by the
Open Journal Project
team
Submitted to eLib: 13th August
1997
This shortened version posted
to the Web: November 1997
Contact
for this report
Steve Hitchcock
|
by the
|
Projects funded
by the Electronic Libraries
(eLib) research programme in the UK are required to submit annual reports
of progress to the funding body and are encouraged to post a version of
the report to the Web. This report covers the second year of the Open Journal
project, from May 1996 to May 1997, although some more recent developments
up to the date of the original submission of this report have been included.
The report builds on the projects's first
year report.
Submitted to eLib: 13th August
1997
Contact
for this report
|
Contents
Introduction
i Project objective restated
ii Highlights of the reporting period
iii Report against proposed Year 2 deliverables
iv Summary of the year
1 Activities and progress
1.1 Technology: an Open Journal framework
1.1.1 Open Journals and
the Distributed Link Service - the redesign and its effect on the project
1.1.2 PDF toolkit expanded
1.1.3 Software agents
1.2 Publishers and resources
1.3 User support
1.4 Outputs
1.5 Successes
2 Open Journals: first releases and evaluations
2.1 An Open Journal of Biology
2.1.1 Evaluation: summary
of outcomes and recommendations
2.2 An Open Journal of Cognitive Science
2.2.1 Evaluation: summary
of first responses
2.3 An Open Journal of Computer Science
3 Learning from experience
3.1 Scheduling innovation
3.2 Putting links in text
4 Project management: summary and financial statement
6 Future development priorities
8 Working with other eLib projects
8.1 Cogprints e-print archive
8.2 Information gateway projects
9 Principal project publications
1996/7
List of Appendices
Appendix 1. Improving user interaction with
the Distributed Link Service
Appendix 2. Evaluation details: an Open Journal of Biology
Appendix 3. Summary of project finances
Appendix 4.
A. Project evaluation plan and timetable
B. External evaluation plan
C. External evaluation, first report
Appendix 5. Exit strategy proposals
Appendix 6. Timetables of work for years 2 and 3
List of Figures
Figure 1. Proxy-based Web document enhancement
Figure 2. User home page for the Open Journal
of Biology showing collected resources
Figure 3. Sequence of linked biology pages
Figure 4. User home page for the Open Journal
of Cognitive Science showing collected journal resources
Figure 5. Sequence of linked cognitive science
pages
Figure 6. User home page for the Open Journal
of Computer Science showing collected journal resources
Figure A1. Link service 'remote controller'
i Project objective restated
To build a framework for publishing applications enabling journals on the Web to be interlinked in ways which build on the traditional qualities and identities of the journals, and which increases the readers' ability to follow, search and access the literature for themed study and research using the maximum available online resources.
ii Highlights of the current reporting period
The following deliverables, accompanied here with a brief comment on progress and notes on where they fit into the current scheme of work as described in this report, were cited in the project's original proposals to FIGIT.
The primary objective of the second year was to release first versions of Open Journals in three subject-based disciplines and to begin evaluations to inform subsequent releases and further development of the Open Journal framework during the final year. Two of the Open Journals have been released to evaluators.
Progress with the third Open Journal is advanced, the small delay being due to more extensive work than anticipated being required to build the linking framework and improve reliability. While evaluations are based on small, focussed groups of test users, arrangements are being discussed with publishers participating in the project to make at least one Open Journal generally available to demonstrate the best features of Open Journals to eLib practitioners and others. In addition, the project has been seeking to apply the linking framework to other eLib projects, notably the CogPrints e-print archive, and the OMNI and SOSIG information gateway projects.
1.1 Technology: An Open Journal Framework
Developers:
The link service technology that supports the project was in transition from a client-based service to a server-side process at the time of the last report. This meant that supporting software services such as the applications for superimposing served links on documents in portable document formats (PDFs), which had been demonstrated, had to be redeveloped.
