.edu, in the veronica directory. Archive-name: veronica-faq Last-modified: 1993/01/29. -------------------------------------------------------------------------- List of questions in the veronica FAQ: Q1: What is veronica? Q2: How can I connect to veronica? Q3: So I get a list of titles; How can I get information about the location of items I am interested in? Q4: Why do I need to use 'item descriptor' to get context and host info? Q5: How can I get my server into the veronica database? Q6: How can I keep my server out of the veronica database? Q7: How often is the database updated? Q8: Where can I get the software to run veronica? Q9: Where can I get the veronica dataset? Q10: Why can't I get a reply from a veronica server? Q11: What does "veronica" mean? Q12: Why doesn't my server show up in veronica searches? Q13: Where are the veronica servers? -------------------------------------------------------------------------- Q1: What is veronica? A1: veronica is a service that maintains an index of titles of gopher items, and provides keyword searches of those titles. A veronica search originates with a user's request for a search, submitted via a gopher client. The result of a veronica search is a set of gopher-type data items, which is returned to the gopher client in the form of a gopher-compliant menu. The user can access any of the resultant data items by selecting from the returned menu. A veronica search typically searches the menus of hundreds of gopher servers, perhaps all the gopher servers that are announced to the Internet. At present, there are no "veronica clients" per se; veronica is accessed through normal gopher clients. veronica is tightly integrated with the gopher protocol. The veronica service comprises two functions: 1). Harvesting menu data from gopher servers, and preparing it for use; 2). Offering searches of that database to gopher clients. These two functions are not necessarily provided by the same host computer. Most users and administrators of veronica search servers will not need to be concerned with the first phase of the process. Operators of veronica query-engines can obtain a prepared dataset for use with the query server ( Q9 below ). veronica evolved as a solution to the problem of resource discovery in the rapidly-expanding gopher meta-burrow. At the University of Nevada, there was an outcry for an easy way to find gopher-based information without doing a menu-by-menu, site-by-site search. -------------------------------------------------------------------------- Q2: How can I connect to veronica? A2: veronica must be accessed through a gopher client. Assuming you have a gopher client, use it to connect to a gopher server which offers a link to a veronica server. If your local gopher server does not already have a link to veronica, use gopher to go to the server at gopher.micro.umn.edu ( port 70 ). Choose the menu item "Other Gopher and Information Servers". Choose veronica from that menu. In step-by-step form it is like this: 1. gopher to gopher.micro.umn.edu by typing: gopher gopher.micro.umn.edu 70 Note that this is a unix-client example. Other clients will have a different way of specifying the server. 2. go to the directory named "8. Other Gopher and Information Servers/" 3. select that item 4. go to the item named "2. Search titles in Gopherspace using veronica/" 5. select that item. This item is a gopher link to the veronica directory at the University of Nevada. 6. There are two searches currently (1-21-93) in this directory: "1. Search gopherspace by _SINGLE_ keyword veronica " "2. Search gopherspace by _partial Boolean_ veronica " When you choose either search item, you will be prompted to enter a keyword or keywords to define your title search. There are at present two versions of the veronica search-engine. One version uses the NeXT's Digital Librarian and accepts only a SINGLE word to search for. The second version uses the WAIS index with Gilbert's Boolean extensions; this version can accept limited Boolean specifications. The veronica directory contains a help-file about the format of the Boolean queries. veronica queries are not case-sensitive. The veronica server will return a gopher menu composed of items whose titles match your keyword specification. These items are culled from the menus of (potentially) many gopher servers. As with any gopher menu, you access an item of interest by double-clicking it or hitting the key. If you want to include veronica service on the menu of your local gopher server, you can find the gopher-item-descriptor for veronica from this menu at Minnesota; ask the administrator of your local gopher server to include a link to that item on your local menu. If you DO NOT have a gopher client, you may access gopher and veronica by telneting to one of the telnet-accessible anonymous gopher clients. This is what the Gopher-FAQ has to say about the telnet sites for gopher: *** Here is a short list, use the site closest to you to minimize network lag. Hostname IP# Login Area ------------------------- --------------- ------ ------------- consultant.micro.umn.edu 134.84.132.4 gopher North America gopher.uiuc.edu 128.174.33.160 gopher North America panda.uiowa.edu 128.255.40.201 panda North America info.anu.edu.au 150.203.84.20 info Australia gopher.chalmers.se 129.16.221.40 gopher Sweden It is recommended that you run the client software instead of logging into the public telnet login sites. A client uses the custom features of the local machine (mouse, scroll bars, etc.) A local client is also faster. *** end of inclusion from gopher-faq -------------------------------------------------------------------------- Q3: So I get a list of titles; How can I get information about the location of items I am interested in? A3: Most gopher clients offer a "get information" command or an "item descriptor" menu choice. On a macintosh it usually is "command i"; on a unix curses client it will be an equal sign "="; on a NeXT it also is a "command i". On the PC gopher clients, choose "item inspector" from the "FILE" menu. This key sequence will give information about the current item on the gopher menu. The item-description will include the hostname of the gopher server which provides the item. The item-description also includes the path (directory) of the item on its gopher server. -------------------------------------------------------------------------- Q4: Why do I need to use "item descriptor" to get host and context information? A4: One simple answer to this is that three-times as many lines would be needed on the screen to provide host and path information as well as title information. The menus returned by veronica are often lengthy, even without this information. Since the item- description function is included in most gopher clients, its use makes for more compact screens. There is a deeper answer, of course ... The veronica server is designed to work with the query-type item of