It’s easy to say that every year’s been a big year, but in our case we’ve got the goods to back it up. Aside from doubling the size of the Securosis team, I added a new member to my family and managed to still keep things running. With all our writing and speaking we managed to hit every corner of the industry. We created a new model for patch management, started our Pragmatic series of presentations, popped off a few major whitepapers on application and data security, launched a new design for the site, played a big role in pushing out the 2.0 version of the Cloud Security Alliance Guidance, and… well, a lot of stuff. And I won’t mention certain words I used at the RSA Conference (where we started our annual Disaster Recovery Breakfast), or certain wardrobe failures at Defcon. On the personal front, aside from starting my journey as a father, I met Jimmy Buffett, finally recovered enough from my shoulder surgery to start martial arts again, knocked off a half-marathon and a bunch of 10K races, spent 5 days in Puerto Vallarta with my wife, and installed solar in our home (just in time for a week of cloudy weather).
It’s been a pretty great year.
I’ve never been a fan of predictions, so I thought it might instead be nice to collect some lessons learned from the Securosis team, with a peek at what we’re watching for 2010.
The biggest change for me over the last year has been my transformation from CTO to analyst. I love the breadth of security technologies I get to work with in this role. I see so much more of the industry as a whole and it totally changed my perspective. I have a better appreciation for the challenges end users face, even more than as a CIO, as I see it across multiple companies. This comes at the expense of some enthusiasm, the essence of which is captured in the post Technology vs. Practicality I wrote back in July.
Moving forward, the ‘Cloud’, however you choose to define it, is here. Informally looking at software downloads, security product services and a few other security related activities over the last 30 days, I see ‘s3.amazon.com’ or similar in half the URLs I access. This tidal wave has only just begun. With it, I am seeing a renewed awareness of security by IT admins and developers. I am hearing a collective “Hey, wait a minute, if all my stuff is out there…”, and with it comes all the security questions that should have been posed back when data and servers were all on-premise. This upheaval is going to make 2010 a fun year in security.
2009 for me wasn’t a whole lot different than the past couple of years from a consultative role. Although I probably pushed the hardest I ever have this year to build security in as architecture (not as an afterthought) I still, quite often, found myself in a remediation role. Things are changing – slowly. The enterprise (large and mid-size) is very aware of risk, but seems to still only be motivated in areas where it’s directly tied to monetary penalties (i.e., PCI and the government / defense side). I hope next year brings better balance and foresight in this regard.
As for 2010 I’m going to agree with Adrian in reference to the ‘Cloud’ and its unquestionable impetus. But it will still be an interesting year of pushing the seams of these services to the limits and finding out where they don’t hold water. Mid to late 2009 showed me some examples of cloud services being pulled back in-house and the use case considerably reengineered. 2010 is going to be a good year for an oft quiet topic: secure network architecture – especially with regards to services utilizing the ‘Cloud’. The design and operation of these hybrid networks is going to become more prevalent as network and transport security are continually hammered on for weaknesses. I’m sure it’s safe to say we’ll see a few cloudbursts along the way.
My research moved in a bit of a different direction than I expected this year. Actually, two different directions. Project Quant really changed some of my views on security metrics, and I’m now approaching metrics problems from a different perspective. I’ve come to believe that we need to spend more time on operational security metrics than the management and risk metrics we’ve mostly focused on. Operational metrics are a far more powerful tool to improve our efficiency and effectiveness, and communicate these to non-security professionals. If after decades we’re still struggling with patch management, it seems long past time to focus on the basics and stop chasing whatever is sexy at the moment. I’ve also started paying a lot more attention to the practical implications of cognitive science, psychology, and economics. Understanding why people make the decisions they do, and how these individual decisions play out on a collective scale (economics) are, I believe, the most important factors when designing and implementing security.
I learned that we shouldn’t assume everyone has the basics down, and that if we understand how and why people make the decisions they do, we can design far more effective security. On the side, I also learned a lot about skepticism and logical fallacies, which has heavily influenced how I conduct my research. Our security is a heck of a lot better when it’s mixed with a little science.
In 2010 I plan to focus more on building our industry up. I’d like to become more involved in information-sharing exercises and improving the quality of our metrics, especially those around breaches and fraud. Also, like Hoff and Adam, I’m here if Howard Schmidt and our government call – I’d love to contribute more to our national (and international) cybersecurity efforts if they’re willing to have me. We need to stop complaining and start helping. I’ve been fortunate to have a few opportunities to work with the .gov crowd, and I hope to have more now that we have someone I know and trust in a position of influence.
This year I learned a lot about database security (thanks, Adrian) and more about DLP too (building on what I had previously learned here). I picked up quite a bit about cloud security (thanks, Rich & CSA), but I’m still not sure how much you can really secure keys and data on VMs in someone else’s physical control & possession. So I guess Securosis is serving its purpose – it was founded primarily to educate me, right?
Sadly, it hasn’t been a good year for our federal government. The long-empty cyber-czar post (and the improved but still inadequate job definition) is clearly the responsibility of the Obama administration. So are 2009’s many failures around health-care and banking reform, and the TSA’s ongoing efforts to prevent Americans from travelling and to keep foreigners away – most recently by assaulting Peter Watts and with their magical belief that passengers who don’t move their legs or use their hands are safer than people who are allowed to read and use bathrooms.
This year, I learned a lot about the differences between risk management in theory and risk management in reality. In particular, I came to the conclusion that risk management wasn’t about predicting the future but rather about obtaining a more informed opinion on the present state of being of your organization. I also learned a lot more about my writing style and how to be a better analyst.
In 2010, I plan on continuing to focus on outcomes rather then controls and trying to figure out how to help organizations do so while simultaneously dealing with a controls focused compliance program. Should be interesting to say the least. I’m also looking forward to other companies releasing reports along the lines of what Verizon has done this year and in 2008. In particular, there should be some interesting things happening in January. Can’t wait to get my hands on that data.
We hope you have a great new year, and don’t forget to check back on Monday, January 4th – we have some big announcements, and 2010 is shaping up to be a heck of a year.