Looks like an interesting book, even if they’re talking about the old way of making software and making that safer the real experience used for making this book is useful.
“Unfortunately, the inherent invisibility of reliability and security in the absence of emergencies means that they’re often seen as costs that you can reduce or defer without immediate consequences. However, the costs of reliability and security failures can be severe. According to media reports, data breaches may have led to a $350 million reduction in the price Verizon paid to acquire Yahoo!’s internet business in 2017. In the same year, a power failure caused key computer systems to shut down at Delta Airlines and resulted in almost 700 flight cancellations and thousands of delays, reducing Delta’s flight throughput for the day by approximately 60%.”
Simpler design is better. Using subsystems.
Security by scoping by space and application layer encryption. For insider, least privileged.
On the developer side, code that has been verified can deploy, use frameworks, testing. Duh. Progressive deployment of patches is a good idea.
Don’t make logs contain sensitive information.
Have a plan for when an incident happens, as its too complex to face otherwise. Chain of command.
They have a procedure for it but I couldn’t find it, I did find this though:
https://services.google.com/fh/files/misc/google_security_wp.pdf
“Security and reliability have a lot in common—both are inherent properties of all information systems that are tempting to initially sacrifice in the name of velocity, but costly to fix after the fact.”
> Governments may also break into systems to police domestic activity. In a recent example, NSO Group, a cybersecurity contractor, sold software to various governments that allowed private surveillance of communications between people without their knowledge (through the remote monitoring of mobile phone calls). Reportedly, this software was intended to surveil terrorists and criminals—relatively noncontroversial targets. Unfortunately, some of NSO Group’s government customers have also used the software to listen in on journalists and activists, in some cases leading to harassment, arrest, and even possibly death.3 The ethics of governments using these capabilities against their own people is a hotly debated topic, especially in countries without strong legal frameworks and proper oversight.
Different kind of actors might want to hack your website, understanding how they operate helps you understand how to defend against each of them.
I like the perspective of considering everyone contributing to an open source project as insiders. Also interesting to hear the kind of stories that happen with employees unable to handle their emotions and themselves, with this being a possibility we have to accept the system should be made in such a way to only give the access necessary, this can of course only be true if we go with a “going from scratch” mentality every so often and take the time to plan it, which is possible in this AI age for sure.
It may not be immediately obvious that your company, organization, or project is a potential target. Many organizations, despite being small or not involved in handling sensitive information, can be leveraged to carry out attacks. In September 2012, Adobe—a company best known for software that enables content creators—disclosed that attackers had penetrated its networks with the express intent to digitally sign their malware using the company’s official software signing certificate. This enabled the attackers to deploy malware that appeared legitimate to antivirus and other security protection software. Consider whether your organization has assets that an attacker would be interested in, either for direct gain or as part of a larger attack on someone else.