all gopher clients. It can only return one set of data in reply to each query, in accord with the gopher protocol. There is no chance for further negotiation between the client and server. The veronica-search server could return visible lines about the host and context of each item, but it would need to do so for all items, tripling the size of the return as mentioned above. The only alternative is to design special gopher CLIENTS which either submit a follow-up query to the veronica server, requesting more information, or which present the item-descriptor in more palatable format. Work is underway to define a client-server negotiation protocol to allow followup queries, using the gopher+ clients and servers. Meanwhile, clients just are not capable of this feature. And deeper yet ... Context means many things ... Is it adequate to know the host, path and type of an item, or do you want to know the last update time, the number of links to that item in the entire gopher network, a unique internet-id for that data item, the names of veronica servers referencing that item, and so forth? The gopher+ protocol and other emerging protocols will allow us to maintain some of this information. The veronica developers at Nevada feel that it is preferable to use these standards as they become available, rather than to define ad-hoc document identifiers, etc. which are specific only to veronica. -------------------------------------------------------------------------- Q5: How can I get my server into the veronica database? A5: Explicit registration is generally not required. The veronica data-collection software will find your gopher server IF it is registered with the Mother Gopher at Minnesota, OR IF it is referenced on the menu of another gopher server which is registered at Minnesota. Of course, the veronica data-harvesting algorithm will not be able to access your server if you have restricted access to your local site. ----------------------------------------------------------------------------- Q6: How can I keep my server OUT of the veronica database? A6: Sometimes gopher-administrators do not want their site to be indexed. The administrator of such a server will need to mail to gophadm@veronica.scs.unr.edu, and request their site not be included. This design is based on the well-founded assumption that any registered gopher server, or a server accessible via the menu of a registered server, is intended to be accessible to users at any internet node. The gopher protocol achieves its utility through this sort of sharing; gopher servers are provided with the "-s" option should it be necessary to implement site restrictions. With the implementation of Gopher+ protocols, the veronica+ protocol will allow gopher administrators to set a "veronica-index-off" flag. -------------------------------------------------------------------------- Q7: How often is the database updated? A7: Heretofore, the database was updated every two weeks. A lot of network traffic is generated by the indexing process. We at U. Nevada thought it not optimal to index more frequently. Most likely, we will index more often in future. The veronica development team is working on a standard protocol for locally-generated indexes, and a standardized method for collating these indexes at a central data site. The post-processed dataset will then be redistributed to sites running veronica query-engines. This approach will greatly reduce the number of gopher-walks and gopher-trees necessary to supply all veronica sites with data. -------------------------------------------------------------------------- Q8: Where can I get the software to run veronica? A8: The NeXT-specific veronica retrieval software can be obtained by anonymous ftp from veronica.scs.unr.edu, in the "veronica" subdirectory. The veronica development team is finishing up the WAIS indexing programs and will have the software available via anonymous ftp hopefully by the end of January 93. The WAIS-veronica distribution will include small modifications to the WAIS index builder, as well as a WAIS-index query engine. The WAIS-indexed version is prefered for three reasons: 1. runs on more platforms. 2. runs faster. 3. allows partial Boolean searches. -------------------------------------------------------------------------- Q9: Where can I get the veronica data set? A9: You can anonymous-ftp the full veronica dataset from veronica.scs.unr.edu, in the "veronica" subdirectory. This data has been processed to eliminate redundant references, to avoid loops in the gopher network, and to remove most data that is known to be highly transient. If you are trying to set up a veronica server, PLEASE take a copy of this data, rather than initiating more gopher-tree processes. --------------------------------------------------------------------------- Q10: Why can't I get a reply from a veronica server? A10: The most common problem is that the gopher client "hangs" because the results of the veronica search includes items of some type not recognized by the client. For example, a search will commence and there is some indication that it is working, like a spinning wheel on a Mac or on a Unix client. However, when the wheel stops spinning the Gopher client says "Nothing available". The problem is that some gopher clients can not handle certain objects (e.g., a PC client is unable to understand a Unix sound file). Many clients are written to treat as empty any directories containing non-standard or unrecognized datatypes. The solution will be in the Gopher+ protocol. That protocol will allow a client and server to have a "conversation" about what types of data the client can and cannot handle. --------------------------------------------------------------------------- Q11: What does "veronica" mean? A11: very easy rodent-oriented net-wide index to computerized archives. --------------------------------------------------------------------------- Q12: Why doesn't my server show up in veronica searches? A12: Most commonly this is because your server was not accessible during the time we last collected data. We will now preserve older data files, and use them in lieu of current data for those servers that cannot be reached during the data harvest. Be sure your server is registered with the Mother of Gophers, or is referenced on the menu of a server which IS registered. Be sure you have not restricted off-site access. If these don't work, send mail to gophadm@veronica.scs.unr.edu --------------------------------------------------------------------------- Q13: Where are the veronica servers? The "partial Boolean" veronica is at the Clearinghouse for Networked Information Discovery and Retrieval (CNIDR), based on a Sun SPARCII platform, using WAIS indexing. The "search for directories only" option is also based at CNIDR. The "simple Boolean" veronica is at the University of Nevada, on a NeXT Cube using the NeXT OS/3.0 indexers (ixbuild and ixsearch). .