The software tools currently being applied as part of the Open Journal linking framework are:
The major redesign to the link service during the second half on 1996 slowed link development during that period but also had some longer-term effects. The redesign was necessitated by the trend towards platform and browser-independent Web applications which are also 'network-centric', that is, applications which place fewer burdens on end-users and exploit network-connected machines for computing power. The link service originally required that software be installed on the reader's host machine to produce extra menus for each client application (Web browsers, word processors and document viewers) so that each application could communicate with the link service. This enabled links to be followed from Netscape, Word or Acrobat Exchange, even when these applications did not support Web links themselves.
The new approach makes the link service transparent to users by embedding it in the Web's document transport system and compiling links into documents as they are delivered to the user by a specially adapted Web proxy server (Fig. 1). This approach requires no extra client software for the user, which is an immediate practical benefit for the project, but it suffers some disadvantages. In particular this behind-the-scenes link compilation is applicable only to documents which are delivered via the Web and which are coded in well-understood document formats that can themselves support some form of hypertext link. Thus linking within the project is now restricted to documents in HTML and PDF.
Figure 1. Proxy-based
Web document enhancement: overlaying links as a server-side process
In addition there remain some constraints on the interaction of users with the link service in terms of their control of link presentation and link authoring. Some new and planned features begin to overcome these problems, but since these developments are not part of the project they are reported separately in Appendix 1.
1.1.2 PDF toolkit expanded
Adobe's Portable Document Format has become widely accepted as the publishers' format of choice for electronic journals that exist in paper form, and with the imminent, dramatic expansion of the number of titles being delivered via the Web, the influence of PDF is growing perceptibly. While the need for more support for linking is also becoming widely recognised among publishers and third-party agents, the project appears to be the only initiative seeking to extend PDF linking for journal applications. This significant development is being is accomplished through 'plug-in' applications developed at Nottingham University. [1] [2]
Adobe provides an application programming interface (API) for the Acrobat viewer applications Exchange and Reader which enables developers to create plug-ins. The plug-ins add functionality to the Exchange product and interact with PDF documents. For the Open Journal application these plug-ins perform database access, examine PDF documents and create links in these documents. The working model described previously was adapted during the current year to accommodate linking via the proxy server.
The initial idea behind the plug-ins was that a publisher would maintain a link database, applying it to PDF files before distribution. Two plug-ins were written, one which included one-to-one links, i.e. any occurrence of a particular word or phrase within a PDF file would be linked to a single destination. The other plug-in enabled one-to-many type links to be embedded into PDFs, i.e. any particular word can be linked to multiple destinations and on selecting the link the user is given a choice of which destination to go to. The act of including one-to-many links on PDF required an extension to the PDF specification so that multiple destinations, link descriptions, etc., could be stored for each link. To make use of these non-standard extensions a third plug-in had to be created which enabled these extensions to be utilised by the system. This plug-in should be distributed to users of the system, the other two plug-ins are only required by the person performing the link inclusion (typically the publisher).
An additional demand for Open Journals, however, is that link inclusions in PDF should provide the same flexibility for user selection and presentation as in HTML and should be controllable from the DLS. As above, this would enable end-users to control which link databases were added to which PDF documents. To achieve this another plug-in that works in association with the structures maintained by the DLS — it enables links to be incorporated into PDF documents without the need for human intervention at the DLS server — was developed. When a link server receives a request to include links it starts Exchange and automatically runs the plug-in, reading in the relevant link databases as required. This method, though, is too slow due to the need to run Exchange on the server. The current stage of PDF development in the project is thus concerned with rewriting this plug-in as stand-alone code that does not require the use of Exchange on link servers.
A complication of this link inclusion process is that it generates temporary files that use limited storage space on the link server, causing the service to crash when this space is full. It is the number of link inclusions rather than the length of the document that causes this, and the service provides large numbers of links. The work-around is to include links in the first page only of the PDF document, although it is anticipated that the user link selection features will also reduce the linking load.
1.1.3 Software agents
Although based around the DLS, the project is also developing a framework of information retrieval technologies. Through the development of subject-expert software 'agents', users are offered a greater range of relevant resources than they could discover alone. The emphasis is on developing agents that support resource discovery and linking. Typical applications for which prototypes have been created include:
Where the emphasis of the previous year had been on acquiring published materials in relevant subject areas and in formats compatible with Open Journal technology, the project's principal activity this year has been the release of the Open Journals themselves with this content.
