Welcome to the SecAware blog

I spy with my beady eye ...

12 Mar 2019

NBlog March 12 - pragmatic information risk management

Over the past ~three or four decades, the information risk and security management profession has moved slowly from absolute security (also known as "best practices") to relative security (aka "good practices" or "generally-accepted security") such as ISO27k.

Now as we totter into the next phase we find ourselves navigating our way through pragmatic security (aka "good enough"). The idea, in a nutshell, is to satisfy local information risk management requirements (mostly internal organizational/business-related, some externally imposed including social/societal norms) using a practicable, workable assortment of security controls where appropriate and necessary, plus other risk treatments including risk acceptance. 

The very notion of accepting risks is a struggle for those of us in the field with high standards of integrity and professionalism. Seeing the dangers in even the smallest chinks in our armor, we expect and often demand more. It could be argued that we are expected to push for high ideals but, in practice at some point, we have no choice but to acknowledge reality and make the best of the situation before us - or resign, which achieves little except lamely register our extreme displeasure.

Speaking personally, my strategy for backing-off the pressure and accepting "good enough" security involves Business Continuity Management: I'll endorse incomplete, flawed and (to me) shoddy information security as being "good enough" IF management is willing to pay enough attention and invest sufficiently in BCM just in case unmitigated risks eventuate. 

That little bargain with management has two nice bonuses:
  1. Determining the relative criticality of various business processes, IT systems, business units, departments, teams, relationships, projects, initiatives etc. to the organization involves understanding the business in some depth, leading to a better appreciation of the associated information risks. Provided it is done well, the Business Impact Assessment part of BCM is sheer gold: it forces management to clarify, rationalize and prioritize ... which gives me a much tighter steer on where to push harder or back off the pressure. If we all agree that situation A is more valuable or important or critical to the organization than B, then I can readily justify (both to myself and to management, the auditors and other stakeholders) mitigating the risks in situation B to a lesser extent than for A. That's relative security in a form that makes sense and works for me. It gives me the rationale to accept imperfections.
  2. BCM (as I do it!) involves investing in appropriate resilience, recovery and contingency measures. The resilience part supports information security in a very general yet valuable way: it means not compromising too far on the preventive controls, ensuring they are sufficiently robust not to fall over like dominoes at the first whiff of trouble. The recovery part similarly involves detecting and responding reasonably effectively to incidents, hence I still have the mandate to maintain those areas too. Contingency adds a further element of preparing to deal with the unexpected, including information risks that weren't even foreseen, plus those that were in fact wrongly evaluated and only partially mitigated. Contingency thinking leads to flexible arrangements such as empowerment, multi-skilling, team working and broad capability development with numerous business benefits, adding to those from security, resilience and recovery.

My personal career-survival strategy also involves passing the buck, quite deliberately and explicitly. I value the whole information ownership thing, in particular the notion that whoever has the most to lose (or indeed gain) if information risks eventuate and incidents occur should be the one to determine and allocate resources for the risk treatments required. For me, it comes back to the oft-misunderstood distinction between accountability (being held to account for decisions, actions and inactions by some authority) and responsibility (being tasked with something, making the best of available resources). If an information owner - typically a senior manager for the department or business unit that most clearly has an interest in the information - is willing to live with greater information risks than I personally would feel comfortable accepting, and hence is unwilling to invest in even stronger information security, then fine: I'll help firstly in the identification and evaluation of information risks, and secondly by squeezing the most value I can from the available resources. 

At the end of the day, if it turns out that's not enough to avoid incidents, well too bad. Sorry it all turned to custard but my hands were tied. I'm only accountable for my part in the mess. Most of the grief falls to senior management, specifically the information owners. Now, let's learn the lessons here and make sure it doesn't happen again, eh?

So that's where we are at the moment but where next? Hmm, that's something interesting to mull over while I feed the animals and get my head in gear for the work-day ahead, writing security awareness and training content on incident detection.

I'd love to hear your thoughts on where we've come from, where we are now and especially where we're heading. There's no rush though: on past performance we have, oooh, about 10 or 20 years to get to grips with pragmatic security!

Meanwhile, here are two stimulating backgrounders to read and contemplate: The Ware Report from Rand, and a very topical piece by Andrew Odlyzko.

No comments:

Post a Comment