Web Magazine for Information Professionals

EPRESS

Stuart Peters on EPRESS text management software tools, currently in development.

The Electronic Publishing Resource Service (EPRESS)[1] is developing an online database system to aid the administration of electronic journals and make information available to a distributed Editorial team. Having identified labour as the greatest contributing cost toward electronic journal publication, EPRESS aims to reduce the burden by automating many of the functions of the Editorial Assistant and the publisher. This article demonstrates the scope of journal services and illustrates the ways in which the administrative chores are reduced to increase the efficiency of the publishing process.

The Rationale behind EPRESS

It is the view of some electronic journal producers that the perceived cost reductions of publishing electronically are an unattainable myth. The 'first copy' costs of traditional publishing do account for a large proportion of the overall production costs (Odlyzko, 1997[2]; Fishwick et al, 1998[3]). However, electronic publishing can become more cost efficient than traditional print publishing by adopting economies of scale.

Experience of Sociological Research Online,[4] an electronic journal set up under Electronic Libraries Programme (eLib) funding, demonstrated that the major costs involved with electronic journal production are for labour and marketing. Marketing is a black hole that will consume a budget of any size. Traditional paper marketing plays an enormously important role even for electronic journals. While web- and email-based marketing can be achieved with virtually no budget, limiting to electronic media will not bring new users who have otherwise not considered electronic journals.

Of the two major costs contributing to electronic journal production, EPRESS set out to reduce the labour costs. EPRESS is the Electronic Publishing and Resource Service, funded for two years by eLib. At the heart of EPRESS is a system aimed at reducing the time required for administrative tasks. The process of publishing scholarly refereed material is largely one of communication. Correspondence is exchanged between the editor, authors, referees, editorial boards, reviewers and so on, all facilitated by an Editorial Assistant (see Peters (1999)[5] for an analysis of these exchanges). The Internet offers a rapid communication medium, so it seems obvious to exploit that potential.

Sociological Research Online developed a series of tools aimed at speeding up administration and at sharing information with the distributed editorial team. These tools had been put together speedily on a 'just in time' basis, were not 'industrial strength' and were often cumbersome. For example, article information was stored in an offline database, data exported and turned into static web pages; those pages were then uploaded to the web. This meant that the Editor was always working with slightly historical data and putting the information online was an enormous task. Furthermore, the original database only recorded information - it did not assist the processes in any other way.

The first task of EPRESS was to take the article and other databases and convert them into live web databases. EPRESS has been developed in PERL, ideal for rapid development, which interacts with an SQL database (PostgreSQL) through a web (Apache server, any browser) interface. Theoretically the system should be platform, operating system and application independent meaning that any SQL database, web server and browser could be used. EPRESS includes four main databases (one for articles, reviews, people and book publishers) as well as databases for additional journal services (subscriptions, noticeboard, discussion forums etc.). Each database can be searched, browsed and edited through a standard web browser thus allowing data to be accessed from anywhere in the world.

With the main databases in place, it is fairly straightforward to add reporting features to allow the Editor to see how many articles are coming in and how quickly they are being turned around. This can be achieved simply by creating a SQL query and returning the data as a table or GIF graph representation. Amongst PERL's many freely available modules are several to produce dynamic web graphics.

The publishing processes of Sociological Research Online are similar to those of any traditional academic journal. Although EPRESS was developed with one journal in mind, the system should suit any peer reviewed journal. Print journals increasingly use electronic methods for the refereeing process and EPRESS could easily be adapted to output print correspondence rather than electronic messages. There are clear advantages to using electronic communication as the databases are integrated with the communication medium - something that is not possible with a paper based system.

'Two Click' Administration

EPRESS introduces many shortcuts aimed at 'two click administration'. For example, when an article arrives and is logged, the author is emailed an acknowledgement and the Editor informed of the new article. The Editorial Assistant achieves this by clicking a button that drafts an email, customised to the new author, and a second click to send this message and one to the Editor. Between the two steps is the facility to edit the email should the Editorial Assistant wishes to add any specific comments to the author.

Once the proposed referees have been entered into the article record, they can be invited to evaluate the article with another single click. The database performs a look-up in the 'People' database to find their email addresses, prepares the customised email telling them where the article is to be found and sets up the messages to be sent with a second click (see figure 1).

