John Kirriemuir
Ariadne Web Editor
UKOLN (UK Office for Library and Information
Networking),
University of Bath, Bath
Somerset, UK
[email protected]
D-Lib Magazine, February 1997
Ariadne is a project funded under the Electronic Libraries Programme (eLib), with the main purpose of providing a regular, parallel, Web [1] and print [2] magazine for the UK Higher Education Library and Information Science communities.
In 1993, an investigation into how to deal with the pressures on library resources caused by the rapid expansion of student numbers and the world-wide explosion in academic knowledge and information was undertaken by the Joint Funding Council's Libraries Review Group, chaired by Sir Brian Follett; This investigation resulted in the Follett Report [3], from which one of the key conclusions was:
"The exploitation of IT is essential to create the effective library service of the future".
As a consequence, the Higher Education Funding Bodies in the UK invited proposals for projects which would "transform the use and storage of knowledge in higher education institutions". The original 30 projects, and the 30 or so that would be funded later, would form the Electronic Libraries Programme [4] (otherwise known as eLib); this programme would be managed by the Joint Information Systems Committee (JISC) [5] on behalf of the funding bodies.
The driving philosophy behind eLib is more holistic than simply providing a set of tools and resources; it is one of instilling a "cultural change" in UK Higher Education (HE), making attitudes, confidence and knowledge in the use of network-based technologies more positive and widespread. Projects are encouraged to publicise their work throughout relevant parts of HE, and to work in an open, consortium-based manner where possible. The large number of Universities in the UK are involved in at least one project; most of these are involved with several.
The overall goal of Ariadne is to disseminate information about, and raise awareness of, electronic library developments, resources and services, to the HE and academic library communities.
Ariadne consists of a two-pronged approach to information dissemination:
The Web version is not the only freely available regular publication for the Library, Information Science and Digital / Electronic libraries communities. In the UK, for example, there is the Newsletter of resources at Heriot-Watt [6], the US, however, is the home of most of the rest, including D-lib magazine [7], PACS review [8], the Journal of Academic Media Librarianship [9], LIBRES [10], the Katherine Sharp Review [11] and several others. Some of these are peer-reviewed, some are not. All have their own individual target audiences, styles and scope of content. We originally hoped that Ariadne would not overlap significantly with any of these publications; instead, we would seek to do collaborative work and cross-publicity wherever possible.
Ariadne's overall goal originally broke down into the following roles:
As the project has developed, so the audience has grown to be more international, with many people from the LIS communities of other countries now reading the Web version. With this in mind, an additional role was added:
Ariadne is produced in two institutions:
In addition to the team, the project has an active editorial board of 16 people, who are all based in strategic positions, institutions, or services in the UK Library and Information Science field. The roles of the editorial board are to commission or write articles, interviews and reviews, to publicise Ariadne, and comment on all matters of content, style and accessibility to the team. The members of the board have been very active and enthusiastic in all of these roles, despite having their own full-time jobs to contend with. For example, Emma Worsfold recently observed, through the evening and night, a parallel Web version of an issue of the Daily Telegraph [20] being produced for a future article in both versions of Ariadne.
None of the staff who work on Ariadne do so on a full-time basis; this, combined with two groups based at diametrically opposed corners of the country, makes physically meeting logistically and financially difficult. Our usual approach is to meet at conferences that most or all of us attend for some other reason, though we have already had one (successful) experiment with video conferencing. Because of the large costs involved, it has only been possible to have one combined meeting of both camps and the editorial board to date; this produced many suggestions regarding content and style which are being implemented in both the Print and Web versions. We would hope, when the Mbone becomes easier to use, more reliable and more widespread, to have regular editorial board meetings using this, or some other video conferencing facility.
When the Web version of Ariadne was designed, with issue 1 in mind, the main principle was simplicity. There were some 15 articles in the issue, so a single level contents page was constructed. Graphics were kept as simple as possible, as we were unsure of how much traffic Ariadne would generate; simple icons and graphics, using the eLib colours of blues and white, were constructed using PaintShop Pro.
