Most of my readers are probably aware of The Congress on the Future of Engineering Software that occurs annually over an April weekend in sunny Scottsdale, AZ at the Scottsdale Plaza Resort (not the Scottsdale Resort like some shuttle drivers would insist) and hosted by Cyon Research. If you are not familiar with the event, I highly recommend you visit their website and save me the trouble of name dropping. I am probably one of the unique few that attend this event that is not a software vendor, component supplier, industry analyst, hardware OEM, or media. Even though I blog about CAD and engineering, during the event I represent my employer as a user. So while some people are there to gauge the industry or take pictures for interesting articles, I'm there to gauge where my business critical applications are heading and figure out how to position my company to best prepare for and utilize the future of engineering software.
This was my second year attending COFES. Last year I knew a few people and got some great advice on how to get the most from COFES. This year, I was actually able to put that advice into practice and my head is still pounding with the wealth of information I took home from the event. Writing down four jam-packed days of information into a trip report or blog post just doesn't do justice to what actually happens at COFES, but I'm going to try anyway.
DaS SymposiumEven before COFES officially starts, there are events. The big event that most interests me is the Design and Sustainability (DaS) Symposium. The key takeaway from last years Symposium was the creation of the Green Keystone Alliance (GKA) containing participants of the Symposium like Sustainable Minds, Ford, and partnerships with NIBS. The purpose of the GKA is to be a central advisory panel to all of the other organizations that are trying to define sustainability. The presentations were about creating long lasting products, getting rid of designed obsolescence (like needing a new cell phone every 2 years or less), planning for the entire life cycle of a product including disposal, and creating products that use less raw materials. But, none of these concepts were able to put an economic model together. They were not able to answer the question on how companies can stay in business if they are not selling new products. Is the market (even in a non-recession) willing to pay 3 or 4 times what they pay now for a cell phone? The Symposium did, though, try to make an argument for putting “green” as part of the accounting value statement. Software vendors are trying to collect data on all environmental costs associated with a product: mining, growing, shipping, processing, packaging, all the things that use energy before the end item is actually shipped, and then how much energy is needed to dispose of said product. A new science termed LCA (Life Cycle Assessment) is being employed in many industries to try to capture this data and put a value to it. I wouldn’t be surprised if graduates start appearing with LCA degrees.
Computer InterfacesThe mouse and keyboard are dead. The physical keyboard and mouse are being replaced by virtual keyboards (like on the iPad) and visual tracking like used by the Xbox Kinect. Because of new technology like Kinect, typing in mid-air is possible on a virtual keyboard. The intelligence of auto correct is getting better (in theory at least) so typos are not a huge concern. The point here is because this change affects the physical layout of offices. Monitors will be replaced with touch screens and cameras will be everywhere. What does that mean for the security of your work environment?
The "Cloud"What discussion about the future of engineering software is complete without mentioned the word CLOUD? I don't know about you, but I'm as tired hearing about this nebulous philosophy as I am hearing about green washing. In my opinion, internet based cloud is not something I would like but knowing that computer hardware technology is also taking a chapter out of history I'm a bit more optomistic. Distributed desktop workstations are going the way of the dodo and being replaced with thin clients. Compute power is more scalable at the server level now rather than the desktop level. Running applications from the server and pushing the input and feedback across the network results in better performance than buying a faster workstation CPU or better graphics card. Compression technology makes this possible without maxing out bandwidth. Maintaining software is easier because there is only one installation to update (on the server). Performance is easier to track and when it lags, easier and cheaper to improve because only the server needs a hardware upgrade. AMD gave a great presentation on what they have available on the market today that already does this with HP hardware. And, unlike another cloud demo, AMD’s didn’t blow up because it was on local hardware, not on some amorphous server farm somewhere in the world that depended on an internet connection. But, that brings me to my second point. Microsoft is pushing their “cloud” solutions via the Azure platform. Microsoft may be the creating the new Windows killer OS. Are we ready for that infrastructure change? Just upgrading to Windows 7 is met with resistance in some corporations and industries.
Systems ModelingOne my more enjoyable aspects of COFES is the small group discussions. I attended one hosted by Allen Behrens of Taxal Limited. The discussion was about how modeling of physical systems is growing with each advance in computing power. No longer is a single FEA sufficient. All simulations require multi-physics. One step beyond mult-physics is systems modeling, where the results from a vibration and braking analysis in an automobile are used as inputs to the suspension system and then the suspension system design is used again in the vibration and braking analysis (and design). These are very complex interactions and very complex systems to model. The only way to model them is with mathematics (the partial differential equations kind of math). Matlab with Simulink kind of does this; Simulia kind of does this; Labview kind of does this. The solution to the problem of systems modeling is creating a library of domain-specific functions so each engineer doesn’t have to reinvent the math (aka wheel) for each analysis. Modelica is a solution provider trying to do just that in an open environment. This is a software tool that is definitely worth watching.
ConclusionIf you can't attend COFES, at least try to keep up with the tweets, blog posts, and other media in order to find out what happens there. Not all of COFES will apply to you, but what does apply is undoubtedly beneficial for future proofing your organization. There are more topics discussed at COFES that I hope to blog about specifically in the near future.
Links to more COFES 2011 info:
twitter hashtag #cofes2011
Siemens PLM Blog