Fig 1: Procedure to invite referees
Figure 1: Inviting referees to review an article

  1. Referees added to article record
  2. Email drawn up
  3. Email sent
  4. Email arrives with referees

 

When all the reports are back, the Editorial Assistant can email them straight to the Editor with another single click. At each step, additional information is logged automatically so that the database tracks the status of an article (whether it is with the editor, author, referees etc), the dates on which correspondence is sent and so on.

Preparing an email automatically may not seem particularly groundbreaking, but this is a remarkably time-consuming task when it has to be done by hand. Even though email clients allow 'stationery' for standard correspondence, it is still time-consuming preparing multiple messages with information pulled together from several fields of a database. In addition, all these actions then have to be logged manually. This integrated approach not only speeds up the process significantly, but also diminishes the chance of 'human error' creeping in when actions aren't logged. Whereas sending a reminder of a forthcoming issue to 30 or 40 book reviewers once took a whole morning, this now takes two 'clicks' to send a personalised message, including bibliographic details, to all overdue reviewers.

As an alternative to automatic emails, the database system can instead produce Rich Text Format (RTF) documents, formatted to the journal's requirements, and then printed onto headed paper. This is necessary in order to produce letters to be sent to reviewers when books are sent out and could easily be adopted for paper based journals that correspond with paper. Sociological Research Online asks authors to assign copyright to the journal and again with a single click, an email is drawn up, it is sent with a second click and includes a fully customised RTF attachment containing the copyright agreement to be signed and posted back.

Good Practice

By using automated processes, 'good practice' can be built into the publishing process. New publishers can therefore adopt good standards without the need to acquire specialist knowledge and EPRESS can encourage the adoption of new standards.

Once the refereeing process has been completed, if an article is accepted , the record should hold all the metadata required for publication. At this point an HTML template can be retrieved containing all the information about the article: the author details, issue numbers, year and date of publication, abstract, keywords and URL. As well as appearing visibly in the article, this information is also embedded as Dublin Core metadata. The Editorial Assistant then inserts the text of the article and marks it into the journal style (see Figure 2).

Fig 2: HTML template
Figure 2: Publishing a book review via HTML template for Journal of Artificial Societies and Social Simulation.[6]

  1. Article record contains full information
  2. HTML template drawn up in colour coded RTF
  3. Review text added and marked up, article ready for publication

 

Journal data can be compiled in different formats for different purposes. Meta-data for Sociological Research Online is available as comma delimited data for an abstracting agency, SGML for a subscription agent and as ReDIF[7] for the Open Archive Initiative[8]. This can be done automatically or on demand, and the resulting files can be emailed automatically or transferred by FTP.

Automatic Data Entry

EPRESS is looking at ways to reduce administrative tasks as much as possible. If data could be entered automatically, then EPRESS would do so. There is one aspect of journal administration where this is already possible. When books arrive for review, the bibliographic details are stored in the database. This information is required not only for administrative purposes, but so that bibliographic information can be inserted automatically into the HTML review template, and a list of 'Books Received' can be published with each journal issue.

The online book services' catalogues can be used as free bibliographic databases. EPRESS has developed a system to query book details by ISBN and return a record if already in the database or, failing that, to query one of the online book services. If details are returned, the results are parsed and a record automatically created with the bibliographic details in place. As publishers generally make bibliographic data available to the online book services prior to publication, this information is often available by the time the book arrives for review.

Journal Services

Many journal services can be administered through online databases. Notices are stored in a database that records start and expiry dates so that each notice displays on the web site only when relevant. This ensures that the noticeboard only includes current notices.

The search engine and journal indexes are controlled from the background administration system so that they are updated when a new issue is published. Background databases allow for the discussion forums to be monitored and edited as necessary. The link directory is controlled by a database so new links can be added with ease, and old ones removed if they expire.

A recent major addition to the EPRESS databases is an integrated facility to handle subscriptions. The database controls access to subscription issues so that when a new subscription record is entered into the system, access is granted with immediate effect. On occasions when subscription agents have telephoned with an urgent request to set up a subscription on behalf of a client, it has been possible to arrange access before the end of the telephone conversation.

