Thermopylae Sciences and Technology provides a variety of geospatial products and solutions that build on DoD research and development as well as their Google Enterprise Partnership. Sensors & Systems (S&S) editor Matt Ball spoke recently with AJ Clark, president of TST, about the company’s work as a Google developer and reseller, and the move toward mobile.
S&S: The GeoInt Symposium is right around the corner, and it will be interesting to note the impacts of sequestration.
Clark: GeoInt is the first big conference for the Intelligence Community since the shutdown started and it comes on the heels of sequestration. It will be interesting, and I think it will be dominated by a lot of uncertainty. I doubt that we’ll have a lot of progress on the budget by that time. I know there will be reduced attendance, but I think that we’re doing well to be having a conference at all, given that some of the larger government shows have been canceled.
I am actually very excited to attend. I think you’ll see a turning point in the industry that will be marked by innovation. Companies that can provide common denominator capabilities across multiple defense services, intelligence agencies, and Fortune 500 companies are the ones that will start emerging as leaders for the next decade. Future spending is projected to focus more on commercially available technology that is ready now, versus spending vast sums on “build your own” solutions so I’d keep an eye out for those starting to get some more momentum as well.
S&S: You’re heavily involved in mobile development. Is that your main focus with the GeoInt community?
Clark: Our company background is in GeoInt, and after six years of operation there are two niche things that give us a broader context in the geo world. We aggregate geo data from imagery to the data on top of it, and to the analytics that you can run on top of that data and imagery. We are also focused on bringing GeoInt to a broader audience, making it relevant to the other 90 percent that weren’t as focused on geospatial technology or information, and helping organizations get a better return on that investment with better collaboration.
That brought us to mobile, because once you have all that data in one place, you start thinking about bringing the data and map to the context of your location. Mobile came of age right at the time that we were starting to mature some of our geospatial solutions. It was a natural evolution of how we extend the capability to an environment where geo really matters.
When you’re moving around, it really matters. As you travel, you watch Google Maps on your phone, and it has more relevance than if you were just sitting in an office.
S&S: You’re a Google Enterprise Partner, so you’re developing a lot on Google platforms and solutions, which must be nice considering how active they are with map updates and adding location functionality.
Clark: We look a maps as in oil exploration, we thought we were running low on oil and gas until we started using new technologies and then we found new reserves because of that technology. Google is creating data and platforms that you can leverage and normalize and start mining that oil. It’s a whole new resource that requires someone like us to help customers exploit for their purpose. For us, the work that Google is doing with Google Map Engine and enhancing their products is the natural resource, and we come in to apply the power of their platform for specific markets.
With the mobile side, it’s a clear win for a government organization or business to bring all this capability out to the workforce in the field. In today’s age, both with the sequester and the sluggish economy, you have to come in with a story where you’re solving three or four problems with one cost and investment.
Google is a huge partner of ours, not only can we leverage the new technology that they’re bringing, there are also a lot of value-added partners that we can leverage. Google provides that new oil, with all their platform development, data creation, and the cloud infrastructure that they have invested in.
S&S: One of the criticisms of tools like the Google Map Engine is that’s it’s great visualization tool, but it’s not a place to really visualize your own data. What is your experience there in terms of integrating an organization’s data resources?
Clark: Google has come a long way with their Google Map Engine since it was first offered. They are solving a lot of data management problems. Many commercial organizations are willing to put their data in the cloud, and don’t have as many restrictions as the federal market.
Geodata management is tough to deal with no matter who you are or where you’re at. Scaling database technologies like Oracle Spatial, PostGreSQL and PostGIS has its challenges. Google has evolved the science in how to handle geo data at scale.
You will find people that aren’t comfortable exposing their data to Google to provide that service for them. I think there are more than enough that it will drive that technology further, and will mature it to a point where certain organizations will have more of a willingness to use that cloud capability because the risks are outweighed by the benefits.
With our services perspective, we try to address multiple issues dealing with data. It could be anything from imagery and terrain data to geo data on top of raster and vector layers to dynamic data of a workforce or soldiers moving around. We look at geo data as dynamic and static, with dynamic being near real time.
I think there are compelling solutions that Google offers on an enterprise level, with the Google Earth Enterprise Server that you can put behind your firewall as well as portable servers where you can miniaturize the server that manages imagery and terrain and put it on as small of a device as an Android phone with a secure operating system. You can load Google Earth globes on a mobile to take the data to the field.
S&S: Are you doing work on other platforms and technology outside of Google?
Clark: We work with MongoDB and with an open source Apache project called Accumulo. We have a geospatial index where we’re trying to apply the same geo functions and PostGIS and Oracle Spatial to open source data stores than can handle hundreds of billions geo objects.
We’re working to address some of the challenges where Google hasn’t convinced organizations to bring their data over. At the same time, Google has improved the quality of their offering.
S&S: On the data integration side, one of the newer military concepts is that the data should be tailored to the individual, with the person receiving the data having control on what it looks like, and what they can see. Google talks a lot about the personal map. Do those two concepts mesh well?
Clark: I think that there are so many elements that we have to think about. The next level of the Web will be interesting. I think that we have to have a level of logic to it–some people call it smart applications, others call it contextual-based computing. Whatever you call it, we need to build systems that have an awareness of things that users commonly need. Our technology has evolved to the point where artificial intelligence helps us gain relevance to other information out there.
The NGA and others are really big on the smart data concept. I think where we’re going is leveraging contextual-based computing in a mobile environment where the data is tagged in such a way where we can give recommendations to users, and expose users to the data that is coming from that area. An example is if you’re at a battlefield and you have a collection of geofenced polygons that are showing you all the ISR sensors that are above you or around you and you can see video or images of what’s around you.
This is all based on how we provide a unique experience. Our data is growing as fast as our hardware is growing, and we need to provide the application to properly interact with it. That’s really where we’ve tried to make progress with our Ubiquity mobile framework where users are pushed applications based on data available and relevant to their mission.
S&S: Your iHarvest product is interesting. That looks to me like a configurable dashboard that looks out at the Internet and helps to understand trends.
Clark: iHarvest has many different things that it provides to a user or organization. We use iHarvest to collect some of that logic that I was talking about. Ubiquity will promote applications to users based on their activity, and iHarvest handles some of the contextual information. They are constantly looking for information on IEDs.
iHarvest can build up those models on thematic or spatial interests and also connect them with other users that have a strong similarity to them. Building communities of interest around topics is what the Web is currently doing almost ad-hoc. You have a blog about hunting dogs, and people go to that site and get to know each other. How can you automate some of those processes where you bring people to a community of interest, and link them to others that are uniquely connected to that interest so that they then become connected.
That’s a capability of interest to the military, but also to investment companies. A lot of times in these large organizations there isn’t enough synergy for them to connect to each other. There’s so much information coming in right now that we’re almost overloaded and we don’t think to walk down the hall to talk to each other.
iHarvest is helping bring people back together, like water cooler conversations that can be digitally enhanced, and digitally replicated.
S&S: We’ve touched some on the limits of mobile computing. With the soldier there are certainly areas where the connectivity is difficult. Is there a good way around connectivity limitations?
Clark: The thing about the network is that it continues to get better, and the world is universally working to build a better network. Until that day, we have to have elegant solutions with a mobile application that goes on or off when there is connectivity.
Ubiquity was designed to provide context-based functionality to users within an application when and where they needed it, but it also was designed to provide a mobile framework for any applications running on the phone where it optimizes the battery life of that device, and makes use of whatever connectivity is available. We throttle what kind of data exchanges take place based on the network capacity at the time.
The other item that we address is offline activity. The commercial world would generally build applications with the expectation that there would be connectivity with either a cellular network or a WiFi network. Our customers in the military certainly work in areas without a network, so we needed to elegantly design an application that works with our without network communications. Ubiquity does a great job of providing native applications that can function like Web applications with an ease of update, but at the same time if connectivity is lost they have 100% of the functionality and use of the application that they had before the connection dropped.
We cache local data, local imagery management, and also data collection that is stored locally. We continue operate so that when you do get connectivity you can synchronize data bi-directionally. That capability is something that also come in handy for some of the other vertical markets that we service, like oil and gas companies, emergency response, or insurance evaluators that go out after a storm has gone through. A lot of field operators find themselves having to work in a disconnected environment.
S&S: I’ve noticed from your website that you have been working with the Google Glass technology. Is that something that is very geo, and augmented reality, oriented?
Clark: Google Glass is definitely very geo, with four or five key functions that it performs right now. I’ve used it for directions and finding my way around, and it’s about 100 times better than using my phone to find my way around. The thing about Glass is that if you’re driving or on a bike there is a challenge trying to hold or position your phone. Google Glass provides a map that you can see, and you can hear clearly right from the speaker near your ear. You have both hands on the wheel, and you don’t have to fumble around with your phone.
If you think about the movements that your body has to go through to get an update from your phone versus Glass, it’s an order of magnitude of a hundred times less. I just have to look a little to the right, versus reaching down and grabbing something and touching the screen.
I think Google Glass will always have a strong geo component to it. We’ve interacted with a number of other wearable computing devices. Google is working to perfect four or five things, and have a ways to go with production and with developers working to build interesting applications and functions on top.
S&S: Where does your success come from, and are you optimistic about your future despite budget uncertainty?
Clark: We have had a lot of success providing capabilities for a broad audience, and then also allowing for the customization of what we’re providing. We are seeing a lot of adoption in the commercial world, building on top of what we developed for the military.
It’s an exciting time right now, and there’s a huge commercial market with definite applicability to the things that we’ve developed. It does take a lot of business savvy to incorporate that functionality into those markets, and that’s where we’re putting some effort right now.