Writing


Security Blog

  • Password weakness

    In the xkcd comic on password strength, Randall Munroe triggers a discussion on the effectiveness of password algorithms. While he raises valid points, the real risk lies in password reuse and breaches. Moving away from passwords and exploring alternatives like SSL certificates or one-time passwords may be a more promising approach.

  • How certificates go bad

    The recent Comodo sub-CA bogus certificate issuance has sparked a loud discussion in the security community. This incident highlights the flaws in the SSL certificate authority (CA) model. Trusting numerous CAs globally creates vulnerability, and compromised registration authorities (RAs) can issue certificates for any domain. Incident response and transparency are crucial, but incidents like this…

  • Architecting for DDoS -Defense

    Designing a robust DDoS mitigation strategy requires understanding potential failures and the efficiency of attacks. While increasing capacity can be a simple solution, it’s crucial to push functionality to the edge, prioritize user authentication, employ caching techniques, and store user-generated content. Instead of focusing on recovery time, design for a Minimum Uninterrupted Service Target to…

  • Awareness Training

    Implementing a robust security awareness program is not difficult if your company prioritizes security. However, if security is not a concern, you have a significant problem. Many programs focus on meeting auditor requirements with annual training sessions and policy acknowledgments. While these are necessary, a comprehensive program integrates security into various activities and encourages employees…

  • NSEC3: Is the glass half full or half empty?

    NSEC3 is a DNSSEC specification that addresses the issue of authenticated denial of existence in DNS. It replaces the NSEC method by using a hashing function and signed hash ranges to prevent easy collection of zone file contents. While NSEC3 improves secrecy, it raises questions about the need for semi-secret public DNS names and suggests…

  • Contracting the Common Cloud

    After attending CSO Perspectives, Bill Brenner has some observations on contract negotiations with SaaS vendors. While his panel demonstrated a breadth of customer experience, it was, unfortunately, lacking in a critical perspective: that of a cloud provider. Much of the point of SaaS, or any cloud service, in the economy of scale you get; not just in…

  • The Adaptive Persistent Threat

    Much ado has been made of the “Advanced Persistent Threat”. Unfortunately, pundits look at the ease of some of the attacks, and get hung up on the keyword, “Advanced.” How do we know the adversary is so advanced, if he can succeed using such trivial attacks? The relative skill of the adversary is actually uninteresting;…

  • Why is PCI so successful?

    At the RSA Conference, participating in a panel discussion on the PCI Data Security Standard, it was evident that PCI has significantly impacted the industry. The standard’s simplicity, broad applicability, and narrow focus on protecting specific data have improved security more than any other standard. It serves as a model for future compliance standards.

  • Why don’t websites default to SSL/TLS?

    HTTP is designed for web administrators to host multiple sites on fewer systems, while HTTPS (SSL/TLS) focuses on security-conscious users. SSL’s design creates scalability issues, although solutions like wildcard and SAN certificates, as well as SNI, aim to mitigate them, pending widespread SNI support.

  • Modeling Imperfect Adversaries

    Brian Sniffen’s paper at FAST highlights the importance of understanding adversaries in risk assessment, particularly in the streaming media space. It explores concepts like defense in breadth and tag-limited adversaries, emphasizing the need to consider different capabilities and attack strategies when formulating security frameworks. His work provides valuable insights into the evolving landscape of streaming…


Leadership Newsletter

  • Kremlinology

    Never attribute to malice what can be explained by stupidity.  Sometimes, your own! There is a natural human tendency towards kremlinology – that is, the attempt to impute motives by observing only a few characteristics or outputs.  In one application, it is called Fundamental Attribution Error, when we assert that someone has ill motives just because we… Read this …

  • The Future of Work

    Here we are.  Three to six months into CoviDistancing – call it lockdowns, social distancing, isolation, shutdowns – and, really, there’s no end in sight. Let that sink in for a few minutes.  It’s possible that there’s an effective vaccine just around the corner – which generally means a year of human trials so we know it’s reasonably safe. … Read this …

  • Moving to Distributed Work

    So, you’re working from home … For a while. You’ve probably worked remotely before, and you’re thinking, “I’ve got this!” Odds are, you’re mistaken. You don’t have this. That’s OK; this is an opportunity to learn new skills. You can think of working from home much like someone moving into an entirely new environment. Your… Read this …


Fiction