Context-Sensitive Documentation in Industrial Process Plants   Table of contents   Indexes   Using the DOM as an XML/HTML repository API
 
 

Panel Presentation
 
Daniel   Rivers-Moore
  Director of New Technologies
  RivCom
Lotmead Business Village
Swindon   SN4 0UY  United Kingdom
Phone: +44 1793 790802
Fax: +44 1793 790802
Email: daniel.rivers-moore@rivcom.com Web: www.rivcom.com
 
Biographical notice:
 
Daniel Rivers-Moore
 Bergström, Peter  
 EuroSTEP 
EuroSTEP Ltd
 Shaw, Nigel  
 Stockholm  
 Sweden 
 
Daniel is Director of New Technologies at RivCom, a publishing services company specialising in publishing structured information in both printed and electronic form. He is chair of the Product Documentation team under the ISO subcommittee responsible for the STEP family of standards, and is joint project leader of the ISO initiative to harmonise STEP with SGML. He has been actively involved in the development of early implementations of XML for the delivery and presentation of structured information within large corporations. His interests include the development of generic data structures combined with flexible and powerful means of delivering and presenting the information to end users.
 
Nigel   Shaw
  Managing Director
  EuroSTEP Ltd

Email: nigel.shaw@eurostep.co.uk
 
Peter   Bergström
  Senior Consultant, SGML
  EuroSTEP
Drottniggaten 71D
Stockholm   Sweden  S-111 36
Phone: 0046 708 111 966
Fax: 0046 708 111 965
Email: peter.bergstrom@eurostep.se
 
Biographical notice:
 
Peter Bergström
 
Peter Bergström is one of the leading experts on SGML and HyTime in Sweden, and is actively participating in the current efforts to make STEP and SGML inter-operable on a practical level. He is now part of the Preliminary Work Item team on "SGML and Industrial Data" in the ISO STEP community. Peter has an extensive record of SGML and HyTime related work, both for defense and civilian use, and he is the main designer behind the FMV Grund-DTD.
 
EuroSTEP is a consultancy company based in Sweden, Germany, England and Finland. EuroSTEP works with global standards such as STEP, SGML and EDIFACT, and their applications in today's business. Another important aspect of EuroSTEP work is "opening up" organizations and people, a necessary step in understanding and utilizing open, standard-based technology.
 ISOGEN International 
 Kimber, W. Eliot 
 
Peter was until this year the chairman of the Swedish SGML User's Group. The home page of the Swedish SGML User's Group is found on http://www.admin.kth.se/SGML/
 
W. Eliot   Kimber
  Senior Consulting SGML Engineer
  ISOGEN International

Email: eliot@isogen.com Web: www.isogen.com
 
ABSTRACT:
 STEP 
 STEP/SGML harmonisation 
 
STEP  (STandard for the Exchange of Product model data) is an ISO standard for the sharing and exchange of product data. Aproduct is defined in an extremely generic manner, soSTEP can be seen as a standard for the exchange of information about anything we wish to describe using a computer system, starting from the viewpoint of the thing described.SGML  (Standard Generalized Markup Language) and its family are ISO/IEC andW3C  (World Wide Web Consortium) standards for structured information about anything we wish to describe using a computer system, starting from the viewpoint of the description.
 
In this session, the panel will explore the status of current initiatives to bring these two important families of standards together. The following questions will be addressed:
  • What is STEP / SGML harmonisation?
  • Who are the players involved (both in setting the standards and in using them)?
  • Why should the standards be harmonised?
  • How is it proposed that this be done?
  • When is it likely to happen?
  • Who cares? (with reasons why the answer is that you should...)
 SGML  
 STEP 
 
STEP andSGML are both ISO standards. Each belongs to a group of related standards that is far-reaching in its scope. Both aim to enable platform-independent and system-neutral sharing and exchange of structured information.
 
SGML has its origins in the world of publishing and is at home in the world of documentation. STEP has its origins in the world ofCAD  (Computer-Aided Design) and is at home in the world product data, where product is seen as an assembly of solid parts. However, just as the scope of SGML extends well beyond documentation, and can be applied to structured information of virtually any kind, so the scope of STEP extends well beyond the description of the shape and configuration of a manufactured product, and covers the whole of the product lifecycle, from design, through manufacture, use, maintenance, and eventual disposal or recycling.
 
 

What is STEP/SGML harmonisation?
 STEP/SGML harmonisation 
 
In February 1997, the ISO subcommittee TC184/SC4, approved aPWI  (Preliminary Work) under the title SGML and Industrial Data . The purpose of this work item is to carry out a study of how the STEP and SGML families of standards can work together for mutual benefit.
 
TheSTEP family consists of
 
  • STEP (ISO 10303)
  • MANDATE (Manufacturing Data)
  • PLIB (Parts Library)
  • Parametrics
 
TheSGML family consists of
 
  • SGML (ISO 8879)
  • HyTime
  • DSSSL
  • Topic Maps
  • the relatedW3C initiatives:
    • XML  (eXtensible Markup Language)
    • XLL  (eXtensible Linking Language)
    • XSL  (eXtensible Style Language)
 
The following statement of scope was formulated at the San Diego STEP meetings in June 1997:
 
 
There is a pressing requirement in many industries for product data and related information, of the kind typically contained in documents, to be managed in an integrated way throughout the product lifecycle. Like product data, document contents need to be modelled, identified, controlled, stored, assembled, configured, queried, navigated, delivered and shared within and among enterprises, in a system-neutral and platform-independent way.
 
