M+E Daily

By Ian Hamilton, Chief Technology Officer, Signiant

The technological landscape of media production and distribution has transformed over the last decade, moving from mostly isolated to highly interconnected networks. Earlier media processing systems were purpose-built and confined to a particular organization, with the only interactions implemented through physical media transfer or over dedicated audio and video connections.

In contrast, modern systems are hyperconnected networks of media processing software that span the globe and mostly run on commoditized IT infrastructure. Hyperconnectivity refers to the increasing digital interconnection of people and devices around the world. Cisco recently predicted that there will be 50 billion networked devices by 2020.

This level of connectivity will impact our lives at many levels, and the expanding use of commodity IT is largely driving its spread.

Because commodity IT is mass-produced, standardized and widely interchangeable with similar types of hardware, it significantly reduces costs and supports far more numerous Internet Protocol based interactions.

Being broadly compatible and interchangeable, commodity hardware can function on a plug and play basis and form the foundation for new connected technologies. The Internet of Things, for example, in part resulted from the reduced cost of WiFi interfaces and the ability to add them to a wide variety of devices; now everything from refrigerators and alarms to heart monitors and traffic lights are connected to the Internet.

Secure by design

Security experts have long advised designing software with security in mind from the start. However, many software vendors are overly dependent on “testing in” security after development through “black box” penetration testing. In fact, many vendors do not seriously consider security until third parties identify vulnerabilities during penetration testing.

Applying secure design principles from the start can greatly improve security in hyperconnected systems. Similarly, the effectiveness of penetration testing can also be enhanced by how it’s conducted. Contrasting “white box” vs. “black box” penetration testing, “white box” testing starts with a detailed understanding of the implementation, where “black box” testing treats the internal implementation as a mystery and typically utilizes a standard set of tests.

A “black box” approach to “testing in” security is insufficient because it typically only finds well-known vulnerabilities.

Understanding how a system is designed and targeting attacks based on that understanding is a far more effective way to validate the security of a system.

On the organization side, many do not scrutinize the security designed into their software and instead rely on building a single impenetrable barrier around their network.

However, in a world where cyber criminals are getting more and more sophisticated while network connectivity is increasing, the perimeter of an organization’s network is very hard to define and can be compromised through side channels that no one ever considered.

The recent cyber attack on Target’s network that stole millions of credit and debit cards is a good example.

According to published accounts of the incident, the hackers got in through an email phishing attack on Target’s heating, ventilation and air conditioning (HVAC) subcontractor, an unsuspected inroad that’s one of many ways an opportunistic cyber criminal can exploit hyperconnected systems.

“Secure by design” or the application of secure design principles during software development is more important than ever for modern, hyperconnected media processing systems.

Here are four secure design principles that are especially critical today:

1. Psychological acceptability
“Physiological acceptability” is a secure design principle that emphasizes the importance of users buying into the security model of your software. But, because connectivity has created a plethora of software choices easily downloaded off the Internet, modern users have very high expectations around usability and features.

If people can’t easily understand the security protection features of your software, they will likely make mistakes when trying to implement them. And if your software’s security mechanisms are inhibiting people from doing their jobs or are even a bit too tedious, users will find ways to bypass them or use alternative solutions that are not IT sanctioned. In fact, the number of unsanctioned and unsecure apps is on the rise.

Netskope’s April 2015 Cloud Report indicated that organizations have an average of 730 cloud apps in use; and over 90 percent of those do not have enterprise class security. Even for smaller media organizations, enterprise grade security is necessary when dealing with hyperconnected, global systems. In essence, it’s important to choose media processing software that is both user friendly and secure.

3. Fail secure
One way nefarious hackers try to breach a system is to find ways to cause it to fail or crash and exploit the resulting failed state.

Within hyperconnected systems, there are more ways for cyber criminals to try and make systems fail. “Fail secure” is a secure design principle that ensures that if a system or component of a system fails it fails in a secure state that can’t be compromised rather than an insecure state that can be easily compromised.

Basically, you have to make sure that if your software does fail, it fails in a secure state.

4. Open design principle
Businesses that depend on obscurity of their software implementation for security are increasingly at risk in a hyperconnected world. It’s much easier to access information, so it’s getting harder and harder to keep secrets. The “open design” principle states that the security of a system should not rely on keeping its design or implementation a secret.

Rather, it encourages designing security functions that maintain security even if the design is known. For example, how various encryption algorithms work is well known, but a specific application of an encryption algorithm depends on an external key to perform the encryption. Keeping the encryption algorithm a secret isn’t important, but keeping the external key used for a particular application of the encryption algorithm is.

In our hyperconnected world where media companies are handling highly valuable assets that have proven to be targets for cyber criminals, it’s essential to make sure that the software you use has been developed using secure design principles. And while all secure design principles should be considered, these four stand out as being particularly applicable.

Click to read the article via .pdf

———————————————-

Ian Hamilton is a founding member of Signiant, where he has led the creation of Emmy award-winning software solutions to address the challenges of fast, secure content distribution over the public Internet and private intranets for many of the media and entertainment industry’s largest companies. Having previously worked with ISOTRO Network Management, Bay Networks and Bell Northern Research, Ian has been developing internetworking infrastructure and applications for more than 20 years.