By the time issue 4 was being constructed, the growth in size of the Web version meant that reading the contents page, as indicated by feedback, was time-consuming. In addition, links to non-article sections, such as "About Ariadne" or the news section, were not clear. A redesign was due.
Each category of articles in Ariadne was given its own sub-menu. From an individual article, the reader could jump to either the sub-menu, the main menu, or one of the non-article sections via a set of icons to be found at the top and bottom of the article. If the article was in a regular column, additional icons allowed the reader to move between the column in different issues, without having to go through any content or issue menu's. A test run of the new format indicated that the navigation was improved, but some people still preferred having one contents page, no matter how long; therefore, both the sub- menu and the one content page system were combined to form the current menu-article structure.
The graphics were also redesigned to look more attractive, while at the same time not being too large as to significantly slow down Web page access. Rather than use pictorial icons, which run the risk of being unclear or ambiguous, icons overlaid with textual descriptions were designed; see Figure 1.
Individual articles and menus are marked up in a deliberately simple form of HTML for two reasons:
Several other changes have been made to the design of the Web version. The two most common negative comments that have been received about articles in this version is that:
From issue 7, a large number of the discussion-oriented articles have a response form clearly attached, where people could either send email to the authors of the article directly, or send a comment to be published as an article response. In addition, URLs are now explicitly referenced in articles. As a point of principle on the equality of network and paper based material, references to all kinds of resources (eg. Web, journal, book) are included in the same reference section.
From issue 1, Ariadne has contained a search facility [21]; however several problems have recently become apparent with said facility:
As new developments with network-based deliverables emerge, we evaluate the more promising examples to see if they can enhance navigation or general accessibility, for the Ariadne reader. For example, we have experimented with a Javascript pop-up window to add an extra method of navigation around a single issue [22] of Ariadne.
In addition, offering articles, or the whole issue, in a format different from HTML, needs to be investigated to see if any of our target audience are better served by one of these formats. The increasing use of PDF as an electronic journal delivery format [23] needs to be monitored, though with issues of the (smaller) print version of Ariadne being around a Megabyte in size in PDF format, there are obvious network implications.
The main criteria of any such access-enabling technology is that it should be relatively easy to use, add significantly to accessibility or navigation, be inexpensive, and should not add significantly to the time taken to produce an issue.
Content for the Web version of Ariadne is obtained from the following sources:
In addition to the main articles, there are several other areas within the Web version:
These areas are due for a review and a redevelopment, and so may be different in style or content in a few months time. One problem we have with any redevelopment is the permanence of URL's; as various Ariadne URL's have been mentioned in the print media, and some people will have bookmarked pages they wish to read at a future date, we have to be careful when redeveloping the structure of the Web pages.
When the first few issues of Ariadne were produced, the number of articles meant that time could be taken over article proof-reading, and negotiating with the authors over changes. However, from issue 4 onwards, with the increase in size of the Web version, the time required to produce an issue has become an acute problem.
Despite not being a refereed journal, there still needs to be some element of quality control, in terms of the content of articles, how each article is constructed and how it is presented to the reader. How the various elements that constitute an issue, such as articles, sub-menu's, multimedia files and navigational controls, are checked and linked together, is also a quality control issue.
Articles that are not in the equivalent print version (and therefore are already proof-read), from regular (trusted) columnists, or produced in-house (e.g. the caption competition) are read to check for potential copyright, slander, technical accuracy and irrelevance. In some cases, articles are sent to editorial board members who specialise in the relevant subject area for their comments; some articles are returned for partial rewrites, or are rejected outright (though we do say to the author, in fairness, why they are rejected).
When an article is deemed to be acceptable, it is put into an appropriate template file. The article then enters a "proof dusting" process, which has been updated for the production run of the next issue. Proof dusting will now be a multi-stage process of spell checking (twice, for errors that creep in during production), HTML checking, and making essential grammar alterations. In addition, the "proof duster", having read the article, constructs a one-line description of the article which is used as the article leader, and the link to the article from the main contents page and section menu. The editor finishes the process by adding and checking any references.
