Wikinomics and Personal Creativity

The book Wikinomics by Don Tapscott explores the exploding trend of all members of society to create and distribute their own products. Wikipedia, YourTube, Blogs, and the Open Source communities have all created the tools that allow everyone to be a creator, not just a consumer of products. He feels that the generation that is now between 10 and 25 will carry with them a mindset of personal creation and modification of all of the products they use. Many of our training system users seem to already have this type of mindset. But perhaps the next wave of users will have it to a higher degree. We should expect future users to stretch the composability of systems like OneSAF to their limits

The Unreasonable Man

“The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man.”

Man and Superman, 1903, George Bernard Shaw

The Long Tail in Training Systems.

Chris Anderson’s book, The Long Tail: Why the Future of Business is Selling Less of More, explains a new side of the economics of music downloads and other digital businesses. In summary, when a product is digital and requires no physical storage, production, and shipping, then it is possible to profitably deliver millions of songs, movies, games, and software apps. We are not limited to the Top 1,000 in each category because we have to recoup costs associated with physical products.

The idea can also be applied to military training systems. We have historically focused our training systems on the people who go into combat and have done so with projects that are $50M or larger. At this price tag we can only afford to address the needs of a small portion of the military – hence the focus on those who are in harm’s way. But if we could lower the cost of training systems significantly, we could potentially acquire training systems for every single Army MOS. In the future, when every soldier has a laptop computer and every unit has a decent network connection, there could be a training application on every single Army desktop. These applications could be as ubiquitous as MS Office. To accomplish this, the training systems must be much smaller and less expensive - $1M, $100K, or even $10K to develop. If we can create valuable training systems at these prices that run in a desktop computer environment, then we may be able to serve all 400,000 soldiers in the Army.

Anderson’s original article in Wired Magazine and his Blog.

Roger Smith's "The Long Tail in Military Simulation Systems" article which explores in more detail how this can be done.

ISR Modeling Categories

In preparing a recent presentation for the IO-ISR Forum at the Simulation Interoperability Workshop, I listed a dozen simulations of Intelligence, Surveillance, and Reconnaissance; ordered them by time; and attempted to identify categories to characterize the evolution of ISR modeling over the last few decades. The categories that I arrived at are:

• Role Player. Initially intelligence was created through the actions of role players who injected information into the training event in the form of voice phone reports, textual reports, and map overlays.
  
• Sensor Models. The first step in modeling ISR was to create sensor models for the native combat objects in the game. These usually posted enemy ID, location, and status information to a global blackboard in the system that made it accessible to all units and players. Examples: JESS/CBS, ADSIMS/AWSIM, ENWGS/RESA.
  
• Intel Side-car Models. The next step was to create side-car ISR models that operated unique units in a model that was separate from the combat model. This allowed separation for classification issues, independent development of ISR sims, and adoption of existing sims that were used in the testing world. But it also created dual environments that were not always well synchronized. Examples: TACSIM, NWARS, JOISIM, JECEWSI, FIRESTORM, IEWTPT.
  
• Extensions. A few ISR simulations have been developed as extensions of the combat model. These use the same infrastructure, data, and models where possible; but add classified data and models as necessary. Examples: BICM, WIM.
  
• Integrated. I was not able to identify any ISR simulations that are truly integrated with their combat sim parent. In all cases, the ISR simulation is developed separately and evolves its own unique features that are not common with the combat simulations. Examples: None. Though I believe ENWGS had an integrated ISR model at one point, but it was removed.
  

The World Needs Only 5 Computers

Greg Papadopolous, the CTO of Sun Microsystems, has been getting a lot of play from the statement that the world only needs 5 computers. This plays against a (purported) famous statement by Thomas Watson Jr. of IBM, when he truly believed that there were only a few customers for big computers. Greg P. has an entirely different meaning. He sees the networking of the world as a force that will concentrate computational power into a few major centers that will serve up most of the services that people will need from computers. As examples he identifies the 5 providers of these services as Google, eBay, Amazon.com, Microsoft, and Salesforce.com.

Militarizing this idea, we could say that the Army only needs 3 computers. These would be networked “hyperscale machines” (Papadopolous’ term) that serve up: (1) Business IT, (2) Mission Operations, and (3) Training Events. As with Papadopolous’ list of 5, further digging will reveal that these 3 computers only cover 80-90% of what the Army needs to do. Perhaps there will be smaller machines for R&D and other functions. Yes, smaller. The amount of R&D computation is probably very small compared to the pooled operations in the top three categories of business, mission, and training.