Significant new materials have been acquired, however, with two more publishers joining during the current year:
These arrangements with new publishers bring important resources to the cognitive science and computer science Open Journals. ISI has provided a large, 500 MB database with abstracts of 200 000 papers taken from 7000 journal issues from 300 journal titles covering the years 1991–1995 in the field of cognitive science and related areas. For the purposes of the project these data are now served from Southampton University.
Chapman and Hall is contributing three journals in computer science, all recently available online, the first a new electronic-only publication, the other two based on existing print journals:
1.3 User support
Subject specialists, including researchers, teachers, information scientists and publishers, have been appointed to evaluation groups for the Open Journals in biology and cognitive science.
Overall direction of the project continues to be managed by the steering group with members from the research centres, publishers and from eLib. The group has met 11 times in total, six times in the current reporting period. Interest remains strong, with attendance invariably three-quarters of the eligible group.
1.4 Outputs
The principal outputs are the Open Journals that have been released for evaluation. These are not yet as widely available beyond the evaluation groups as we would like. Discussion with publishers on wider access rights are continuing. Instead the project remains visible principally through presentations and its publishing activities. The emphasis in the first year on publicity, promotion and generally raising the profile of the project created an impact. Consequently, during the second year interest in the project has grown noticeably. In particular we have responded to numerous invitations to give conference presentations, giving 10 presentations in total, including four overseas in Sweden, Germany, Finland and the USA.
These presentations have been fully documented with online materials, and a representative example can be found in [3]. Three major papers [1] [4] [5] have either been published or are awaiting publication (see section 9 for recent project publications).
1.5 Successes
To demonstrate the concept the project is producing three Open Journals
in the areas of biology, cognitive science and computer science. Each consists
of original journals from a number of different primary and secondary publishers,
served from a number of different sites in a number of data formats (principally
HTML and PDF). Dictated by differing cultures, academic disciplines show
various degrees of acceptance of online publishing, adopting different
approaches. This is borne out by clear distinctions between the three Open
Journals.
2.1 AN OPEN JOURNAL OF BIOLOGY
This was the first Open Journal to be released. The field provides the project with its longest list of participating journals (Fig.2), and one of the project objectives in this case to begin to link these journals into non-journal data resources on the Web.
Figure 2. User home page
for the Open Journal of Biology showing collected resources
Biologists have access to a diverse range of online materials, from
journals to books, databases of gene sequences and other molecular structures,
libraries of graphical images and teaching resources. The information is
invariably well catalogued and often freely accessible. The databases in
particular form a focal point for the field, lending it an online coherence
that few other disciplines can match and attracting a higher proportion
of the field’s prestigious journals online than other fields.
The spectrum of link creation options that could be supported by the DLS includes highly pertinent, hand-crafted links such as might take a user from a biology journal page to a graphical molecular database. Links can also be created en masse by a batch computational process, for example, in citation or keyword linking.
The majority of the 12 000 keyword links that are currently being demonstrated in the Biology Open Journal (Fig. 3) were created by computational methods and demonstrate the power of the once-only authored generic link, a powerful link type developed in the Microcosm linking model on which the link service used in this project is based. This is a cheap way of providing a database of links for a journal archive by creating a link from any occurrence of a specific word or phrase within a literary corpus to any paper with that keyword specified. This approach is also used for terms from the online Dictionary of Cell Biology, so that the occurrence of a key dictionary term anywhere in any document is automatically linked directly to its definition.
To create a database of these links requires a source of metadata for the articles of interest: extracting the keyword fields is a small programming effort which leverages links out of another’s authorial or editorial effort. In practice, many of the project’s resources have not been provided with metadata records. The programming effort required to extract the keywords from an encoded document display format such as PDF has been a lot higher, but still less than creating the links manually.
To date not all journals have been linked using keywords, i.e. all journals can display links from the link service, but those which do not provide keyword data have no links pointing to them. This more complete integration, via links, of the various journal resources in the Open Journal is among the actions proposed for the next stage of development of this Open Journal.