Why "proof dusting"? One of the attractions emphasised to potential authors of material for Ariadne is that their work will be only lightly edited (with, of course, checks make for spelling, technical accuracy, grammar, copyright infringement, slander, and relevance). This is especially important in reviews, or discussion articles, where very minor changes in words or structure can potentially have a large effect on the overall tone of the article, and may not be to the author's liking. In addition, intense proof reading would result, with time limitations, in articles either having to be submitted several months before publication, thus losing their immediacy and "news worthiness", or for a drastic reduction in the number of articles accepted for publication; neither of these scenario's is compatible with the aims of Ariadne.
As with evaluating technologies for their use in improving accessibility or navigation to the reader, so we have to evaluate technologies that help improve the quality and efficiency of issue production. For example, we are currently evaluating "Bobby" [26]. This is a validation system with a difference; the service, given a URL, checks the associated Web page to see how compatible it is with Web browsers used by the visually impaired. In addition, the page can be checked for compatibility against a wide range of versions of popular browsers. The downside of this service is that it is painfully slow to use, probably because of it's location rather than any problems with the service itself. To use such a service on all the articles in an issue of Ariadne would be very time consuming, though we are considering obtaining the source code. Ideally, such a service should be mirrored in the UK for easier evaluation of its usefulness.
Publicity of Ariadne is a vital task, more so when one of the criteria is for Ariadne to help effect "cultural change" in Higher Education.
The Web version of Ariadne is publicised in several ways, including:
A lot of feedback is received about Ariadne in general, either through email, or verbally at events. Unfortunately, rather less feedback is received about individual articles, despite easy-to-use feedback forms being included in several of the articles from issue 7 onwards. Feedback tends to come from a core group of enthusiastic members of the HE library community, and is mainly concerned with the overall content of Ariadne (usually, though not always, such feedback is positive). A survey was also carried out in both the print and Web versions which, though not producing a large response, was still useful. The survey results, which concentrated mainly on how much people liked individual sections in the print and Web versions, editorial board comments and other feedback have led to a minor reorganisation of the structure and content of issue 8 of the print version (with a subsequent effect on the "common material" in the parallel Web version).
Finding out where Ariadne has been mentioned has become an important task, because of our nature as a Web-based magazine. There have been several instances of print publications and list emails mentioning the Web version of Ariadne. Complete with incorrect URL. We have managed to work around some of these errors, by setting up new domain names (www.ariadne.ac.uk, for example, currently points to the UKOLN home page, from which Ariadne is clearly flagged), or dummy home pages. However, domains such as www.ariadne.edu, we can do little about. Mentions of Ariadne are often indicated by a flurry of emails or requests from related readers/organisations - for example, several emails from Russia which recently arrived over a few days indicates some probable mention of Ariadne in the former Soviet Union.
Links to Ariadne from prominent LIS sites no doubt help casual readers to stumble across Ariadne, as well as raising its profile. By using the command link:http://ukoln.bath.ac.uk/ariadne/ in search engines such as AltaVista and Infoseek, it is to identify Web pages that have links to Ariadne. Pleasingly, there are many of these, and they include prominent sites such as the Library of Congress [33] and IFLA [25] resource listings.
Some information on how readers navigate through Ariadne can be gathered from analysis of the log file of accesses. For every file accessed in Ariadne, be it an article, picture, icon or some other object, an entry in the log file notes the date, time, file ID, client browser used and various other information. However, the use of caches at either a national level (in the UK, a significant number of file requests from academia are served from the Hensa [34] cache), or at an organisational level (such as a University or an Internet Service Provider), means that many accesses to Ariadne are not recorded in our log file. Analysis of the log file has to be undertaken with the caveat of these "hidden readers" kept in mind; it is more therefore more accurate to usually look for "trends" in any statistical analysis, rather than hard numerical access statistics. This is very frustrating, as we cannot establish exactly how many people in the UK HE and Library communities read the Web version of Ariadne.
