The Pros and Cons of Industry-Standard DTDs   Table of contents   Indexes   From Mainframe to Intranet
  Rankin  George 
 

Croner & Croner & SGML — the first 3 years: opening the envelope!

 

Introduction

  Today’s presentation is absolutely non-technical! It is aimed at those managers, directors or project managers whose task is, or might be, to implement SGML in their organisation. The presentation will be in the form of a case study and will look at how Croner Publications tackled the whole issue of adopting the SGML standard. Covering the period 1993 to 1997, the presentation will highlight some of the problems encountered, how the business benefits of SGML were perceived and how the business case for SGML was sold to the publishers at Croner. The future is also addressed and an outline of Croner’s future strategy will be given. The theme of the presentation is simply that it took courage to finally grasp the nettle and “open the envelope”!
 

Croner: a case study

  Croner Publications is a successful publishing company based in London and is market leader in the UK in areas of publishing information relating to tax, employment law, transport, education, health & safety and the environment. The publications interpret the law for business managers and directors and the text of the publications is mainly in the form of commentary. Today, Croner publish over 100 loose-leaf books, 75 newsletters, 40 special reports, 70 bound books and 50 electronic products. Publications are updated frequently, including weekly, fortnightly, monthly and quarterly.
 

Croner in 1993

  4 years ago, the production department was forced to review its position to take account of a number of factors. Firstly, the plans had already been laid in 1992 to launch a number of electronic products, starting in early 1994. Secondly, a review of the in-house typesetting systems was felt appropriate to establish if themodus operandi was the best available. 80%]] of typesetting was performed in-house in 1993. There was concern at that time about how the production department would cope with building electronic versions of our hard copy publications. The answer of course was that the process would entail complex and expensive data conversions.
 

Review of publishing systems

  It was decided to conduct a review of all the production systems and the services of an independent consultant were commissioned. The report looked in-depth at alternatives to the systems then in place ( Xyvision and Ventura) and also at the processes involved in the production cycle. The report concluded that no changes should be made to the typesetting systems, although subsequently the systems were upgraded, but it did throw up one acronym with which I was to become very familiar indeed — SGML!
 

SGML working party

  Emanating from the review of publishing systems, Croner set up a working party to look at the mysterious SGML. The party comprised key personnel from production, editorial and IT. The working party looked in detail at what other companies in the group (Wolters Kluwer) were doing with SGML and finally reported at the end of 1994. The key recommendation was to run a major pilot project in 1995, during which the principal goal would be to establish the business benefits of SGML for Croner.
 

Why did Croner look at SGML?

 As already mentioned, the advent of electronic publishing at Croner was a significant factor in deciding to investigate the adoption of SGML. The fact that Croner was dependent on proprietary based typesetting systems was also a factor. It was felt that a more neutral solution would have to be found to cope with the future, which was clearly heading down the electronic information route.
 

The Pilot Project

 The business benefits of SGML were clarified during the pilot project, although they had been identified beforehand. The pilot ran for most of 1995 and into early 1996. The report produced on the completion of the pilot project highlighted these benefits:
 
  • no dual updating
    •  
  • simultaneous publication of folio and electronic products
    •  
  • reductions in production time and resources
    •  
  • cost savings
    •  
  • automation of typesetting operations
    •  
  • allow editors to access information from pcs
    •  
  • elimination of paper proofs
    •  
  • reduce dependency on proprietary systems
    •  
  • re-use of information
    •  
  • product development opportunities
    •  
  • quicker response — faster information
    •  The pilot project was organised in several phases to allow specific issues to be addressed and to ensure that the pilot did not progress too quickly before meeting the critical success factors defined in the report. The pilot phases were:
     
    1. Review of Procedures Reviewed existing production and editorial procedures.
    2. Prototype Construction Converted one section of a loose-leaf book from Xyvision to SGML. Using a common domain DTD, adapted for the pilot, the prototype phase aimed to evaluate the conversion to SGML and the output of SGML coded data to the typesetting system.
    3. Conversion All of Croner’s loose-leaf publications were assessed at a high level and 3 titles in particular were identified for document analysis (with a view to conversion). The phase also entailed writing Croner’s first DTD and preparing scripts for the conversions. 1 title was converted to SGML during this phase.
    4. Document Management System Review Before the pilot project began, it was acknowledged that some form of document management system would be essential. This phase looked at 4 options and produced a recommendation as to which system to purchase.
    5. Authoring and Document Management Phase 5 was when the SGML authoring software was purchased, tested and demonstrated to editorial staff. This was a crucial part of the pilot project as the impact on their working practices could be clearly illustrated.
    6. System Construction This phase involved pulling together the various components of the SGML system ( authoring tools, document management system, database) and to establish key issues such as level of granularity of data storage, indexing, QA processes, etc.
     