2.1.1 Evaluation: summary of outcomes and recommendations
The evaluation of this Open Journal, the first to be released, indicates the rapidly escalating demands on links in the biology literature, highlights some new ideas and refinements for a later version of the Open Journal, and points to some remaining difficulties.
Biology is perhaps the most richly resourced academic discipline on the Web and is a major opportunity to demonstrate the Open Journal concept. Equally, it is one of the fastest moving fields, especially with respect to linking [4], with others seeking to exploit the potential, so the Open Journal must be judged in the context of greater progress and sophistication in biology.
Figure 3. Sequence of linked biology pages: a unlinked PDF page, b same page with links added by the link service via the project's PDF plug-ins, c a single link can point at multiple destinations, d following a link on 'embryo' to the Dictionary of Cell Biology, a remote resource (note how this resource also has links added to it by the link service, so although it is remote it can still be an integral element of the linked Open Journal)
Eight specialist evaluators were recruited for the Open Journal of
Biology in February 1996. After some delay due to complications with the
development of the linking framework, the Open Journal was released on
13 March 1997. This delay has affected the findings.
The concept of an Open Journal has been generally well grasped, and the novel set-up needed to view links, that is using the proxy link server, created few problems for users, with one important exception, library users.
Although some of the problems that delayed release were resolved, some solutions were only partial (PDF plug-ins) and some problems recurred (reliability of the link service). Crucially, the development of this particular Open Journal, as the first to be produced and one with a relatively high proportion of content in PDF, was especially dependent on the availability of the framework for applying links in the form of the DLS and PDF plug-ins. As a result the release version demonstrates fewer link types, e.g. no citation links or links to gene sequence databases (although not too few links in number!), than had been anticipated and as a result lacks the full integration between resources that should characterise the Open Journal approach.
The following major actions are suggested for the Open Journal of Biology:
The project is not a content owner, so the emphasis is the development of a flexible, responsive and powerful framework for linking journal resources, a framework that will meet future demands. Given development elsewhere of the key features of citation and sequence linking, it is the former characteristics of the framework and not just linking that must be demonstrated. The next release of this Open Journal will thus need to be justified in terms of: time to develop, ease of updating, maintainability for archiving and cost, areas in which the project is most able to demonstrate significant advantages over other services.
.
2.2 AN OPEN JOURNAL OF
COGNITIVE SCIENCE
This example forms the basis of the project’s most extensive demonstration of the use of a link service for citation linking, i.e. linking journal papers from reference lists.
The archives of one of the first Internet journals, Psycoloquy, edited by Stevan Harnad and launched in 1990, together with its print counterpart Behavioral and Brain Sciences, which recently began an online preprints service, are the nucleus of the project’s cognitive science Open Journal (Fig. 4). Papers in these journals are linked to the 200 000 abstracts in the database of bibliographic information provided by ISI.
Figure 4. User home page for the Open Journal of Cognitive Science showing collected journal resources
The DLS can be used to overcome the lack of links on most Web journals
by dynamically matching a fixed database of pre-defined links against the
contents of an article, inserting a link whenever an explicit match occurs.
For the purpose of citation linking this involves matching citations and
bibliography lists in a document against the database of bibliographic
information.
This activity is undertaken on behalf of the user by the citation agent introduced in section 1.1.3. When the DLS processes a document which is defined to be in the Cognitive Science Open Journal, it launches the agent to act on the document before adding other links in the usual way.
As well as recognising the citation, the agent has the problem of retrieving the cited paper, i.e. where to link the citation to. We have chosen to link the citation in the body of the document to the bibliographic reference at the end of the document (an internal link) and then to link that reference to an external bibliography database. This database, although not linking to the full-text articles, provides a complete set of metadata for the cited article including an abstract and the citations given in that cited document. Making that record available, with each citation linked back into the database, allows the user, in theory, to browse the complete literature (in summary form), with each reference instantly available (Fig. 5). In addition, the database is indexed and linked to 'forward' references, that is, papers that have subsequently cited a given paper.
