CWIS Access

Antony Barry & Deidre Stanton

Paper submitted to Australian Library Review. issue on Australian IT and Electronic Library Developments.

It is questionable in the long term whether there will be such a thing as a campus wide information system (CWIS) any more there is a telephone system. While there are a group of services which form a core of what we now regard as a CWIS the degree to which they need to be, or in fact are, organised as a monolithic service is in question as network publishing capability grows. Currently developments in such services are dominated by the Internet Gopher software developed at the University of Minnesota and much of the discussion below revolves around this.

1. What is a CWIS?

Different areas of a campus view a campus wide information system in different ways. For some it is a mechanism by which the management of the University delivers information to the university community. For others it is a mechanism by which information available from elsewhere can be delivered to the desks of individual students and academics. For a few, but increasing number, it is also a mechanism by which individuals and groups on a campus can publish their research and fruits of their scholarship to the network community which increasingly is becoming synonymous with the world community of scholars.

It is also not always clear when viewing a campus from the network whether there is a "Campus Wide Information System" in any official sense at all. With the increase in electronic publishing of internal directories and policy documents and network access to all of this it is open to any individual with a workstation to produce a coherent view of these resources which can be functionally equivalent to many less well stocked "official" systems 1. While in the longer term it is likely that there will be one information server on a campus to give a view of "official" information this information may be sourced from information servers in many areas. The view provided may only be one of many . Whether the library is included in the "official" CWIS or not will not affect whether Library and official information are available and only to some extent the view of that data which is available.

1.1. CWIS as Guide to the campus

As a provider of internally oriented information services CWIS systems are able to provide access to day to day administrative material such as directories, procedural manuals, minutes of administrative committees. They assist in providing a store of the campus 'memory' for that material normally duplicated with varying degrees of completeness and currency on many desks and shelves throughout the organisation.

The provision of a central source of authoritative data, equally accessible to all has clear advantages for administration. A mechanism where news, circulars and other general public announcements can be disseminated as widely as a mailing list or notice board but with far less effort, clear advantages ensure.

1.2. As gateway to the net

The view that a campus information systems can be used to provide organised access to networked resources of interest to the users of the campus still holds some validity but increasingly we will see individual views of the net being published. The craft of the bibliographer and the scholar gains new emphasis in an environment where vast quantities of semi organised material are available. While Librarians might see the world through the blinkers of a classification scheme 2 individual academics can mount what are in effect 'live' bibliographies which deliver the documents and services they describe in a manner of interest to them and their colleagues 3. This extends even to personal publication list where David Green and ANU academic has not only published some of his academic work via WWW 4 but also his C.V. and a linked bibliography of his publications 5. We will see more of this.

1.3. As a publishing mechanism

Desktop publishing is an unfortunate misnomer. Desktop publishing software, far from providing the ability to publish, only provides the facility to do desktop typesetting and formatting. Material can be assembled in a form which can then be passed to a publisher for printing and later distribution. The software created to run current CWIS systems is in the public domain and will run on small individual workstations or microcomputers. The facilities of the CWIS are available to anybody with a networked device on their desk . Through this software they are able to do true desktop and network publishing.

Far from there being a single monolithic CWIS at each institution each individual can publish direct from the desk to their networked colleagues. It is likely that the amount published in this way will be much greater than that published on paper and that there will be an explosion in preprint and ephemera. The future task of the library profession could well include that of managing and simplifying access to networked publications. But also as central publishing organisations weaken in their ability to influence, control and filter what gets published there will be scope to assist individuals in their network publishing endeavour. If libraries help provide a view of the network for their own internal users there is a converse responsibility for the library to provide a coherent view to the inquirer from outside the institution who is seeking publications of the University. There is scope for the Library within an institution to become a significant player in the publishing process.

While the internet mainly reaches the academic and research sector this is rapidly changing in the US as there part of the network starts to commercialise. Even in Australia despite resistance to opening AARNet up individuals are starting to spread access.6. The limitations in the distribution of electronic publishing are likely to rapidly decrease.

