Issue: Volume: 23 Issue: 2 (Feb 2000)

The Future of Engineering Software




Imagine that gravity and friction have suddenly been reduced to 25% of their current values. You wake up one morning, and there it is: You now weigh 50 pounds instead of 200, and everything works much more smoothly. Outside you can hear cars crashing and emergency vehicles arriving on the scene. You bound out of bed-and bang your head on the ceiling. What changes would you have to make? What wouldn't work anymore? What would be better? What would be worse? How chaotic would the transition be?




Well, while gravity and friction haven't changed, their business analogs have-all because the Internet has reduced the "friction" of commerce, while the plummeting cost of computers has "lightened" their financial burden. And though the change has really been happening for years, it has only now reached critical mass, so we feel as if we just woke up one morning, and there it was.

It is now much, much easier to start a business, to initiate a construction project, to develop and farm out the production of a new product, than it ever was before. That's why it feels as if "gravity and friction"-the traditional factors that lend stability to business-were suddenly greatly reduced.

The world of engineering is only beginning to absorb the impact of these changes, so chaos is the order of the day. Global competition for engineering services of all kinds is a fact. You can hire designers in India, programmers in Russia, and architects in the Philippines, without leaving your home. And you can assign them work and get the results without ever meeting them.

While so-called "extranets"-the application of Internet-based technology for communications among organizations-is a universally acknowledged way of escape from this chaos, no one has yet even figured out the best way to package or sell one. In fact, as of this writing, more than 90 companies are offering non-comparable extranet products and services whose prices range anywhere from hundreds of dollars per seat to free.

On the positive side, the radical transformation of the engineering field has served as a wake-up call for the engineering-software community. The need to examine engineering software at large-to determine what needs to change, what doesn't work anymore, and what can be made to work better-has become plain.

To that end, the consultants at CounterEntropy Strategies (CES) thought it would be helpful to bring together a hand-picked group of leading users and vendors of engineering software as "ambassadors." They would first discuss these matters, then set an agenda for everyone in engineering to consider at a later conference. The agenda would help users, educators, and vendors bring some order to the chaos threatening to overwhelm engineering.

So, CES invited 64 of the most influential thinkers in the industry to Chicago last November, for the Summit on the Future of Engineering Software. Decision-makers from Intel, Microsoft, Ford, Autodesk, Bentley, Georgia Tech, Raytheon, the Federal government, and many other organizations attended a structured series of workshops and discussions.
The "ambassadors" and organizers of the Summit on the Future of Engineering Software held recently in Chicago.




At the end of two days of deliberations, a vote was held to determine how the assembly would rank the most critical issues in engineering software that had surfaced in the round-table and other meetings. Spokespersons from each of eight groups presented their issues. Then everyone voted on them. Each person received a hypothetical $30 to distribute over as many or as few issues as they liked-indicating on which issues they would spend their resources in coming years to improve engineering software.

The deliberations of eight topic groups yielded a total of 35 individual issues that fell into the following nine basic areas-listed here in order of importance to the voters:

  • User Interface. When it comes to engineering software user interfaces, the shoe still doesn't fit. Software tools are simply too hard to learn and too hard to use; they still try to make people adapt to them instead of the other way around. This issue garnered 18% of the votes-more than any other.
  • Web Implications. 16% voted for this topic and agreed with the statement, "All engineering projects of any substance will be managed with the help of an extranet-a project- or product-web-within three years." Business-to-business e-commerce is fomenting a revolution throughout engineering, and software vendors are still not quite sure what to do about it.
  • Interoperability. "It's not going away soon," said 15% of the voters. Both the technical side of file exchange and the intellectual-property aspects are still at issue. In the latter case, for example, if a vendor encrypts a portion of the files created with its software, users may not be able to access their own data if they cease using that software. No one has been willing to take responsibility on behalf of users for either of these interoperability problems.
  • Barriers to Implementation. Technology is not the major obstacle to improvement, fear of change is. Some 14% of the votes indicated that the dearth of full "art to part" and "trees to keys" implementations is due more to change resistance than to technological challenges. Of course, this reflects back to the first (user-interface) issue, above.
  • Knowledge Capture. 13% of the votes were used to express the sentiment that capturing and managing a company's engineering expertise-stopping the "brain drain"-is already important, but will become much more so in the future.
  • The Channel. Just over 8% of the votes were given to software-distribution matters. Here, the hot issue was application service providers (ASPs)-groups that are employing the Internet to rent software, services, and storage. Strong feelings were evident that ASPs will change the software sales picture by reducing the role of systems integrators, people who recommend software, and other "middle men."
  • A "Dream Machine." PCs and workstations are not the hardware that ultimately will be on the engineer's desk. Whether the so-called dream machine is a thin client of some sort, whether we revert to application-specific hardware, or whether we go in some other direction, 6% of the votes were in favor of setting a course by defining the kinds of systems that will be needed by engineering professionals in the coming years. This will be important both to users of the systems and to people developing software for them.
  • Failure of "The Grand Unification Theory of CAD." The ambassadors used 5% of their votes to demonstrate agreement with the idea that there will probably never be a standard CAD format or a standard set of CAD commands. The interest in maintaining proprietary file formats is simply far too strongly entrenched for many vendors to want to cooperate. Therefore, the Grand Unification theory appears to be wishful thinking on the part of some users.

    It is interesting to note that four of the top nine issues for the new millennium are precisely the same ones that we have been dealing with for several decades: user-interface issues, interoperability, barriers to implementation, and the need for better tools. Most poignant is the over-arching role played by user-interface issues in most of the other topics. Why can't we get it right? Why don't system builders do a better job of understanding users?

    The most surprising outcome was that few of these prominent industry leaders and experts felt comfortable attempting to look more than two or three years ahead. All the ambassadors were keenly aware of the chaos engendered by the inexorable march of Moore's Law (the number of active devices on a microprocessor chip doubles every eighteen months), and by the deck of wild cards that is the Internet.

    So where do we go from here? What did we learn from the Summit? Principally, that the perplexity each of us has felt in attempting to think about the future is shared by many, and that user interfaces, interoperability, and the Web must be creatively re-thought by users and vendors of the software-chiefly as seen from a user's point of view.

    Now the ambassadors to the Summit have outlined a millennial agenda for the engineering software industry. But to fill in the details and refine that agenda, more work-and input from more people-will be needed. So, the second step will be for the group to meet again in a larger body (500 to 1000 participants) in May of 2000, in Scottsdale, Arizona, at the Congress on the Future of Engineering Software. (For more information, see www.cadsociety.org).

    Joel Orr (joel@joelorr.com) is president of Orr Associates International, Chesapeake, Virginia. W. Bradley Holtz (wbh@wbh.com) heads WBH Associates, Bethesda, Maryland. Evan Yares (evan@yares.com) is executive director of the Open DWG Alliance, Phoenix, Arizona. The three convened the Summit on the Future of Engineering Software on behalf of CounterEntropy Strategies last November. That firm will host the Congress on the Future of Engineering Software this coming May.

Back to Top
Most Read