Extracting the citation is a more or less difficult task for the agent depending on the format in which the paper is held. Highly structured document formats based on SGML are in principle the simplest as all the separate components of the bibliography data are marked explicitly. At the other end of the scale, simple ASCII provides problems in recognising both the existence of a citation and the boundaries of its component parts (for example, where the author names end and the title starts). HTML, the format of the articles used for this Open Journal, can provide extra clues implicit in the formatting markup (for example, titles may be rendered in italics). Details of the implementation are described in [5].
The process that matches a citation to an entry in the database was originally performed at the time the user follows the citation link. In other words, the software identified the places in the document where the citations occurred and put the links in as queries to a database search script. The advantage of this approach was that it delayed the relatively slow database look-up until the user had explicitly requested it by following the link. The strategy of adding links blindly, however, often resulted in the user being presented with an apology from the database instead of the abstract of the cited paper, because the domain of the database is limited, principally in time but also in breadth of journal coverage.
For this reason the links are now verified before they are added to
the document, and each citation is checked individually for a unique resolution
in the database before the link is created. Also, the link is created directly
to the correct record for the article, instead of indirectly to a query
processor as before. This extra processing is now performed as the document
is delivered to the user and slows down delivery, but since HTML documents
can be streamed (the user can see the start of a document while the DLS
is still processing its end), and since the bibliography section is invariably
at the end of an article, this has not proved a significant disadvantage.
Figure 5. Sequence of
linked cognitive science pages: a An original paper from Psycoloquy
without links, b citation links added by the Distributed Link Service,
c citation link followed to bibliography section, d bibliography
link followed to ISI abstract in database (note how some citations associated
with this abstract are also linked by the DLS, and the 'forward referencing'
links, i.e. to other papers which cite this article, as indicated for the
selected abstract and by the linked ^ symbol to the left of articles cited
within this abstract
Current indications are that some 60% of the citations in the nucleus of the Cognitive Science Open Journal are outside the time-range of the database and hence are ignored. About half of the remainder are successfully recognised and linked, with the rest accounted for by citations to books or citations out of the core of the discipline. This is perhaps not surprising for such an inter-disciplinary subject, with patterns of success rates emerging for different authors and different themes within the journals. Further work is being conducted to improve the hit-rate.
Although the project philosophy is to give users access to links distributed across a network, it has found that access to localised, formattable data is necessary for creating large linkbases computationally. Selected abstracts data made available to the project by the Institute for Scientific Information (ISI) required an intermediate index for linking, but the resulting links from journals such as Psycoloquy are proving reliable and relatively complete. The real power of this approach is that once created, this referencing linkbase could be applied to other cognitive science journals from which reference data can be parsed wherever they are on the Web and wherever the abstracts database is held. All that is required is that user is able to access the resources, e.g. as a subscriber, as well as the link service.
2.2.1 Evaluation: summary of initial responses
This evaluation is ongoing with the full report anticipated in September. First responses have been generally positive and encouraging. The 'forward referencing' feature, i.e. 'articles that cite this one', was described by one user as 'great'. Another evaluator reports: ' Very good first impressions. This will be a boon to those like me who are not able to literally travel around for information for medical reasons'.
Set-up seems to have been straightforward for most users, although one possible problem that shows signs of recurring, as in the biology demonstrator, is the use of the proxy link server. In one case a company firewall policy prevented access to the link server, and a successful workaround has been implemented for this user. Elsewhere, a user notes a difficulty caused by having to set a competing proxy to access her Internet Service Provider (ISP) but claims to have been able to view the links anyway.
The most common request from users: 'more (full-text) articles, more
links'. A satisfying endorsement in this respect of the project philosophy
and something that publishers may wish to take note of.