A more difficult problem is the blurring of the concept of a document. If information is provided electronically from a server it may be that the author, for some types of documents, will wish to, and will, modify the document to take into account additional information and ideas as they come to hand. Where a document is changing in an essentially continuous fashion it may become difficult even to talk of editions of a document only the content at a particular date and time.

With the cost of disk now an order of magnitude cheaper than paper, with networks making remote information easily available it is now quite feasible to publish the source data upon which the conclusions and reasoning in a paper are based so that they can be checked. Indeed, in an academic environment, to pressure to make working data available can be quite great. The networks will increasing see data of this kind and certainly in the fields of biology and space science we are seeing this now.

The end result of these trends is the creation of a central organised source of documents and data serving the needs of the host institution including both material from within the institution and outside. In short a library.

2. History

The early history of the introduction of CWIS systems has been summarised by Judy Hallam 7. In retrospect the development of what we know as campus wide information systems was inevitable. The initial spread of terminals for academic and administrative purposes provided the groundwork for what was to come. While these remained linked to unconnected computers the synergy required for the development of the CWIS could not occur. The revolution introduced by microcomputers and the need to network them to achieve economies in the use of expensive peripherals led to a rapid increase in networking on campuses. Once sufficient of these local networks and machines supporting a large terminal load became interconnected through a common campus network the framework was there to consider the possibility of providing common information services across the network as a whole to all devices. No doubt this was driven by enthusiasts and visionaries but the prospects or reduction costs, particularly in central printing services, and courier costs must have been attractive to administrators.

While the these campus networks remained free-standing, and while no standardised easily used information retrieval software existed at an affordable costs these service could only look inwards to the problems of providing internal information to local clients and they followed the model of delivery natural to the times and for which software could be made available, a central service provided from a mainframe.

2.1. Workstation level solutions

Some sharing of software did occur in this period and the advantages of a client-server approach was recognised. Both Princeton network news (PNN) and MIT's Techinfo were developed in this period but both still largely served internal needs and had not made the conceptual leap required to support an externally oriented service. This had to wait for the Gopher team at Minnesota.

It is ironic that the developments at Minnesota would appear not to be based upon a desire to achieve what they eventually produced but motivations to create a campus information systems which could be achieved on multiple workstations and distributed across multiple workstations in order to cut costs rather than a mainframe approach 8. The introduction of one simple concept, that of allowing one server to chain to another, fundamentally altered the concept of what a CWIS could be and do. It generated the capability of multiple sites sharing information and being able to link information services together into one integrated whole without keeping multiple copies of the data or extensive coordinating mechanisms between sites. This could be done between not just within institutions, but between institutions as they became linked through the Internet. With the extension of this linking concept to support gateways to other network tools like anonymous FTP archives, Wide Area Information Server (WAIS) databases, remote terminal connections, usenet news an extremely powerful set of tools was created to deliver internal information and information from other institutions. By putting this software in the public domain, Minnesota, not only followed the traditions of the Internet but they also served their own interests by ensuring that information published by other institutions using their software became accessible

The shift to an outward view for CWIS systems could not have evolved until these software developments had occurred. Even before then, interest had arisen and was stimulated by the CWIS-L mailing list on BITNET, along with the PACS-L mailing list was one of the earliest e- mail lists serving the campus information service community. This group is still active and it is mirrored to Usenet news 9

2.2. Outmoded Concept?

In many ways the concept of a CWIS is already outmoded even though still actively pursued 10. Like the distinctions between LANs, WANs; networks and internets which was once clear, and are now largely merged as far as the individual user is concerned; the once clear distinction between internal services on a CWIS and external services have similarly been smoothed away. It is for good reason that the terms "GopherSpace", "the net" or "the Web" are used. The services of the Internet are merging into one coherent whole, into a global information service.

3. CWIS Content

As campus information systems are still relatively new, there is little in the way of standardisation of their content and menu structures However certain elements are becoming common to all such systems. There is some consensus as to what information a campus should provide over it's network. These are discussed below.