    Feedback from editors

     A number of questions were raised by editorial staff during the various presentations made by production staff during the pilot project. As the question of how editors’ working practices would be affected by SGML was considered critical, the comments were taken very seriously. They included:
     
  • who will write and maintain DTDs?
    •  
  • can the DTDs be changed at a later date?
    •  
  • why can’t editors view the effect of changes made in SGML automatically? (pagination issues)
    •  
  • what effect will SGML have on editors’ time and resources?
    •  
  • which publications will be converted to SGML?
    •  
  • what are the proposed training schedules?
    •  

    Live implementation of SGML

     Although a number of issues remained partially unresolved at the end of the pilot project, there was sufficient confidence within the project team to proceed to live implementation of SGML in 1996. It was decided that a multi-phased approach would also be best for the introduction of live SGML to Croner. Phase 1 was planned to allow the introduction of the SGML production system and, by the end of the phase, would see a number of loose-leaf books converted to SGML. Phase 2 would deal with the further enhancement and customisation of the system. 7 conversions were either completed or in progress by the end of 1996. By the end of this year, 18 loose-leaf books will be held in SGML.
     

    The future

      Croner’s SGML strategy very much mirrors their electronic publishing strategy. By the year 2000, 50%]] of Croner’s loose-leaf books should have been converted to SGML, with the full conversion programme running until the year 2005. Next year, newsletters will almost certainly be produced in SGML. New loose-leaf publications will be produced in SGML as from mid-1997. The future is very much based on providing electronic information as quickly and as frequently as possible. Croner already has a number of newsletters on its Web site. By next year, all newsletters will be available on-line.
     The benefits which were hoped for in 1993 are gradually coming to fruition. The learning curve for all concerned, whether production or editorial staff, is steep and long. Editors will take time to adjust to their new working practices and not all will find it an easy transition. The enormity of the task must not, and will not, be underestimated. However, we are now producing some electronic updates from SGML, concurrent with the folio update.
     

    Problems

     Several problems have been encountered along the way and have caused difficulties. Retaining the integrity of page boundaries in loose-leaf books once converted to SGML has been the biggest issue. This has been partially resolved by re-issuing pages for certain titles, but budgetary constraints will prevent this happening in all instances. To summarise the problems:
     
  • pagination of loose-leaf books post conversion to SGML
    •  
  • changes to working practices
    •  
  • steep learning curve
    •  
  • costly to implement
    •  
  • savings are not apparent in first 2–3 years
    •  
  • forms, tables, graphics
    •  Different solutions are required for forms, tables and graphics in an SGML environment. Croner’s team is still working on this aspect. Implementing SGML is complex and requires intensive training and support. This can be frustrating for production and editorial staff alike, especially in the early days.
     

    Costs

     Costs of over £750k will have been incurred in the 5 years since Croner started the pilot project (ie 1995–1999). These costs are principally in the following areas:
     
  • data conversions
    •  
  • consultancy costs
    •  
  • training
    •  
  • hardware
    •  
  • software
    •  
  • re-issuing loose-leaf books
    •  

    Savings

     Savings have been identified, but will not realistically begin to kick in until the 3rd year of the project. The savings will be made by avoiding costly conversions, simultaneous despatch of folio and electronic updates, performing more operations in-house and by reducing typesetting resources. Print runs will also be reduced in future as more subscribers opt for electronic delivery (eg newsletters will be on-line, therefore print runs will decline). It is estimated that over £1 million will be saved between 1997 and 2001 as the projected savings begin to take effect.
     

    Selling the idea to publishers

      There must be obvious benefits to any organisation before embarking on a project such as implementing SGML. The case has to be made at director level and publishers must be convinced the benefits are real. Financial benefits may be attractive, but the case at Croner was not sold on the grounds of cost alone, but more on the basis of being able to meet the demands of the future (ie an ambitious electronic publishing programme). CD-ROM and on-line are the future.
      In a highly competitive business, customers must receive information quickly and frequently. Producing more sophisticated CDs and putting information on-line creates new product opportunities for publishers. Editors must have unrestricted access to the information they need to create these new products and re-use of information should be easily facilitated.
     

    Conclusions

     Most of the anticipated benefits of SGML have been confirmed, although the problems mentioned already have still to be fully overcome. The most important achievement was winning the confidence of the company’s directors and convincing the publishers that SGML would bring benefits to them and their editors. Once the case was won at director level, it was easier to then work closely with other senior managers and editors to explain our plans and involve them as much as possible in the project. Financial benefits can look attractive, but a realistic case must be presented, which points out the possible pitfalls as well as the financial and non-financial benefits.
    The Pros and Cons of Industry-Standard DTDs   Table of contents   Indexes   From Mainframe to Intranet