Several facts and trends do emerge from a cursory analysis; we have logged accesses from over 100 countries; of these, substantial numbers originate in the US, Australia, Canada, Scandinavia, Germany and the Netherlands. 42% of the logged accesses are from the UK; however, as a relatively large proportion of UK accesses are served by caches, the real proportion is probably significantly higher.
In the near future, we will be starting to use one of the many Web statistical analysis packages now available to carry out more detailed analysis of the logged accesses. Several questions we hope to answer are:
Ariadne was originally funded for two years, from the summers of 1995 to 1997. At the time of writing, this gives us only half a year under our existing funding remit, which has forced us to ask the question: "how can you financially sustain a newsletter for the Library and Information Science communities"? To even start to answer this, we have to look at the bigger picture.
The issue of future funding is set against an increasing backdrop of political, economical and technological turbulence and uncertainty in the United Kingdom. On the technological front, emerging bandwidth hungry applications, such as the Mbone, the increase in student numbers and the use of network-based technologies in mainstream academic research and teaching are resulting in unprecedented demands on bandwidth, and for hardware and software. Which costs money; a lot of money. On the political and economic front, 1997 is election year, with the outcome uncertain; the loser of our two main political parties seems likely to suffer deep trauma and undergo fragmentation and/or coalitions with other parties. However, the winner seems barely better off, with some painful choices to make regarding tax, what gets funding in the public sector, and what doesn't - the question is, how much of that pain will be borne by academia?
Upshot: more money needed; less money available; much uncertainty. However, amidst this uncertainty, a few facts can be stated with confidence:
This isn't just cultural change; like it or not, it's cultural revolution.
For Ariadne, this cultural revolution is a curate's egg. First, the bad parts; how does the Web version of Ariadne, and other similar types of publication, get funding to continue? What are some of the alternative possibilities to a centrally-funding scenario?
All of the options above have disadvantages - the first three would require a significant amount of extra administration, possibly bigger than the current project (remember, Ariadne currently consists of a few part-time staff; a marketing department would be a significant development). Overall, there would seem to be a trade-off between making people or organisations pay to access Ariadne, and the perceived or real editorial independence of such a publication.
To complicate matters, there is the relationship with the Print version; how should that be funded? Differently? Not at all (i.e. by revenue brought in by the Web version)? And how does the content interchange between both versions alter - should they be completely different versions? Or become completely identical? Or the Web version offer "tasters" (snippets) of material in the Print version?
To complicate matters further, what are the other LIS publications of a similar nature up to? Are any of them duplicating to a large degree what we cover (current answer: occasionally, in some sub-areas, but never frequently)? Will someone in the UK launch a "rival", free, Web-based publication for the academic LIS communities, covering the same sub-subject fields (current answer: who knows)? Deciding on the best strategy, in terms of acceptable funding (for our audience, and for us to sustain Ariadne), editorial independence, change of content, style and produce delivery and change of interaction between the Web and Print versions is very much a "non-trivial" task.
But what of the good parts of the curate's egg? Well, journalists say that "no news is bad news", and with the cultural revolution we find ourselves stuck in, there is no problem for the foreseeable future in finding relevant, useful content for our core and greater audiences. Sitting down every week and thinking "what's happening in the digital library movement", looking at what is announced and what people have submitted, the problems are not trying to flesh out Ariadne, but deciding what to include, what is relevant and useful now, and what will be relevant and useful in the near-future. This is good; we have a purpose, and we have content.
What of the future? As many eLib projects (and perhaps Ariadne) start to emerge from under the blanket of eLib funding, blinking and groping in the harsh and brightly lit world of self-funding autonomy, one lesson most will learn is that evolution does not end with eLib funding. System development, transfer of technologies, responding to rapidly changing audience/customer requirements and demographics; all of these will be ongoing and unending.