3.1. Directories

Directory information is very common, typically e-mail and phone directories. Organisational problems will arise on campuses where the compilation of the phone directory and the phone service has been divorced from computer operations and the production of a phone book may not create data in a form suitable for placing in an automated directory. Historically the responsibility for phone systems may be entrenched in an area of a University's administration loath the lose control or be merged with directories of other types.

Not only are there likely organisational problems, the standards for directory compilation and access are not yet settled with complete clarity. From the OSI world comes the x500 standard. Currently the University of Adelaide is providing access to this service for AARNet users 11. Many institutions have opted for an older 'standard' the CSO server which is reputed to be easier to install and which is better integrated into gopher 12. Still others have opted for a seemingly simplistic method of using WAIS software to index their directories as a text file 13. While this may seem quick and dirty, it works. The various approaches used with Australia have been tabulated at ANU 14.

A CWIS will certainly contain other listings and directories but there are more site dependent.

3.2. News

Many universities have mounted their internal news, typically their campus newspaper and other types of newsletters. A more interesting trend has been the use of systems to automate parts of the maintenance of such systems, either by links to local Usenet newsgroups where contributions may be placed and then linked to the CWIS or through maintenance of individual directories via e-mail.15

3.3. Service information

Another area which is almost universally provided on campus wide information systems service information for service bodies, such as libraries and computer centres. These service centres usually have publicity material available promoting their services readily available for inclusion on a CWIS. This may include copies of forms which can be e-mailed or printed and filled in when applying for various services. It will normally also provide direct links to some online services such as Library catalogues.

While integrated library management systems trail behind the mainstream of network developments as they continue to be provided via telnet links rather than by using the newer protocols of Gopher or Z39.50. A number of groups are making movements in this area, notably the Norwegian library systems which provide direct Gopher access to their bibliographic records 16 and the work of the Z39.50 Implementors Group 17 which is leading to the deployment of native Z39.50 catalogues rather than those using the now outdated 88 version of the protocol utilised by WAIS software.

3.4. Gateways

The origin of CWIS systems in the provision of centralised sources of internal information leads to a conflict in the provision of access to material sourced from outside the institution. The very success of gateway systems in presenting information in a seamless fashion can easily lead to confusion in the sources of the data. Problems encountered in access to, or accuracy of, external sources can easily be blamed upon the provider of the gateway to the service rather than the service provider themselves. This has been countered in a number of ways.

The developers of Gopher and World Wide Web client software have added item inspection facilities to display the linking information to reveal the source of the information. With the Gopher+ protocol this has been extended a step further so that the server will display additional information about the source such as the size of the data, the date it was last changed and the name and address of the administrator of that part of the system.

A number of simple expedients have been followed by those creating Gopher menus. Two extreme approaches for the top level have been followed with every variety in between. One approach is to only have one menu item at the top through which ALL external information can be obtained 18. The second is the reverse. All menus but one go to external or "unofficial" sources with one going to internally provided material 19. Both approaches have problems. Where the real desire may be to isolate "official" university material from other material, finding a dividing line is not easy as more groups on a campus that are remote from the central services staff publish their own material. Libraries in particular tend to be less concerned with the source of the information and more concerned with this subject area and hopefully, the usefulness of the information.

3.5. Interactive Facilities

The direct deployment of interactive facilities when users can engage in a dialogue with a computer via a CWIS system which was available with the earlier deployed software is not yet prevalent on current systems. Various kludges have been employed, but with little general acceptance. For instance it has been easier to exit to an interactive terminal session to achieve this end. When Gopher+ becomes more generally deployed it will support query/answer sessions and this may change. It is not yet possible to cleanly interface, for example, an automated enrolment systems into a CWIS.

4. Management

The management of a CWIS has still not settled to a generally accepted model. Certainly the full panoply of distributed information servers in a campus will be essentially unmanageable in all but a technical sense set by the logical and physical needs of the network. The degree of management will need to be strongly circumscribed by the needs of personal and academic freedom.

For 'official' organisational information some patterns may be emerging. Firstly, input is needed from those whose expertise lies with the media over which the system travels - the computing staff. So far their input seems to have been predominant. Secondly, those groups concerned with the presentation and organisation of information , librarians and staff in public information areas, need to provide guidelines. Librarians in particular are developing a role in the provision of campus wide information systems.

Lastly, the actual information providers need to be involved, the compilers of directories, handbooks, agendas and minutes. To date their involvement seems minor, but their role must become a central in the long term if these systems are to provide current, useful information.

The need for coordination that is associated with the official electronic publication activities of a University is also to be found with the less formal network publishing activities occurring on campus. However this often seems to amount to less, rather than more coordination than that found for official publications. How it is done , the degree of control exerted and the formal organisational structure set up to do this will vary in response the local organisational culture rather than any generic model which can be imported.

4.1. Decentralisation

The layers of administration between the information providers and the end users of the information should be minimised if a service is to run smoothly with clarity in responsibility in service. Fortunately, Gopher software provides for this, even though most sites still seem to be operating a centralised model, where systems staff place material on CWIS systems rather than the information providers carrying out these tasks. There is no need for this to be the case and downward delegation can easily be achieved either through-

4.2. Stylistic Considerations

With a distributed information system, that allows the content to be distributed over multiple computer systems, it becomes important to identify: Because electronic information is usually regularly updated regularly , the need to indicate when the material was last updated becomes more important than with paper sources which, by their nature, are expected to be out of date. The following elements are often seen on information servers. -

  1. A document which explains what each menu is about, who maintains it and when it was last updated. Normally there would be a phone number and email address for the person responsible.
  2. Each menu item may give an indication if the information is local or provided by an external source.
  3. An indication of the format and size may be given as a warning on difficulties in downloading or viewing - for instance for platform specific files.Various examples of this approach can be found in documents on the Murdoch University Gopher 21 and the University of Canberra CWIS 22.
There is an advantage in each document containing information on it's provenance. Normally this is displayed in a header or template, as this is the first part of the document to be displayed when viewed online.

This header might contain -

  1. Title
  2. Author's name, email address and phone number
  3. Date compiled
  4. Date of expiry if any.
  5. Alternate formats available.
  6. Who to contact for further information and how as this may differ from the author of the document.

4.3. Format and Layout

Gopher systems are able to deliver documents in a single format whether this be binary, graphical, sound or text. However most documents are platform specific. As it is not only impossible but undesirable to restrict all workstations to the one platform type at a single institution, the limitations of the lowest common denominator workstations determines what can be delivered. While we are still limited to terminal interfaces, only unformatted text is common to all. Formatted material in postscript or RTF formats is almost universal but must be downloaded first before it can be viewed or printed.

There are further limitations on the type of textual material which can be delivered due to the different characteristics at the client end. To ensure global accessibility the following guidelines need to be followed -

  1. Tab characters must not be used.
  2. Text line lengths should be limited to 72 characters
  3. Only standard printable ASCII characters should be used. vt100 graphics and other terminal escape sequences will destroy a display on some clients for instance.
  4. Layout should be done on the assumption that a fixed width character set will be used for display.
  5. Documents should be limited in length to 200 lines if they are to be displayed on the screen rather than downloaded.

5. Software

Many of the monolithic CWIS systems still exist and can be accessed over the net 23 , and are listed by Judy Hallam 24 there appears to be no more of such systems being installed. Not only has Gopher swept the pool but it has replaced or runs in parallel with existing systems. Some sites using the PNN software have switched to gopher and Yale has virtually dropped it's Enterprise system in favour of Gopher. Only one of the original system seems to still have vitality and that is MIT's Techinfo which shares Gopher's client-server advantages and has explicit information provider updating capability that Gopher lacks 25.

5.1. Gopher

In it's design, Gopher is an internet lego set, as it is designed so that a menu item on one Gopher server can connect at any level of any other Gopher server. This capability has enabled nearly all Gopher servers to cross connect to each other in a single entity in what is described, not unreasonably, as "GopherSpace". Indeed this concept has been taken one step further by implementing a gopher client as a multiple users dungeon game (MUD) called MOO. Users can access material linked by Gopher, moving from menu item to menu item, as if they were geographical areas, and also exchange messages with other simultaneous users of the same material within common meeting areas 26.

