Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 1
DEVELOPING DATA-INPUT STANDARDS FOR COMPUTERIZED MAINTENANCE MANAGEMENT SYSTEMS: (Summary of a Symposium/Workshop) HOW IT COULD BE - A PROPOSAL AND A VISION Joseph H. Thompson Electronic Product Information Corporation It is a privilege to be invited to speak at this Federal Construction Council symposium on Data-Input Standards for Computerized Maintenance and Management Systems. My name is Joseph H. Thompson, and I am President of Electronic Product Information Corporation, or EPIC. My topic this morning is How It Could Be: A Proposal and a Vision. The sources I will use for this presentation are the Wall Street Journal, the Software Publishers Association, UnixWorld Magazine, and Micro Times Magazine. I founded EPIC in 1986 and after several years of growth on our own revenues we made the decision to seek venture capital in 1992 to expand our business and increase the company 's impact on the construction industry. Our principal investors are the Massely Burch Investment Group and Quaestus. EPIC's mission is to develop expert product information systems for building product manufacturers which manipulate text, graphics, pricing, and other data. EPIC's product group experience includes gypsum wallboard products, skylights, aluminum windows, wood windows and doors, commercial sectional and rolling doors, roofing products and systems, and waterproofing products. Clients in these product areas include United States Gypsum Company, Velux-America, Wayne-Dalton Corp., Caradco, Kolbe and Kolbe, Peerless Products, GAF Building Materials Corporation, Wasco Products, and W.R. Grace. In thinking about this presentation I thought it would be useful to begin by discussing how standards have been set in the past from a traditional perspective. Normally, standards are recommended by committees, groups of individuals, trade associations, or governmental bodies and the recommendation of standards historically has been based on the use of those standards by the group in question. In this scenario standards are recommended, followed by the development of information sets or products based on the recommendations. Once some number of such products has reached the market then a “standard ” is claimed, and further use by the market establishes those standards. With respect to the computing industry, however, I think a different paradigm drives the sequence of these elements.
OCR for page 2
DEVELOPING DATA-INPUT STANDARDS FOR COMPUTERIZED MAINTENANCE MANAGEMENT SYSTEMS: (Summary of a Symposium/Workshop) Scott MacNeilly is the President and CEO of Sun Microsystems. His view on standards is “A standard is anything that is shipping in volume.” This is an interesting perspective from someone who has been heavily involved in the design, manufacture, and marketing of UNIX-based workstations in a variety of markets. I've also included a quote from John Walker, the founder of AutoDesk. In an interview in Micro Times Magazine, when asked about Windows and graphical user interfaces, his statement was “What is happening is that workstation class operating systems are becoming available to the masses.” This is a significant statement, and I think it captures the focus of this symposium and this council: computing standards or information standards relating to the computing industry are driven by the market. There was a series of graphs in the January 27th issue of the Wall Street Journal that showed the performance of three companies who are busily setting standards in the market by virtue of market share and corporate performance. One graph shows Microsoft's performance beginning in 1988. It shows dramatic jumps in 1991 and 1992; those increases followed the introduction of Windows 3.0 in May of 1990. Another graph shows a similar performance curve for Intel Corporation, the manufacturer of microprocessors. The combination of the Windows graphical interface with the MS DOS operating system, and personal computers which are built around Intel chips, has resulted in this kind of performance and those components contributed to Intel 's ability to be able to claim nearly seventy-five percent market share in the microprocessor market. A third graph shows Novell's performance, with smaller numbers in the millions as opposed to the billions but Novell has established a seventy percent market share in networking software, based primarily on focused marketing towards personal computers and their proliferation in the market. Here is a quote from Casey Powell, the President of Sequent Computer Systems, Inc. “Computer makers have become just channels of distribution for Intel and Microsoft.” This is a bellwether statement about where the industry is heading, and it is interesting that it came from a computer manufacturer. Two companies have missed the evolution of standards in this fast-paced market. One of them is IBM and as everyone has observed, IBM has taken a terrible beating because of its inability to adjust its products, corporate structure to accommodate a new and fast-paced market which ironically it created. The other is Digital Equipment Corporation, which also had a difficult time during the past three or four years responding to the 'downsizing' trend as companies in every industry move away from mainframe and mini computers toward distributed computing networks composed of inexpensive, high powered PCs. These are the market realities of PCs in the market and I think by any measure, PCs can be considered an existing standard. Personal computers are the dominant computing hardware platform. In 1992 more than 40 million units were sold. There is an estimated total installed base of 100 million personal computers in the domestic market.