[1] The Web version of Ariadne, as described in this article: http://www.ukoln.ac.uk/ariadne/
[2] The print version of Ariadne is described in some detail in a companion piece to this article, "The Professional Magazine and Parallel Publishing", written by the Ariadne Magazine (print) Managing Editor, John MacColl: http://www.dlib/org/february97/02maccoll.html
[3] The Follett Report, commissioned by the Funding Councils of the UK in 1993, with the aim of looking at how to deal with the pressure on library and information resources caused by the large increase in student numbers: http://ukoln.bath.ac.uk/follett/follett_report.html
[4] The Electronic Libraries Programme (eLib): The programme, managed by JISC, which seeks to develop and implement Electronic Library resources, infrastructure and skills for the UK Higher Education sector: http://www.ukoln.ac.uk/elib/
[5] The JISC (Joint Information Systems Committee) funds and manages the eLib programme. JISC is funded by the Higher Education Funding Council of England: http://back.niss.ac.uk/education/jisc.html
[6] The Internet Resources Newsletter, produced at Heriot-Watt University, is targeted at academics, students, engineers, scientists & social scientists: http://www.hw.ac.uk/libWWW/irn/irn.html
[7] D-Lib, the Magazine of Digital Library Research, is produced by CNRI. A monthly magazine, it contains stories, briefings and in-depth articles on leading edge Digital Libraries research: http://www.dlib/org
[8] PACS (Public Access Computer Systems News) Review. Issued on an irregular basis by University Libraries, University of Houston, this has a wide scope of material from the fields of networking, IT and CD-ROMS: http://www.lib.ncsu.edu/stacks/p/pacsn/
[9] MC Journal: The Journal of Academic Media Librarianship. A peer- reviewed journal, articles for MCJ focus on a variety of issues facing media professionals in institutions of higher education, such as cataloging the Internet, copyright and distance learning: http://wings.buffalo.edu/publications/mcjrnl/
[10] LIBRES (Library and Information Science Research Electronic Journal). LIBRES is an international, refereed, electronic journal devoted to new research in Library and Information Science: http://www.lib.lsu.edu/stacks/l/libres/
[11] The Katherine Sharp Review, a peer-reviewed e-journal devoted to student scholarship and research within the interdisciplinary scope of Library and Information Science: http://edfu.lis.uiuc.edu/review/
[12] A review of the BUBL (Bulletin Board for Librarians) service, in the first issue of Ariadne: http://ukoln.bath.ac.uk/ariadne/issue1/bubl/intro.html
[13] A reply from the BUBL team, displayed in Ariadne, in response to the previous reference: http://ukoln.bath.ac.uk/ariadne/issue1/bubl/intro.html#r1
[14] John Lindsay's article on the evolution of the UK academic network infrastructure in Issue 6 of Ariadne, which contains come forthright opinions on current Electronic Library developments: http://www.ukoln.ac.uk/ariadne/issue6/lindsay/
[15] Cartoon. A regular cartoon, with a networking or IT theme, appears in both versions of every issue of Ariadne: http://www.ukoln.ac.uk/ariadne/issue7/cartoon/
[16] Sideline. A column in issues 1 to 7 of both the print and Web versions of Ariadne. This is an off-beat look at the trials and tribulations of attending conferences: http://www.ukoln.ac.uk/ariadne/issue7/sideline/
[17] Caption Competition. From issue 2, every issue of the Web version featured a caption competition, which has proved to be very popular. Using fill-in forms or the mailto: function, most readers can quickly send in a caption to be considered: http://www.ukoln.ac.uk/ariadne/issue7/caption/
[18] UKOLN, the UK Office for Library and Information Networking. An organisation, jointly funded by JISC and the British Library to carry out research and dissemination in areas including Metadata, Resource indexing, Public Library networking and Bibliographic Management: http://www.ukoln.ac.uk/
[19] Information Services at the University of Dundee, Abertay, where the print version of Ariadne is produced: http://www.tay.ac.uk/www/library/ualibwww.htm
[20] The Electronic Telegraph. A parallel version of the largest circulation broadsheet newspaper in the UK. The Web version contains all of the material in the print version, as well as links from each story to related network-based resources. An article on the production of the ET will appear in issue 8 (mid-March 1997) of both versions of Ariadne: http://www.telegraph.co.uk/