A gopher server maps the file system of it's host computer into the menu structure it displays, each directory presenting a separate menu. Files in each directory are presented as retrievable menu items. Special files in each directory can add links - gateways - to items from other servers. The Unix server is best developed and can offer menu items which provide gateways to -

Gopher has an extensive range of capabilities to deliver files to local clients, although what the client can do with the files varies with the operating system. While this feature is mostly used to deliver displayable text files it can also be used for -

The huge expansion in the amount of material available via GopherSpace has led to extensive work to provide an indexing capability. Two software tools have been developed, Veronica from the University of Nevada and Jughead from the University of Utah 28. Both tools provide for multiple levels of keyword indexing of menu items ranging from global, to just the directories on the local Gopher server. Both provide for Boolean search capability. Their limitations stem not from the software, but from the choice of words used by local administrators to describe their menu items. Detailed discussion on improvements expected in this area are beyond the scope of this paper.

Server specific indexing systems exist for Next systems which have indexing built into the operating system and the gn, a Gopher and WWW server, has an item level indexing systems built in 29 as has the ftpd server for the Mac 30.

The first moves to integrate Gopher into Integrated Library Management Systems have been reputed to have begun. Both VTLS and NOTIS are alleged to be working on a capability to build gopher into their OPACs so that the full text of an item can be retrieved via the Gopher protocol from within the OPAC.

There is an active news group comp.infosystems.gopher 31 and a variety of specialised mailing lists, one of which GO4LIB, is of particular interest to Librarians , as it supports discussion of Library related Gophers. A fairly complete list of Gopher servers is set out below and these may be accessed via the ANU Library Elisa gopher.32

Institution                                  Host
____________________________________________________
AARNet Archive Server                        archie.au
ANU - COOMBSQUEST Soc.Sci & Humanities
     Inf.Facility at ANU                     cheops.anu.edu.au
ANU - Electronic Library Information 
     System at ANU                           info.anu.edu.au
ANU - Research School of Biological 
     Sciences (RSBS) Biology Gopher          life.anu.edu.au
ANU Campus Information System                cwis.anu.edu.au
Austin Hospital PET Centre Gopher            pet1.austin.unimelb.edu.au
Australian Defence Force Academy             gopher.adfa.oz.au
Australian Environmental Resources 
     Information Network (ERIN)              kaos.erin.gov.au
Australian National Botanic Gardens          155.187.10.12
Centre for Design at RMIT                    daedalus.edc.rmit.edu.au
CSIRO Australian National Telescope Facility crux.rp.CSIRO.AU
CSIRO ITS Branch Information Server
(commsun.its.CSIRO.AU)                       commsun.its.CSIRO.AU
Curtin University of Technology              info.curtin.edu.au
Flinders University                          frodo.cc.flinders.au
Griffith University                          gopher.gu.edu.au
James Cook University                        marlin.jcu.edu.au
La Trobe University                          gopher.latrobe.edu.au
Monash University Gopher                     cwis.monash.edu.au
Murdoch University Library                   infolib.murdoch.edu.au
Queensland University of Technology          Gopher.qut.edu.au
University of Adelaide ITD 
     Experimental Gopher                     jarrah.itd.adelaide.edu.au
University of Canberra                       services.canberra.edu.au
University of New England - Northern Rivers  alsvid.une.edu.au
University of New South Wales                usage.csd.unsw.oz.au
University of Queensland                     brolga.cc.uq.oz.au
University of Southern Queensland            helios.usq.edu.au
University of Sydney Law School              sulaw.law.su.OZ.AU
University of Sydney Library                 gopher.fisher.su.oz.au
University of Sydney, Australia              gopher.su.edu.au 
University of Tasmania                       info.utas.edu.au
University of Western Australia              uniwa.uwa.edu.au
University of Western Australia 
     Virology Gopher                         virus.microbiol.uwa.edu.au
University of Western Sydney                 ob1.uws.edu.au
University of Western Sydney, Hawkesbury     hotel.uws.edu.au
University of Western Sydney, Macarthur      tscc.macarthur.uws.edu.au
University of Western Sydney, Nepean         gopher.nepean.uws.edu.au
University of Wollongong                     gopher.uow.edu.au
Victorian Institute of Forensic Pathology    brain.vifp.Monash.edu.au

