Cogosense Blog

Cogosense's Driver Behaviour Blog

Introduction to Cogosense - Part 3

cover_image

In an Introduction to Cogosense Part 2 I covered the transition from Aegis Mobility to Cogosense Technology, and as an aside I thought it would be fun to discuss the company names as an introduction to this post. When I joined with my cofounder to create Aegis, he had already picked the name, Aegis the great shield of Athena and Zeus. He had already registered the name, so I thought he was fair game for a good ribbing. It was a great name for our company, but I felt I had to point out to him. in Greek mythology the Aegis is actually the goat skin bag the shield was placed in. (To be fair it has been interpreted as both the bag and the shield, but being named after a piece of smelly goat skin appealed to my British sense of humour). When we named Cogosense I decided all forms of animal fur should be avoided, so I turned to our Roman friends for the verb Cogere (to collect or gather). Cogo means I gather, as in gathering data from sensors, hence the name Cogosense.

When we started Cogosense we were in the enviable position of having an unstable product, a loyal but fatigued customer base and a revenue stream in constant jeopardy. From day 1 we were self funded and our goal was to achieve product stability. To achieve this we had to drastically cut costs. We avoided executive positions and salaries, opted for virtual office space, and fully embraced the concepts of cloud computing and automated DevOps. We have no infrastructure of our own, we use SaaS services to deliver our own SaaS offering. The discipline required to implement this model of business has brought us to the point where I could pick up any employees laptop and drive a nail through the flash drive in the knowledge no business critical information would be lost and full productivity would be achieved within an hour of receiving a replacement laptop. I mention this because business continuity is a key ingredient of product stability. For me it has been a liberating process shedding the old methods for the new, biting the bullet and just doing it.

The FleetSafer product also under went a revamp. As mentioned in Introduction to Cogosense Part 2 Aegis was consumed by the burden of maintaining access to private APIs in iOS to perform the enforcement functions necessary. Cogosense decided to return to the roots of the product as described in Introducing Cogosense Part 1. Originally the concept had been a simple feature on top of a network that could perform the necessary policy enforcements. This feature we called the Aegis COntext Request Node, or ACORN . It was a node in the network that tracked the contextual state of millions of mobile devices. It provided contextual awareness to the mobile network.Contextual awareness can be defined as a simple question "What where when?". This ACORN platform was still the engine that enabled all our product offerings, it just need a complete repurposing. 

We could not return to a carrier integration, first and foremost working with carriers is expensive and time consuming, second, most communication are now App based, the carriers are carrying the encrypted data, making it difficult to sort and categorise communication sessions at the network layer. Thankfully in the years since we started Aegis, the Mobile Device Management (MDM) platform has emerged as the primary mechanism for managing and applying policy to large fleets of mobile devices. An MDM can be used to define what applications a device is allowed to run, what networks a device can connect to, to remotely lock and wipe lost devices to protect corporate data, and provide APIs to automate actions usually performed manually in the MDM user interface. What MDM's lack is contextual awareness. 

The first thing we did for FleetSafer was pull out all the private API's. We then added an MDM integration layer to enable FleetSafer to be a contextual layer on top of existing MDMs. This means FleetSafer can instruct an MDM to lock a device when it is moving, or it can instruct it to display a different home screen, with a subset of applications. The concept can be extended to time, geographical place, or bluetooth beacon detection. For example a tablet in the driver's cab of a locomotive can be configured to only show a time table or an operations manual, yet anywhere else in the train you can play Candy Crush. These capabilities can be provided on top of nearly any MDM. Currently AirWatch, Cisco Meraki, Miradore and MobileIron are supported, with Microsoft Intune support coming soon. Other MDM platforms are added as requested, the flexible integration architecture has been designed to reduce the time needed to add a new integration.

FleetSafer is the tool needed to put the emphasis on Mobile in the term Mobile Device Management. It was satisfying to come full circle and resolve the root cause of long term stability issues in the FleetSafer product. The MDM interfaces are highly stable, unlike private iOS APIs. We can now concentrate on the task of gathering sensor data.

10082 Hits
0 Comments

Introduction to Cogosense Part 2

Introduction to Cogosense Part 2

Introducing Cogosense Part 1 of this blog described the initial idea of Aegis, and how due to small thing called the iPhone, we got left in the dust as carriers rushed to shore up their networks. Part 2 will be dedicated to the calm and calculated pivot that Aegis executed.

First our CEO resigned. He was my cofounder, and is still a good friend, so no names will be used here. The consequences of this action would make a great case study on how not to handle an investment in a tech company. Suffice it to say a cofounder will treat an investor's money as if it was his own. Founder and Angel Investor interests are usually completely aligned. In the best interests of the company we should have stuck a paper bag on his head and propped him up in the corner till he calmed down.

We did carry on, the solution was modified to remove the carrier integration and an "over-the-top" solution was developed. Instead of intercepting voice, text and data media streams in the network (where they should be intercepted), we wrote software to intercept them on the handset after they had been delivered. The FleetSafer app was born. This worked very well on Android, and not so well on that iPhone device that had already caused us so much grief. Well that's ok we thought, 80% of the market is Android, we just need an iOS solution that is good enough. What we discovered was, the 20% of iPhones were in the hands of the decision makers. Literally at one logistics company we lost a deal for 5000 truck mounted Android tablets because the CEO had an iPhone and we couldn't prevent his device ringing.

This led to a lot of pressure to solve the problem, which led to the abyss of "private APIs". Anyone familiar with iOS will know that Apple only condones the usage of published documented APIs, but with a bit of digging you can find a wealth of undocumented functionality. By the time iOS 5 was released in 2011, we could make iOS match Android feature for feature. The problem was in iOS 6 Apple commenced waging a war against illicit usage of its devices. It seemed a funny concept back then, not being able to use a device you owned as you saw fit, but Apple doubled down with each subsequent iOS release. The amount of money poured into staying ahead of the game became more than Aegis could bear. In February 2015 the company shutdown, the investors cut their losses. I don't blame them, but I think a paper bag could have saved them a lot of money.

A dedicated team of ex employees kept the services going for the Aegis customers. Freed of the burden of executive salaries, leaseholds and server farms, we reduced monthly expenditures 10 fold, we migrated everything to the cloud and automated all key tasks. By June just 5 of us were keeping the lights on. The decision was made to found Cogosense in September, and negotiations started with Aegis to acquire the technology. By Christmas we had a deal.

From this whole experience I discovered it is extremely hard to execute a pivot when you lose half your vision. Over the years of Aegis it became clear Distracted Driving is just not enough to justify a standalone product, but the idealogical inertia had set in at senior levels and we couldn't shake the self image we had created of being a "Distracted Driving" company. The two of us had it right in the beginning, it is a "call forward on drive" feature. I always wonder what would have happened if we had applied the paper bag, kept calm and carried on.

In part 3, I will talk about Cogosense: what we do and what we did with FleetSafer.

9940 Hits
0 Comments