[21] A page from which Ariadne can be searched. Two facilities are available, that of just searching Ariadne locally, or that of searching Index Morganagus, which is an index of Ariadne and 35 other LIS resources, remotely: http://www.ukoln.ac.uk/ariadne/search/
[22] Javascript pop-up. If your browser (a recent copy of Netscape or Internet Explorer will do) can recognise Javascript, selecting this resource will generate, in addition to the contents page for Issue 6, an extra browser window. Through this window, you can jump to any of the main sections in one go, no matter where you currently are in issue 6 of Ariadne. We are currently evaluating our experiments with Javascript: http://www.ukoln.ac.uk/ariadne/issue6/test.html
[23] Electronic Journal Formats, Judith Wusteman, Program, Vol 30 No. 4, pp 319-343. An excellent review and comparison of the multimedia document formats currently used in electronic journals. Email: [email protected]
[24] OMNI Corner - a regular column in the Web version of Ariadne that covers issues regarding health and medicine informatic. OMNI is an eLib projects dedicated to building and maintaining a subject gateway for medical networked information; the column is written by Sue Welsh, the project officer. http://www.ukoln.ac.uk/ariadne/issue7/omni/
[25] IFLA Digital Libraries Resources and Projects Index. A superb and authorative listing of many of the key projects and developments in this field: http://www.nlc-bnc.ca/ifla/II/diglib.htm
[26] Bobby is a Web-based application that will tell you if your Web pages are compatible with browsers used by the disabled. The advanced options allow you to check across a range of browser versions, and also to check against several levels of HTML: http://www.cast.org/bobby/advanced.html
[27] lis-link mailing list. The main mailing list in the UK for academic librarians: http://www.mailbase.ac.uk/lists/lis-link/
[28] lis-elib mailing list. The main mailing list for announcements, nontechnical discussions and other matters arising from, or relevant to, the Electronic Libraries Programme: http://www.mailbase.ac.uk/lists/lis-elib/
[29] The We4Lib mailing list is an electronic discussion for library Web managers and those people either affected by, or developing, Web-based resources for Libraries: http://sunsite.berkeley.edu/Web4Lib/
[30] The Liszt directory of electronic forums, such as mailing lists and Usenet newsgroups. The list is searchable, and indications are given on how to join and participate in individual forums: http://www.liszt.com/
[31] Netskills. An eLib project, like Ariadne from the Training and Awareness area of the Programme: http://www.netskills.ac.uk/
[32] SOSIG - Social Science Information Gateway. An eLib project, from the Access to Network Resources area of the Programme. This is an acclaimed gateway to quality, assessed resources in the wide area of the Social Sciences. The gateway allows you to browse or search the assessed resources; amongst several options are the ability to distinguish between UK and non-UK resources: http://www.sosig.ac.uk/
[33] The Library of Congress listing of Library and Information Science Resources, with sections for journals, national libraries, general resources and so forth: http://lcweb.loc.gov/global/library/library.html
[34] Web pages describing the UK Academic National Web Cache at HENSA Unix: http://www.hensa.ac.uk/wwwcache/
[35] AltaVista vs. Lycos. An article from issue 2 of the Web version of Ariadne comparing these two search engines: http://www.ukoln.ac.uk/ariadne/issue2/engines/
[36] Pricing Electronic Journals, Hal R. Varian, D-Lib Magazine, June 1996. An examination of pricing strategies and models for Electronic Journals: http://www.ukoln.ac.uk/dlib/dlib/june96/06varian .html
[37] "In Cyberspace, no one hears you scream", Emily Bell, The Observer (national UK newspaper), 26 January 1997. An article that gives several examples of free Web-based publications that have suffered problems when moving to a charge-based system. For example, the Wall Street Journal went from 600,000 subscribers to its free edition, to 50,000 subscribers when it began charging a modest annual subscription. Email: [email protected]
Approved for release, February 14, 1997.
Spelling correction made at the request of the author, Editor, February 17, 1997.
hdl:cnri.dlib/february97-ariadneweb