Papadopolous’ original article

Simulation Products to Every Soldier in the World

As computer and network resources become more reliable within military units, it will become feasible for every soldier in the Army to have a computer and a connection from which they can access and make use of a simulation products. Delivering training services to the edge of the force will require hosting most of the computation at server sites and making it accessible through a lightweight client on the soldiers’ PC. Simulation will have to be configured as a service rather than as a product that is installed on all of these computers. The desktop client application might fall into three different categories: (1) clients that are web pages and Flash content that load in real time – like the Phosphor flash-based game, (2) clients that are relatively small but need to be installed on every computer – like Google Earth, and (3) heavy clients that are the size of a full game - like Americas Army. Delivering content to these customers will require large compute and data centers that manage dozens of servers-side applications and hundreds of scenario databases. These servers can also be connected to virtual simulators to allow the soldier desktop to interact with virtual simulators like CCTT and AVCATT.

The goal is to deliver training to every soldier in the Army when he needs it, not when it is possible for him to go to a training center. IT-based solutions work well for constructive and game-based simulations. They can connect to Live and Virtual systems for collaborative training as well.

Military Simulation Wikipedia

For decades the military has maintained catalogs of the simulation systems that have been developed, what their capabilities are, and who to contact to get them. The earliest version that I saw was a 2-column printed set from the late 1980’s. Since then, this catalog has gone electronic, distributed on floppy disks in the 1990s; and more recently has gone online at MSIAC. However, it is something that is easy to neglect and to let fall out of date. Perhaps we can turn authorship of the catalog over to the community at large using a Wiki. The Wikipedia has shown that such open communities tend to create documents that are accurate, detailed, and current – most of the time. The interests of the many serve to police the documents to eliminate or correct inaccuracies in the entries.

Robert Ballard on the Titanic and Future Explorations

Robert Ballard, located the Titanic and discovered geothermal vents on the bottom of the ocean. He pointed out that we spend more money exploring space in one day than we spend exploring the oceans in an entire year. His work has been supported by NOAA and the Office of Naval Research and has bridged exploration and military missions.

• Newly declassified fact – the search for the Titanic was a cover story for a classified mission. His team was really searching for the USS Scorpion, a nuclear submarine with nuclear weapons that was lost in 1968. The Navy was not exactly thrilled when he actually found the Titanic and brought in media attention from around the world.
  
• The hydrothermal vents explain the chemical composition of the ocean (why it is different from fresh water) and show that life can exist without photosynthesis. They may also point to the origins of life on Earth.
  
• Using unmanned robot submersibles he is now surveying the 50% of the U.S. land mass that is under the ocean, primarily the areas around the Hawaiian and Marshall islands. He also uncovered nearly perfectly preserved ships in the Black Sea and next year will return to try to locate and extract the preserved bodies of its crew members.
  

Robert Metcalfe on Military Networks

Robert Metcalfe, the inventor of Ethernet and founder of 3Com, presented some interesting ideas at a Navy conference. He described his perspective on where the Internet is going and provided advice on how to build military networks:

• The Internet is growing into Video, Mobile connection, and Embedded networks in devices. Video and Mobile connections are stressing the foundations of the Internet right now and he thinks it will take 5-10 years to build out the net so that it can properly handle all of this content. However, the growth of devices with embedded networks is more important. Last year 10 billion devices were sold with embedded computer chips, most of them do not have network connection. He believes that in the future all of these devices will be on the Internet. The current net is not ready to support this volume of traffic and TCP/IP is not an appropriate protocol for that last mile to the device. He is pushing Zigby as the protocol that will bring embedded devices to the Internet.
  
• The Internet was created by graduate students with no interest in security. He believes that the Internet is most vulnerable because it has no authentication of users at its lowest level. Routers do not verify the origin of messages, which allows spam and viruses to spread with little record of their originators.
  
• We do not currently monitor the status of the oceans, rather we sample it with a few missions that take data at one location at a given point in time. If we really want to understand the oceans by monitoring them, then we need an underwater Internet to make it possible for sensors to reside and report from there permanently.
  

Commercial/Game-based Low Cost Virtual Simulator

How close are we to being able to create an entire virtual simulator using commercial and gaming technologies? If we were to sponsor a competition to create a CCTT-like or AVCATT-like module (just the crew compartments and supporting computers, not the supporting tools) that trains the same skills, but is built using low-cost commercial items, what could industry accomplish? What price range could this reach? Is it possible to build an acceptable training device like a CCTT module using game engines, their associated development environments, computer peripherals, and low cost enclosures – which can be delivered for less than $50K each?