 

Who are the players involved?
 
The Preliminary Work Item is being carried out under the auspices of ISO TC184/SC4, the Industrial Data subcommittee under the Industrial Automation Technical Committee. The Data Architecture Working Group (TC184/SC4/WG10) is overseeing the work, much of which is being carried out within the Product Documentation task group (TC184/SC4/WG3/T14).
 
Such is the importance of the work that the SGML Working Group, (ISO/IEC JTC1/WG4) has requested a formal liaison with TC184/SC4 in order to ensure that it is actively involved. Informal links are also being forged with the XML and XSL Working Groups under W3C . Given that there are already strong links between the XML and SGML Working Groups, we have here the basis for a very solid and broad-based consensus to be built around the STEP/SGML harmonisation effort. The following diagram illustrates these links:
 
 
 

Why should the standards be harmonised?
 
The harmonisation of the STEP and SGML families of standards can bring the following benefits:
  • integrated lifecycle management of data and human-readable information
  • effective configuration management and change control of product components and of document components
  • robust identification and referencing of product and document components, and reliable linking within and between both
  • technical documentation will become a snapshot of the data, automatically assembled and presented for use by human beings.
 
 

How it is proposed that this be done?
 
To answer this question it is necessary first to give a brief outline of some aspects of STEP , beginning with an overview of the EXPRESS data modelling language, which is a cornerstone of the STEP family of standards.
 
 

A comparison of EXPRESS with SGML
 
The following is a small example of an EXPRESS schema:
  
  SCHEMA animals;  ...  ENTITY dog   SUPERTYPE OF ( ONEOF       ( pedigree,  mongrel) )  SUBTYPE OF (mammal);    name: STRING;    owner: person;    WHERE    legal : licensed(SELF);  END_ENTITY;  ...
 
The following figure shows same schema fragment in the graphical notation known as EXPRESS-G:
 
 
EXPRESS and SGML are both ISO standard information-modelling languages. EXPRESS uses a database paradigm, with networks of relations between objects, whereas SGML uses a document paradigm, with a linear or hierarchical structure. EXPRESS has strong data typing, and is good at expressing complex constraints and rules.
 
Despite these apparent differences, there are in fact very strong parallels between STEP and SGML, as shown in the following table:

A comparison of STEP with SGML
STEP family SGML family
EXPRESS language SGML and HyTime
Integrated Resources Common DTD fragments
EXPRESS model DTD (or Architectural DTD)
Application Protocol Standardised DTD (or industry DTD)
Part 21 physical file Document Instance
Set of STEP data Grove corresponding to model
 
 

STEP/SGML harmonisation tasks
 
The following tasks are currently being carried out as part of the PWI on SGML and Industrial Data.
  • Develop a property set for data described by an EXPRESS schema. This will allow HyTime to be used to establish hyperlinks into product data.
  • Develop EXPRESS schemas for the SGML property set and the HyTime property set. This will allow STEP tools to navigate SGML documents.
  • Liaise over the handling of character sets across XML and EXPRESS. This is important in order to ensure that both standards support the same international character sets.
  • Investigate how XML can be used as an alternative exchange format for STEP. This should allow XML -aware tools to access and manipulate data drawn from STEP databases.
  • Develop an EXPRESS model for storing any type encoded data. This will allow documents and data in formats other than EXPRESS (such as SGML documents, graphics files, native CAD data, word-processor files etc.) to be stored in STEP databases. STEP would thereby become considerably more open, and multi-media documents would be able to be held and managed effectively.
 
 

When it is likely to happen?
 
The Preliminary Work Item should complete the first stage of its work at the forthcoming STEP meetings in Bad Aibling, Germany, in June 1998. This will include giving some practical demonstrations of the capabilities that STEP/SGML harmonisation will open up.
 
From that point on, one or more New Work Items may be set up, to move towards standardising the capabilities demonstrated by the conceptual and prototyping work done so far. Like all standardisation work, this will take some time. We can probably expect to see real practical results coming out of this work in another year to eighteen months, though any new ISO standards that may be necessary to consolidate the work may take a few months more than that.
 
 

Who cares?

 
The STEP community is watching this work with great interest, and includes some very large industrial players. The aerospace, automotive, defence and oil and gas industries are strongly represented. The STEP standards meetings,which take place three times a year, bring together 200 to 250 representatives, including senior IT managers from some of the world's largest companies. The principal benefit they see in this activity is the possibility of significant productivity gains and cost reductions in the area of technical documentation.
 
Bringing the power of HyTime or XLL linking to the STEP world will also have significant benefits, in improving the ability to manage the complexity of some of the very large data sets that are involved in such things as an aircraft's maintenance history, or the operating history of an oil refinery.
 
At the same time, the SGML Working Group and the XML-Data initiatives are hopeful that the EXPRESS language's ability to express formally complex constraints and business rules will bring to SGML a richness of semantic power that is currently missing.
 
STEP andSGML need one another. Each is incomplete without the other, but together they can do powerful things. Imagine being able to configure a power plant through the use of a DSSSL or XSL stylesheet. This is the kind of bold scenario that a true marriage of STEP with SGML should make possible in the forseeable future.
Context-Sensitive Documentation in Industrial Process Plants   Table of contents   Indexes   Using the DOM as an XML/HTML repository API