Sociological Research Online adopted an unusual subscription model whereby access via institutional networks is by paid subscription only, but access is granted freely via commercial Internet Service Providers. The two main advantages of this model are that the journal does not need to handle many individual subscriptions, thus reducing overheads, and readers who cannot persuade their institution to subscribe still have a way to continue reading the journal. Authentication is by network or domain so there is none of the administration associated with username and password systems. Readers can request a free subscription via a web form, the results are written directly to the subscription database along with additional details about the network they submitted from. The subscription can be activated as soon as the network has been checked as one belonging to an Internet Service Provider and that opens access to all other users on that network.

The subscription database can check for expired subscriptions and a facility to draft invoices will be developed shortly, as well as tools to track payments.

Step-Through Management

Other facilities, some recently added to EPRESS whilst others are still being developed, introduce issue management tools. When an issue is published, there are a number of 'standard' pages that need to be published alongside. These include content pages for articles, reviews and other special features, a list of 'Books Received' for review, subscription access files, a list of contributors and an updated 'Home Page' to reflect the arrival of a new issue.

A number of tasks must also be undertaken immediately following publication: the status of article and review records needs to be changed to reflect that they are published, data must be sent to subscription and abstracting services, the search engine and journal indexes must be updated, an email notification sent out to readers and copies of book reviews sent to the book publishers.

These tasks can all be prompted by a step-by-step series of actions, taking the administrator through the processes and reducing the need for technical knowledge about the underlying procedures.

For example, building the content pages for an issue is achieved through a 4-step process. First, the administrator selects which issue to create the content page for. Second, a list of the articles for that issue is presented, and beneath that, a list of additional articles that are also almost ready to be published. Articles can be added or removed from the upper list until it contains all the articles to be included. The third step presents the list of articles for the issue with a text input box. The administrator puts an index number into each box to reflect the running order. When this is finally submitted, a content page is produced (see Figure 3).

Fig 3: Preparing contents page
Figure 3: Drawing up a contents page for Sociological Research Online

  1. Assistant chooses volume and issue number
  2. Assistant selects which articles are to be included
  3. Running order 'index' applied to each article
  4. Table of contents returned

 

Total Integration

EPRESS offers a totally integrated solution that takes publishing from the point of submission, to the point of publication. All information is available immediately through a web browser, therefore making it available from almost anywhere in the world. The Editorial Assistant, Editor, Review Editors or other authorised 'players' can access the state of submissions, subscriptions, information about the authors, referees, potential reviewers, the emerging issue, historical data, readership statistics, the performance of reviewers, referees, book publishers and any of the journals' electronic services. Data is entered once throughout the system and becomes available in many different ways. No special hardware or software is required in order to access information, only a computer networked to the internet and capable of running a web browser.

EPRESS Availability

EPRESS is funded until October 2000. New users are invited to examine the system and evaluate it for their purposes, and anyone wishing to adopt the system is welcome to do so. A fully working demonstration of the major databases is available from:

http://www.epress.ac.uk/demo

Users wishing to take advantage of the demonstration databases should contact Stuart Peters for a username and password, and direct any further enquiries to him. Although the demo database is not a cut down version of the complete system, some of the facilities will not be available because there is no live journal behind it.

References

  1. EPRESS is at: http://www.epress.ac.uk/.
  2. Odlyzko, Andrew (1997) 'The Economics of Electronic Journals', First Monday, 2, no. 8, ht tp://www.firstmonday.dk/issues/issue2_8/odlyzko/index.html.
  3. The report 'Economic implications of different models of publishing scholarly electronic journals for professional societies and other small or specialist publishers' to the Joint Information Systems Committee by Francis Fishwick, Louise Edwards and John Blagden is at: http://www.ukoln.ac.uk/services/elib/papers/ta vistock/scholarly-journals/cranelib.html.
  4. Sociological Research Online is at: http://www.socresonline.org.uk/ .
  5. Peters, Stuart (1999) 'EPRESS: changing the way electronic journals work', Vine, 111: 31 - 36.
  6. Journal of Artificial Societies and Social Simulation is at: http://www.soc.sur rey.ac.uk/JASSS/JASSS.html.
  7. ReDIF is at: http://openlib. org/acmes/root/docu/redif_1.html.
  8. Open Archive Initiative is at: http://www.openarchives.org.

Author Details

 Stuart Peters
EPRESS Technical Officer
Department of Sociology
University of Surrey
Web site: http://www.epress.ac.uk/

Editor's note: Although Stuart Peters has since moved from University of Surrey, any readers interested in EPRESS can contact Nigel Gilbert on n.gilbert@soc.surrey.ac.uk