5.2. WWW

Providing a superset of Gopher functions the World Wide Web system can be described as Gopher on steroids. World Wide Web is a creation of the physics community at CERN in Switzerland although lately NCSA at Illinois has done much of the development. It provides hypertext links across the Net between documents mounted on WWW servers.

A WWW or Web document is annotated in Hypertext Mark Up Language (HTML) a derivative of SGML 33. The tabs used in the document can specify general format information about a document such as heading levels, text style and paragraphs. Graphics can be included in the text. Links to other documents can be embedded in the text as Uniform Resource Locater (URL) references. These other documents can be of all media or multimedia types. They do not need to be included on a WWW server but can, for instance, be linkages to documents available from anonymous file transfer (FTP) sites or via Gopher All material in GopherSpace is therefore available to users of the Web.

Despite it's power, the use of WWW software has been limited until very recently because of two factors.

Firstly, client software was only available for either dumb terminal interfaces that are unable to access formats other than text or for X-Windows systems that are financially out of reach of most potential users.

Secondly, WYSIWYG editing software was unavailable. This situation is improving with converter software written for formats like RTF which can be output from a number of word processors such as Microsoft Word; and the release of specialised editors.

The use of WWW on the Net is now growing much faster than that the use of Gopher. Documentation for WWW is more extensive than that available for Gopher 34,

It's development is supported by a newsgroup comp.infosystems.www and a number of mailing lists.

At the time of writing there were only nine Web server in Australa most of them at ANU 35.

5.3. Mosaic

Mosaic software is the Swiss army knife of the internet. The software is produced by the National Centre for Supercomputing Applications (NSCA) at Illinios. Mosaic is written to be able to 'read' all formats on the Net from the most heavily used such as telnet and FTP to the most specialised such as Techinfo. With the release of the Mac and PC versions at the end of September the applicability of WWW as a network publishing mechanism has been greatly enhanced. Not only is client software available for all major network platforms, Unix 36, X 37, Windows 38 and Mac 39, but these clients greatly help the editing process as they can view document being developed locally independently of the existence of a server to deliver them.

6. Future

The likely developments in the future are for-

November 1993.

A. Barry and D. Stanton

Footnotes

Many of the footnotes below contain Universal Resource Locater references (URLs). These provide what are effectively network call numbers for information resources. They indicate which computer holds the information, at what location, and how it can be retrieved. The URLs can be directly entered into Mosaic software to retrieve and view the document described.

1 At the time of writing the National Library did not offer a 'CWIS' of it's services but it was quite possible to provide these from an external point through a gopher server. gopher://info.anu.edu.au:70/11/OtherSites/au

2 ANU's Electronic Library entry on it's Elisa gopher is organised in this way, gopher://info.anu.edu.au:70/11/elibrary

3 Dr Matthew Ciolek through the CoombsServer offers a coherent view of networked material in the Social Sciences, gopher://cheops.anu.edu.au:70/1

4 David G. Green 'Landscapes, cataclysms and population explosions'. Mathematical and Computer Modelling, 13,6 (1990).:75-82. http://life.anu.edu.au:80/edmonton.html

5 David G. Green - PUBLICATIONS , http://life.anu.edu.au:80/people/dgg/dggpapers.html

6 Zik Saleeba (Comp.) 'Australian Public Network Access FAQ v1.3'. Release version 1.3 -93/08/04 http://life.anu.edu.au:80/australia/publicfaq.html

7 Judy Hallam . 'Campus-wide information systems'. Advances in Library Automation and Networking, Vol 5. 1993. gopher://gass.cc.uow.edu.au/00/Library/tech/hallman.txt

