Machine Design

Show Notes COFES 2008: From Quantifying Innovation, to a New RP Format, and More

IT guys, programmers, industry analysts, visionary entrepreneurs, software company CEOs, and on-the-job engineers filled the Scottsdale Plaza Resort in Arizona for the Congress on the Future of Engineering Software.

Cyon Research Corp., Bethesda, Md., sponsored the annual three-day event. About 200 participants built relationships, listened to industry experts, and attended technology briefings. The focus: engineering technology and key strategies to help companies succeed today and into the future. A small sampling of the event follows.

Keynote speaker Karl Ulrich of the Wharton School at the Univ. of Pennsylvania in Philadelphia spoke on how to quantify innovation. “Valuable innovations are usually based on statistically exceptional ideas or opportunities,” he says. “Many industrial organizations use a tournament-like method to find these opportunities. Basically, companies generate many candidate ideas and then develop and filter them until only the best remains.”

A good example comes from a widely used toothbrush, says Ulrich. “Company designers first proposed over 200 designs. The company narrowed these to the best 50, and made foam models for testing. Another filtering process brought the number down to five for lab testing. Here, personnel used mirrors to analyze every aspect of toothbrush use while people brushed their teeth. Out of this came the Oral B toothbrush, which now has the largest dollar share on the market.”

Ulrich adds that although a tournament is not always necessary, the structure is a useful as a way to think about innovation statistically. “To really exploit statistics, companies should take more draws from a larger distribution, increase the mean of the number of ideas, the variance in quality, and accuracy in evaluating opportunities,” he says. “Because companies don’t have the resources to tackle many such problems, I developed software that makes it easier to do rapid peer evaluations with solid statistical properties. A recent good idea winnowed from the software was brokering medical tourism for semi-elective surgery.”

Later, at a quiet table, Deelip Menezes of Sycode software in Goa, India, showed a few of us one of his latest innovations: A free, alternative file format for RP data exchange that is compatible with, yet better than .STL, the industry standard. “The .STL format has problems,” he says. “For one thing, it uses inefficient methods to store data, so files are large. Large files are hard to transmit via the Internet and waste storage space.”

Worse yet, .STL doesn’t have any inbuilt security mechanism, says Menezes. “.STL files are usually created from Nurbs models to be shared with others,” he says. “Back when .STL was created, there was little worry about anyone extracting the underlying Nurbs model. Today’s reverse-engineering software makes this easy, resulting in huge businesses losses. Moreover, many RP service bureaus store .STL files on servers as part of online quoting systems. This leaves valuable data vulnerable to theft from hackers.”

Menezes says he started OpenRP, a nonprofit initiative, to give industry free software that can read and write a new .RP file format. “An .RP file is created from an STL file (ASCII or binary) and contains exactly the same geometric data,” he says. But .RP significantly compresses the data. An .RP file is 97% smaller than its ASCII .STL counterpart and 90% smaller than its binary counterpart. In either case, compression is much greater than if the .STL file were compressed with WinZip or any similar compression program.”

The file-security problem is also reduced because .RP has two levels of security: file and user, says Menezes. “At the file level, RPs are encrypted. At the user level, security is provided with an optional user-defined password. Should an STL file reach the wrong hands, it is useless without the password. Best of all, there is no data loss due to compression and encryption. When you create an .RP file from an .STL file and then convert it back to a .STL file, you end up with exactly the same .STL file.”

In another presentation, a curious audience listened as Mills Davis of Project 10x in Washington, D.C., asked: “What if everything we think we know about innovation is wrong? For example, hot new products are supposedly the thing,” he says. “But this is shortterm thinking because hot ideas are swiftly copied and commoditized. Innovation is often thought to come from being creative, but it actually comes from being disciplined. Likewise, innovation is deemed expensive, but failure to innovate is even more costly.”

Davis says to think of information like a newspaper — it represents the changing face of reality. Knowledge, though, is something learned and used daily. “A cultural shift from information-centric to knowledge-centric computing would help foster innovation” he says. “The Web is now the main engine of global innovation. The next iteration combines communication, computing, and distributed intelligence into what we call Web 3.0. This will soon let people and machines connect, evolve, share, and use knowledge on a grand scale and in new ways, creating multibillion- dollar markets for Web 3.0 products and services.”

Make contact
Cyon Research,
Karl Ulrich,
Project 10x,

TAGS: Archive
Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.