Interoperable Architectures vs. Simulations vs. Customers

Interoperability between diverse simulation systems has been a major focus since the 1990's. We have made huge strides since then. The limits that we experience in interoperability today are not due to poor designs of architectures or infrastructures. DIS, HLA, TENA, and others offer more opportunities for interoperability than we are able to exploit. Limitations in interoperability are generally due to the need to keep old (and expensive to replace) simulations in the inventory. There can be no magically engineered architecture that can pull together all of the extremely different data, model, and interface representations in dozens of existing simulations. To create a distributed simulation environment that is tightly linked (or highly interoperable) we have to create new architectures and new simulation systems that match each other. Tying old models to a new architecture or infrastructure will always be a suboptimal solution.

Staff Training via C4I System and Without a Simulation

Is it possible to create and run an entire staff training exercise using only C4I systems and the tools that they have on them or that can be installed on them? Could that community create their own training environment without relying on dedicated simulation systems and tools at all? What quality of event could they create if allowed to augment their standard C4I applications with a few additional applications or interface machines drawn from the public domain? One of the reasons I am interested in exploring the future of constructive simulation is that current C4I systems have many of the same tools and the units that use them may be able to train without traditional simulations.

Game-based training system lifecycle support

We see systems like DARWARS AMBUSH! pressing to become a program of record. But, with a game-based system that is burned onto a CD and largely self-explanatory, what type of lifecycle support will really be necessary for it? Perhaps these types of games can survive in the military with a much smaller organizational support footprint. Perhaps one Project Director could be responsible for 30 game-based systems deployed through the Army. What type of support does a game really need? Is it significantly different than our more traditional systems/devices?

DARWARS AMBUSH!: http://en.wikipedia.org/wiki/DARWARS

Data Mining: The 4th Way of Doing Science

Peter Freeman at NSF suggests that data mining will be the 4th way of doing science. For example, in astronomy, scientists will no longer look at the heavens to discover a new planet. Instead they will look into massive databases that have been collected. New forms of analysis will be able to discover where planets should be based on this data much more reliably than by looking through a telescope.

Simulation as the 3rd Way of Doing Science

Within DOE they have developed a new perspective on the importance of modeling and simulation. They call M&S “the third way of doing science”. The first way is through observation. The second is through direct experimentation. Faced with so many significant problems for which they cannot do experimentation or perform observation, they are turning to M&S as the tool to understand the behaviors of all of their systems. In their community a “simulation” is an exact, physics-based representation of the real system; a “model” is an aggregate or stochastic representation that estimates behaviors at a higher level because they do not understand exactly what is happening. M&S is the primary method of conducting energy research.

The origins of the phrase “the third way of doing science” seem to be in one of the two documents below

• DOE Publication
  
• Science Magazine
  

Idealized Design

Russell Ackoff, Univ. of Pennsylvania, has written a book called Idealized Design based on his experiences at AT&T in the 1970’s and several decades of consulting work with industry. In this case “design” does not mean the early phases of a product creation. Instead, he is talking about strategic transformation of a company or business unit. Rather than looking at where the business is now and designing forward to a future goal, Ackoff teaches his clients to imagine that the organization is right now in the ideal state they are looking to attain. Once that state is defined clearly, he has them work backward from there to where they are now. He claims that the advantage is that this prevents them from (1) being stumped by current organizational problems they are facing, (2) following dead end branches that traditional forward chaining methods would lead them down. It also retains their enthusiasm for the possible future throughout the exercise and execution because people are not immediately disarmed by the current state of the organization.

The concept, minus the process, is captured in: http://knowledge.wharton.upenn.edu/article.cfm?articleid=1540

MEMS for Weapon Orientation

We have studied the current state of MEMS devices that can serve as orientation and movement sensors. These devices are currently light, consume little energy, and cost little to produce. However, their accuracy is much worse than the larger, more traditional devices. This is the profile of a disruptive technology. The new devices have inferior performance in the area that is most important to current customers, so they do not adopt the devices. However, they provide a capability at such a low power, weight, and cost that an entirely new market will open up for the devices. The revenues from this new market, as well as global R&D, will improve the technology so that eventually it will provide better performance than the established systems and drive those out of business.

More Info: http://en.wikipedia.org/wiki/Disruptive_technology

Simulation as an IT Service

Simulation systems are becoming increasingly complex. This complexity limits our ability to deliver them to facilities and units. Because large systems require significant professional expertise, large hardware investments, and continuous software updates, there are very few sites that can host these effectively. Imagine that we deliver simulation as a service instead of a product. If the system were hosted in a central, professionally staffed and equipped location, its capabilities could be made available to all soldiers in the world through our growing networks. This would make products accessible to many more units and soldiers than can use them today. The system would also be staffed with the most experienced personnel available, no matter what the location of the soldiers. I have begun investigating the feasibility of delivering simulations in this manner. This must also include determining whether existing products are architected in a manner that will allow this to be done.