8 Rich Wiggins, 'The University of Minnesota's Internet Gopher System: A Tool for Accessing Network-Based Electronic Information', The Public-Access Computer Systems Review 4, no. 2 (1993): 4-60 (to retrieve this file, send e-mail messages to LISTSERV@UHUPVM1 or LISTSERV@UHUPVM1.UH.EDU containing GET WIGGINS1 PRV4N2 and GET WIGGINS2 PRV4N2)

9 news:bit.listserv.cwis-l

10 ANU Information Technology Directions Statement. Campus Wide Information System Committee Report 1993, gopher://cis.anu.edu.au:73/11/Admin/Documents/ITD/ITD4

11 University of Adelaide X500 link, gopher://jarrah.itd.adelaide.edu.au:70/11/Dirs/X.500 and AARNet and also National Directory Services project .Final Report, 1993?, 68p., gopher://plaza.aarnet.EDU.AU:70/00/ftp/projects/directory-services/report/REPORT.ascii 12 ADFA maintains a CSO directory . gopher://qi.cc.adfa.oz.au:105/2

13 ANU used WAIS for both its phone and email directories; gopher://cis.anu.edu.au:70/11/ANU-dir

14 ANU phone & email directory collection; gopher://info.anu.edu.au:70/11/Refbooks/phone

15 University of Canberra supports calendar information in this manner ; gopher:///11/News/Calendar/Events

16 Norwegian National Library; gopher://gopher.bibsys.no:70/1

17 Draft rfc for z3950 over tcp/ip. gopher://libmac1.anu.edu.au:70/00ftp%3aftpd%20gopher%3aNetwork%20Tools%3aWAIS%3adraft%20rfc%20for%20z3950%20over%20tcp/ip

18 The ANU campus information system limits external access on it's top menu to one item for externally provided data and another for external links.; gopher://polly.anu.edu.au:70/1.

19 The Hong Kong University's gopher server only provides one link to it's services. gopher://info.csc.cuhk.hk:70/1

20 Prentiss Riddle gmail ftp://info.anu.edu.au/gopher/Unix/GopherTools/gmail-1.01.shar

21 gopher://infolib.murdoch.edu.au:70/1

22 gopher://services.canberra.edu.au:70/1

23 These can be access via hytelnet which is available for all platforms even WWW http://info.cern.ch/hypertext/WWW/HytelnetGate/Overview.html.

24 Judy Hallam List of CWIS systems ftp://sunsite.unc.edu/pub/docs/about-the-net/cwis/cwis-l

25 Techinfo link http://info.cern.ch/hypertext/WWW/Daemon/TechInfoGate.html

26 MOO software and information: ftp://info.anu.edu.au/gopher/GopherMoo/

27 The Australian National Botanic gardens has linked Oracle databases to its gopher server gopher://155.187.10.12:70/00anbg-data/ibis/gopher-dbms-gateway.txt

28 gopher://cheops.anu.edu.au:70/11/Socioinf-query/JugheadVeronica 29 gn documentation ftp://info.anu.edu.au/gopher/Unix/gn

30 ftpd link or indexing explanation. ftp://archie.au/micros/mac/info-mac/comm/Network/ftpd-220.hqx

31 news:comp.infosystems.gopher

32 Access to Australia gophers is obtainable at gopher://info.anu.edu.au:70/Othersites/au/

33 A beginners guide to hhtp. http://www.ncsa.uiuc.edu/demoweb/html-primer.html

34 WWW documentation is available at http://info.cern.ch/hypertext/WWW/TheProject.html

35 Australian World Wide Web ervers http://life.anu.edu.au:80/links/ozweb.html

36 Link to Unix software http://info.cern.ch/hypertext/WWW/Clients.html

37 Link to X Mosaic ftp://archie.au/NCSA/Mosaic/

38 Windows Mosaic ftp://archie.au/PC/Mosaic/

39 Mac Mosaic ftp://archie.au/NCSA/Mac/Mosaic/