2.3 AN OPEN JOURNAL OF COMPUTER SCIENCE
This will be perhaps the most difficult of the three Open Journals to build. Computer science offers large volumes of disparate online resources with little organisation apart from some loosely catalogued indexes. Few established computer science journals are available online, or have significant online archives, so in this case the project aims to identify pockets of coherence that can be extended through the application of links, starting with specialised journals provided by publishers collaborating in the project together with citation linking to Computing Abstracts Online (CA) from MCB University Press. This is a smaller database of abstracts than that available in cognitive science and it does not incorporate citation data in the way that ISI does.
A citation linkbase has been created from the records in CA, and the citation agent developed for the cognitive science work will be applied here as well to recognise links within the references of journals in Fig. 6. In the first instance this work will concentrate on just one or two of these journals, to establish the overlap between citations in the journals and abstracts within CA; also to establish the effectiveness of the agent with a new data type. This may form the basis of the first release of this Open Journal, but the work will quickly be extended to include the other journals when the data has been analysed for the best means of linking and the agent optimised for the application.
Figure 6. User home page for the Open Journal of Computer Science showing collected journal resources
In a project with a large number of partners and users seeking to apply
new technology to present information in, sometimes, unconventional ways,
it is not surprising that some difficulties are encountered. This report
highlights two important problems that have been tackled in the areas of
project management and technical implementation and presentation.
3.1 Scheduling innovation
As last year, progress with link creation and editing for the Open Journals
remains dependent on the development of the linking framework. Innovative
technology can be notoriously unpredictable, but the main difficulty this
causes is in scheduling work and keeping to the provisional timetable.
Equally, with the release of some Open Journals, development of the technology
now has to be responsive to the needs of users. In the case of the project,
which uses a number of software components such as the link service, plug-ins
and agents, this is a severe complicating factor and it has had an impact,
especially on the development and subsequent reaction to the release version
of the Biology Open Journal, since this preceded all others.
3.2 Putting links in text
The emergence of the Web has popularised the application of hypertext,
but among the Web's vast user population, and among the project's user
groups, expectations of hypertext differ. This has become evident through
evaluations and conference presentations, i.e. outside the research laboratory,
in the occasional reaction by users against certain types of links. Whereas
citation links are welcomed because they are familiar in the journal culture,
keyword links are not so easily assimilated. There are two reasons for
the reluctance of users to accept unusual links within text: at a basic
level they alter the appearance of the page and can be distracting for
users, and they could potentially conflict with an author's intent. This
could change through greater experience of users in handling links, greater
assurance of the quality of published links of the kind that users expect
of published text, and improved methods of controlling the presentation
of links. For example, if constrained to appear within the journal from
which the keywords were extracted, keywords links could play an effective
role in indexing that journal.
The project is slightly underspent on its budget for the reporting period.
This money will be carried over to the final year and may be used to accelerate
current development where necessary. A financial statement for the project
is included in Appendix 3.
As well as the evaluations of the individual Open Journals, the project
is itself being independently evaluated by an external evaluator. This
evaluation is ongoing. Appendix 4 shows the project's evaluation plans
and includes the first report of the external evaluator.
6 Future development priorities
Linking tools
Briefly, the options include the creation of one or a series of software
products based on the work of the project. Another approach might be to
ally these products with a service-based business to support the development
of commercial Open Journals. The project is meeting with publisher partners
to discuss these developments.
8 Working with other eLib projects
The project continues to maintain informal contacts with a number of
eLib projects. We have met and shared information with eLib colleagues
and generally found these meetings to be useful, and this would seem to
be reciprocated as two articles in Ariadne (Jon Knight in November
1996, and Ian Budden in January 1997) testify. We have also offered to
work with projects to support link implementations over their work where
appropriate. It is likely that the Nottingham PDF linking work will be
tested within the Parallel Processing of Transactions (PPT) project when
both have made more progress. In this report we highlight two examples
where work with other projects has begun.