OCR for page 3
DEVELOPING DATA-INPUT STANDARDS FOR COMPUTERIZED MAINTENANCE MANAGEMENT SYSTEMS: (Summary of a Symposium/Workshop) The concept that I believe this symposium and this committee should consider is that data-input standards should be based on a market-driven standard. This graphic illustrates the impact of 100 million personal computers in the industry representing many dollars already expended by businesses: in the construction industry, this market segment includes building product manufacturers, architects, specifiers, contractors. If standard setting for data-input is to take advantage of existing market momentum then I believe concept of a market driven standard has great merit. Microsoft Windows is fast becoming the dominant interface for PC based applications. There are an estimated 20 million Windows users currently. Generally, in graphical user interfaces 9 of 10 new PCs ship with Windows, 20 million Windows users already exist, and another 1 million plus copies of Windows are sold each month. There are more than 2500 Windows applications currently in the market and, perhaps even more substantial, 100,000 Windows 3.0 or 3.1 developers kits have been shipped to date. More than 40 thousand Windows NT developer kits have been shipped to date, and this number is already well ahead of the pace for Windows version 3.x. To put some of these numbers in perspective I would like to step back a little bit and outline my understanding of how this whole evolution has progressed. In the early to mid 1980s, EPIC developed product selection systems for manufacturers which required the use of a CAD system for display and manipulation of product-related graphics. User proficiency was also a requirement for the productive use of such systems. Secondary technology was introduced in the late 1980s and early 1990s which utilized object oriented programming code built with C or C+ + programming languages. This advancement allowed information systems developers to link CAD environments with information which might be in an external data base or structure. An intermediate step was the application of systems shells so that a user could switch out of the CAD environment and access other information provided for a particular system through a C source code program. Beginning in 1991 and extending on into the future, what has happened is that the CAD environment itself can now be transcended by a graphical user interface (GUI) that has as its underpinnings a C+ + engine which can present and connect product information to operating maintenance and repair information. Such an interface eliminates the need for the user to have either a CAD system or the necessary expertise to operate such an environment. Part of the appeal of these new GUI's is the broadening information audience, composed of virtually every member of the design, construction, and maintenance team. By using this kind of system architects, specifiers and their clients, owners and tenants, contractors, facilities managers, distributors, engineers, and product manufacturers and suppliers all can handle the same kinds of information and exchange that information effectively and efficiently for the benefit of the team. In terms of the information model for computerized maintenance management systems there are essentially four basic components: text information, graphic
OCR for page 4
DEVELOPING DATA-INPUT STANDARDS FOR COMPUTERIZED MAINTENANCE MANAGEMENT SYSTEMS: (Summary of a Symposium/Workshop) information, some sort of interactive tool set and a user interface through which the information can be accessed, viewed, and manipulated. The structure of the data in the most basic form is ASCII for text, AutoCAD or DXF or product drawings, BMP files for product illustrations and product schedules, and bills of materials in tabular or listed form. The structure of this information should be developed in an object oriented environment. Programming in C+ + code will take advantage of dynamic data exchange, dynamic library links and object linking and embedding, all of which are components of object oriented programming. Additionally, object oriented technology provides platform independence such that any machine with any operating system can compile a basic C+ + routine and run that routine with the same basic data. At the most basic level, standards-responsive software is based on graphical user interfaces or gooeys' as they are called. Some examples of GUI's in the market today are the MS DOS operating system with Windows as an interface, Macintosh interface with the System Seven operating system, the Unix operating system using either Open Look by Sun Microsystems or X-Windows, a more standardized graphical interface for Unix. Applying the market share model to these options, the only viable direction in developing a data input standard with an operating system and a graphical user interface is Microsoft, which has 89 percent of the market, compared to OS/2 with about 1 percent, Unix with about 2 percent, and Apple with about 8 percent. Another bellwether measurement of software development is sales activity between the first quarter of 1991 and the first quarter of 1992. The DOS software market amounted to about $643 million for a negative 4 percent change from 1991 and the Windows applications market grew to $311 million for a 98 percent change. In the third quarter of 1992, DOS applications sales totaled $644 million, Windows applications represented $487 million, and Macintosh programs generated $243 million. Again the driving force behind this kind of market activity is an expanding information audience. Many of the individuals represented by these market segments need information in an easy-to-use format. Although the information needs vary from one group to the next, individuals in these diverse groups are buying inexpensive, powerful personal computers that come pre-loaded with Windows which then provides both incentive and demand for Windows-based information products and systems. EPIC's own approach in developing product information systems for building product manufacturers is to follow the market share leaders and develop appropriately responsive systems. EPIC's components include the Windows interface, C+ + object-oriented code structure, flexible selection systems allowing individual customization by specific manufacturers for specific products, and the availability of database records so that product selection systems can be linked into product pricing systems, order entry systems, general ledger systems, etc.
OCR for page 5
DEVELOPING DATA-INPUT STANDARDS FOR COMPUTERIZED MAINTENANCE MANAGEMENT SYSTEMS: (Summary of a Symposium/Workshop) Between 1986 and 1988 the availability and need for product information in electronic form looked something like this. During this period manufacturers primarily exchanged electronic product information with distributors and dealers and focused on pricing of the products and inventory levels. The majority of the construction industry market segments were not yet automated, so there was no demand for electronic product information systems. Between 1988 and 1991 the markets for all building product manufacturers expanded in response to changes in the industry itself. Many more individuals in different market segments needed information in electronic form. The exchange of information between distributors, dealers, and manufacturers expanded to include facilities managers, design professionals, home owners, contractors, and home builders. The dramatic decrease in the price of personal computing power coupled with the success of Windows to bring workstation class operating systems to the masses, as John Walker states, which in turn has increased the demand for construction information systems which meet these parameters. Beginning in 1991, the amount of information that had to be handled by these individual groups and separate market segments virtually exploded. This event paralleled an increase in the number of computers in all of these market segments, thereby creating the demand for product information and in fact bringing on the need for data standards and data input standards. Every market segment and every user of information needs access to the same information in a standard format, and the use of the information should parallel individual activities. Setting standards in the computing industry automatically sets standards for data and its use within the industry. Because market share plays such an important part in setting standards in computing, the traditional procedure just doesn't apply in this situation. In essence, what is illustrated here is a reversal of the traditional standard setting process. In this case, standards are first established through market share performance for operating systems, platforms, and interfaces. Once those standards are established then individual industries can claim market share in their own area. Recognition of these standards then drives recommended responses by committees, standards groups, trade associations, and governmental bodies. Following these recommendations, applications can be developed which follow both established market share standards and the response to those standards. This is a very different paradigm for setting standards but it is one that has merit particularly in the computing industry and one that I believe is geared for success.
Representative terms from entire chapter: