Kersplody was amongst four emerging Internet of Things (IoT) technologies taking the top prizes at the 2016 CES IEEE global competition. Understanding and using IoT is a focus for enterprises where it is essential to perform real-time interactive analysis integrated with automated decision systems. Moreover, security throughout the organization and on the edge is paramount. I had a chance to talk with Kersplody co-founder Chris Howard about his recommendations for the enterprise.
Who is Chris Howard?
Christopher Howard is an entrepreneur passionate about transforming how people interact with distributed systems. He is the CEO and co-founder of Kersplody, a start-up that solves many of the integration challenges involved with cyber security and big data deployments. At his previous position at Lockheed Martin, Howard developed software and solutions for numerous intelligence, civil, and defense projects including those involving big data, Multi-Int sensor data fusion, cyber security, and net centric warfare.
To listen to the full interview you can go to the non-profit ACM Learning Center podcasts or click on this MP3 file link in the learning centre
Here are extracts from the full interview which is long:
Ibaraki:
You are an authority in big data, Multi-Int sensor data fusion, cyber security, and net centric warfare. What are the top trends? What can enterprises do?
Howard:
Fast data is the emerging concept that is the next step beyond big data, and the best way to describe it is that you need to not only be able to work with terabytes, gigabytes or exabytes of data, but you also have to make decisions on that data really quickly as the data is coming in. This is compounded with the challenge that the things that are producing the data are producing it in exponentially increasing amounts so this has had to do with the explosion. It’s really a different type of big data processing that we are dealing with right now, where instead of building your machine model in a central environment, testing it, then pushing it into the live environment. You almost want to get to the point where the analysts are actually working with the real live data as it’s streaming in.
You need to have all the tools and capabilities in place so that analysts are able to react to it almost immediately. So the faster you can make the business decisions, the more advantage you are going to have as a business. Businesses should start to think of things in terms of intelligent analysis. Intelligence is your business intelligence data, sales data, competitors, the weather; it’s whatever relevant information associated with your business and your value stream, and how you understand how that is translated into some sort of business operation.
When you are starting to pull in all this data in real time and filtering it and trying to make sense of it, you really have to adopt practices that start to look an awful lot like Multi-Int fusion. Huge opportunities for the technology transfers and for companies that get there rapidly to get ahead of information processing processes — far beyond what some of the leading governments are doing right now.
Computers are now your cellphones, IOC devices and come with humongous security challenges. There’s always a huge trade-off between security and worker efficiency and trying to figure out that right balance is crucial in this business space. You associate that with the challenge that you have no edge to your network anymore because of all these new cloud devices and things are coming in, so there’s a humongous gap between the intrusion and the security systems that exist today.
Associated with that security challenge is a medium to encrypt everything as a standard. Right now encryption is the only protection method that we are aware of that properly protects data. Careful key management and encryption and separating the data across local networks is the only real approach that companies have to continue to keep data reasonably safe (or in the hands of as few people as possible). It’s really challenging because it’s the ultimate dual use technology — encryption makes your business considerably safer and more effective in managing control of who gets data for what reason and what times, and at the same time the same encryption can enable bad people to do bad things, so how do you properly deal with that balance between those two extremes, especially when you are dealing with legal and regulatory environments.
Lastly, you have the challenge of the internet being everywhere. That completely changes the way you collect data, the way you deal with security, the way you interact with and have to manage your data stream.
Ibaraki:
What are the top resources that help you in your work?
Howard:
One of the most useful resources I’ve come across is the combination of specific websites. In particular, Reddit and Ars Technica are probably the two websites I find the most useful just because of the community, people who are involved with it and the quality of the content is generally fantastic.
There are a number of really high quality podcasts out there that are good for getting a large number of perspectives on a number of different topics. I’m able to get perspectives that I normally would not come across in my daily life, especially when it comes to complex technologies and social issues that may need to have some bridge or divide.
Another great resource are Meetups. If you are in a large city area, meetups.com (and several others) is great to figure places to actually meet professionals in a number of different areas. In my area (which is the Denver area) some of my bigger meetups right now are the Denver Tech RIoT meetings (which is an emerging innovative things community that is growing out here); the New Tech group (which is a large group of entrepreneurs) and the Big Data group (that is really good for understanding the large data context and the activities that are going on in the big data and science communities).
There are also a number of associated groups that are more closely related to universities and other organizations. I’ve got a lot of good resources from the Data Science Association in the Denver area. They are interesting and also very concerned about the sociological implications of big data and what it means for society. And then combining that with data research because there’s a whole host of really interesting data science and challenges associated with that.
Google is always a good resource because you can find everything, and when you know how to use Google (especially Google’s advanced search), you can do an incredible job of finding information quickly. You just have to be very careful with Google filtering the results. As far as the technical side of things, great resources I’ve had are AWS being able to quickly spin a cloud environment to run an experiment relatively inexpensively; being able to hack things together very quickly with Python; a lot of good luck combining that with things like Docker, Virtual Machine (and a few other things) to be able to very quickly take an experiment from my personal PC and throw it out to a hundred worker machines out there on the cloud to run on a much larger datacenter.
The number one resource is personal fitness; and just getting outside and letting my brain integrate problems. Most of the fitness things I do are where I can go out without a cellphone or distractions and just sit back and think — it’s amazing the ideas and clarity you can get.
Ibaraki:
What are the top opportunities that are out there and why do you believe that they are great opportunities?
Howard:
Right now probably the top opportunity is cyber security. We can see that because there’s lots of churn in the industry and newcomers seem to have significant advantage over the established players in this space, and this is really because the traditional model of how we approach cyber security is completely falling apart. It’s a very interesting area for rapid growth and innovation for the next couple of years.
The cloud is moving to the edge of the networks, as opposed to the cloud sitting at the centre of the network. And that’s more or less going to be forced because you do not have the bandwidth necessary to get the data that you want to access and interact with to a central cloud to do traditional data science activities with it. So all the tools, technologies and capabilities necessary to handle that transition from the centralized network to the distributed network is really going to take a front seat in innovation. I suspect that’s probably going to get really hot around early 2017, however there’s going to be a lot of interesting early experimentation in the immediate future.
Associated with that is the command and control networks necessary to be able to manage and understand, and do data science on real time streaming data as opposed to centralized data at rest. Huge opportunities in the space right now have to do with how you deal with the next generation of workforce, now that traditional employment is starting to very quickly evaporate in a large number of areas of the economy. We are moving to the point where companies are having just-in-time labour, where you have a business and you treat everything as micro-projects with temporary hired labour, as opposed to using internal company resources sitting on the shelf. I can see that trend accelerating with the millennial generation, who have been forced to deal with challenging the labour situation across many of the very different disciplines, where they are contractors by default but where contractors are the norm not the exception.
In addition, now we have all these great digital resources that are out there like Wikipedia, OpenCourseWare, and the many micro academies. There are numerous things like this run by accelerator organizations where the most useful skills that people can learn are either being self-taught or being taught in micro academies as opposed to traditional university settings. Now you can get full access to courses online and all the materials related to those courses. The traditional university model is quickly becoming slightly obsolete, with the exception of the opportunities to make the personal connections and having access to some of the complex labs. It’s getting to the point where everybody right now on earth has access to almost the same resources, and resources and that combined with the opportunities to do short-targeted programs (as opposed to full university programs) significantly reduces the cost of getting a top-tier education.
Ibaraki:
Talk about your top ideas around building prototypes in promising new areas?
Howard:
I have talked a lot about big data and moving cloud to the edge and that’s really where my focus is right now. That is where I think there are the largest and most fascinating opportunities to be able to get the information that you need to make significantly better decisions with pretty much everything going to a sensor-driven economy.
Closely associated to that is the idea that data is being generated everywhere and very little is being captured and utilized today. We will always have more data than we can store and the gap between these is always increasing. How do we extract the right information from the right place at the right time? How do you take the data-driven economy and take it to the next level by changing the decisioning timelines from hours to days and change it from milliseconds to seconds and what you can do with those types of improvements? There’s a lot of interesting ideas and prototypes surrounding that, and the company I founded recently called Kersplody was created to start to experiment and look at challenges in the specific space.
Ibaraki:
We have many seasoned developers in the audience, please provide your top software engineering and developer tips?
Howard:
It really comes down to doing proper system engineering right now for all your software development. Understanding the context to your code as well as the cost to your code may slow down your ability to write raw code, but allows that code to be relevant for a longer period of time and generally be higher quality. That comes down to several very different specific points. Define what’s good enough and where that good enough point is so that you always know when to stop optimizing or stop coding and work on other projects that are associated with it.
The other thing that’s key to software development is you can de-complexify and write the best code you can with as few number of interfaces as possible, because the more interfaces and configuration points you have, the more complex your software product ends up being. It requires a lot of discipline and discussions with everyone involved in the software development project (not just the software developer) to create the design that enables reduction of interfaces and complexities so you are able to get to the simplest product possible and much more quickly to arrive upon an optimal solution.
Associated with that is even when you are doing ridiculously complex operations you can still have intuitive interface. Focusing on that and getting to the point where a brand new user can utilize the app almost as effectively as an expert user is really the gold standard of where things need to be. The user is always right. If users can’t use your product it’s not the users fault, it’s probably your fault and you can probably do something better about it.
Ibaraki:
You choose the topic area. What do you see as the top challenges facing us today and how do you propose they be solved?
Howard:
Cyber security. Airplanes are hacked, cars are hacked, vulnerabilities are in a breathtaking range. Combine that with the fact that there’s approximately 110 million Americans (or a quarter of the population of America) that got their IDs compromised in some way in 2014. This is a major issue that is ridiculously broken and the technology that’s out there is not keeping up where the companies need to be.
Most companies are just starting to realize how much threat they are in the middle of and how much risk they are actually exposed to. The solution to this space really has to come down to several things. Interoperability — different capabilities between the tools you need to defend your system and these may include: threat-sharing system, an endpoint protection system, audit system, analysis environment. There are various different classes of tools that all need to be combined together because there is no one vendor-fit-all solution, and all these tools need to fit together in a cohesive manner.
Associated with this, doing a much better job of getting a shared cohesive body of all of these threats so that a company can understand when they see a particular thing occurring on the network what is the appropriate action they should take and what’s the best remediation practice for it. The innovation needs to be how to put the right capabilities in there to enable either humans in the loop or human-on-the loop type activity with either full or partial automation, with the end goal being able to effect the right decisions in the shortest timeframe possible.
I think it’s an industry-wide problem that needs to be fixed going forward because we need to get away from everyone having a unique cyber-security solution, and get to where we have at least a common set of standards where you can share data between those individual solutions and make it so the solutions start to look more common and interchangeable. Cyber security is really expensive and complex to do right and we need a revolution to make it affordable and accessible to everyone (not just the companies that are willing to spend 10 to 20 millions of dollars a year protecting every single computer interface within their company). It’s a really interesting complicated area where I’m excited to see the innovations that are coming out.
Ibaraki:
Chris, you have so many outstanding contributions with Lockheed and now with your own ventures. Thank you for sharing your considerable expertise, deep accumulated insights and wisdom with our audience. Please describe your journey from a young age, milestones in that journey and some lessons learned.
Howard:
Some of my first memories involve me taking toys apart, trying to figure out how everything worked and then almost getting things together, but always having some remaining parts. I was always into math, science and engineering and really neat things. I got my first computer experience when I was about five years old, programming what I believe was on a Sinclair system. I was fortunate to be in a place where I had a lot of early opportunities to get involved with technology.
Another big part of growth that allowed me to gain a lot of perspective and wisdom very early on was being part of one of the best Boy Scout programs in the Pacific Northwest. It was an environment that was youth directed and lightly supervised, so we had a group of people that very quickly had to learn leadership skills and direct a very entertaining program that everyone was able to be a part of. Because of this environment the older Scouts always mentored the younger Scouts, and there was always energy and new ideas being interjected into the process which made it so everyone really added value to the overall experience of the team. My expertise ended up being orienteering and I helped both teach and lead my troops to multiple victories in orienteering challenges. As well as teaching responsibility to the team, you had to have the mentality that no person got left behind and everyone in the group got to participate even if they weren’t as strong or as smart, so you could get some team cohesiveness. Out of that you ended up getting some of the best leadership skills you could have gotten.
Middle school and high school I was also very deep into our jazz band program. We were learning music skills, but the more important thing was we were learning by association what is sometimes called the Jazz Fusion process, where you are part of a group and there were rules that you have to follow but everyone is a allowed to improvise on top of the basic structure (and even that basic structure can be modified as long as everyone follows the rules). By doing that, magical things happen and you are able to make incredible music out of something that on its face would appear like complete chaos.
Towards the end of high school, I got my first job right at the end of the first tech bubble as an IT intern at an Oregonian newspaper, this gave me a very unique and early perspective on business. I earned my degree at the University of Colorado, and while doing that I participated in the Space Grant program helping to put together the Three Corner Satellite (and a couple of other satellites for the school program). It was a very interesting introduction into science and mathematics. I then did an internship at NASA at the Dryden Flight Research Center and there were a lot of really cool projects that were going on there at the time so it was an excellent learning opportunity.
I interviewed and got hired at Lockheed Martin. It was timed really well when they were doing mass growth of their teams and I got hired for a program that was cancelled a month after I was on board. I then switched over to the Space Radar program as a Multi-Int engineer and was part of a team that was in charge of taking data from multiple satellite systems, ground systems and aircraft systems and putting that information together in a way that would make sense to someone consuming the information. That was an awesome introduction to the world of large scale, complex, chaotic data and many of the challenges that we faced back then actually have not been solved by the big data processes of today.
I then went into doing graphical modeling for satellites. I eventually transitioned and got involved with big data in the company and ended up leading the big data team for a Shark Tank team. That takes me very close to where I am today, where I got the opportunity to take an idea and run with it, and I am now in the process of trying to take some very advanced concepts in big data and put it all together in a way that really has the opportunity to change the world.
Last is figuring out the best way to push the policy to make things faster, better and cheaper while being careful never to push things that will annoy the customer or break laws or contracts. In addition, it’s knowing exactly where that line is because there are always opportunities to make things better and to push the system to accept change. It is also quickly learning to figure out when change is and isn’t possible and figuring it out through all the complex politics. Technology is easy, but its process and getting acceptance of new ideas is hard.