8.1 Cogprints e-print archive
This work has parallels with the development of the Cognitive Science
Open Journal and is expected to use a similar approach to linking, but
there are other factors to be considered. For example, the linking agent
is demanding of the accuracy of reference data provided by authors, while
papers in the archive will be unedited and therefore the citation data
effectively unchecked. Instead, software is being developed that assists
authors to provide strictly formatted bibliographic data for papers as
they are submitted to the archive. The author has the option producing
this data (it is not required so as not to discourage submission by being
too complex or time-consuming) semi-automatically: the software produces
a table of pre-parsed database entries corresponding to the date, author,
journal, etc., fields in each item in the paper's bibliography. Any mistakes,
due to typographical errors by the author, for example, can be rectified
by the use of a form, and the human-verified database stored in the archive
along with the paper. It is anticipated that this information will then
be used to link to full citation databases as in the Open Journal, although
no arrangements for the provision of such databases have been finalised
8.2 Information gateway projects
For practical reasons the objective of the approach adopted in this case has been to seek to create and apply useful linkbases as independently as possible from the projects, in this case OMNI and SOSIG, aside from seeking the necessary permissions and establishing the prior interest of the other project. A feature of the planned demonstration would not simply be the application of a, hopefully, useful third-party service, but that it could be developed without detriment to the current work of the other project. A full set of data has yet to be retrieved to the linking site, however. Attempts to retrieve the data using a robot have proved incomplete, despite the relaxation of the robot exclusion mechanism on the OMNI site for the purpose.
9 Principal
project publications 1996/7
[1] Probets, S., Brailsford, D., Carr, L., and
Hall, W. (1997) Dynamic link inclusion in online PDF journals. To be
presented at EP'98, the seventh International Conference on Electronic
Publishing, Document Manipulation and Typography, St Malo, France,
April 1998
[2] Probets, S. (1997) PDF, databases and
the Internet. Acrobatics, Vol. 2, No. 1, March, 3-7
[3] Open Journal Project (1997) Can links
become the new publishing model for the Web? The Open Journal approach.
A compilation of conference presentations
http://journals.ecs.soton.ac.uk/newpub/
[4] Hitchcock, S., Quek, F., Carr, L., Hall, W.,
Witbrock, A., and Tarr, I. (1997) Linking everything to everything:
journal publishing myth or reality? ICCC/IFIP Conference on Electronic
Publishing, Canterbury, UK, April (to be published in Serials Review)
http://journals.ecs.soton.ac.uk/IFIP-ICCC97.html
[5] Hitchcock, S., Carr, L., Harris, S., Hey,
J. M. N., and Hall, W. (1997) Citation linking: improving access to
online journals. Proceedings of the Second ACM international conference
on Digital Libraries 97, Philadelphia, USA, July ( New York, USA: ACM)
pp. 115-122
http://journals.ecs.soton.ac.uk/acmdl97.htm
*Full list of the
project's Web papers
Appendices
Appendix 1. Improving
user interaction with the Distributed Link Service
To enable users to communicate with the link server may require at least some configuration beyond the defaults set up by the service. This takes the form of a 'link remote controller' (Fig. A1), a development of an earlier proxy session controller. The link controller is an HTML form displayed as a separate window the input from which is interpreted by a module in the link server and retained on the proxy's persistent storage.
Figure A1.
Link service 'remote controller'
The controller establishes a dynamic session (binding user and host together with a set of link server parameters) which is used to control the behaviour of the link server from that point in time onwards for that particular user. It is intended that the user will invoke the controller just once to set his or her preferred configuration, and only again afterwards to adjust the configuration — links will always be added automatically to the documents according to the last settings of the controller.
The purpose of the controller is to give to the user the ability to choose how links are displayed and used within the processed documents. The DLS allows the user to manipulate links directly to control presentation and navigation.
This link controller, although implemented and available to users, has not featured in the initial evaluations as users have been directed to the default service. The next releases, however, will make explicit use of the controller. Users will be encouraged to refine link selection, for example by specifying which link databases are switched on and off.
Another aspect of link presentation is discrimination, to distinguish between links added by the link server and those native to the document. The DLS could modify links by colour or font style, for example, to differentiate authored links from computed links. Link presentation and differentiation can be controlled by applying a Document Style Semantics and Specification Language (DSSSL)-like implementation that, in contrast to Cascading Style Sheets which sets typographic attributes, uses a two-stage process that can revise formatting or add new content to act as a link anchor.