Happy Anniversary—Bang My Head Against A Wall
Over the last month or two I have received several invitations to go speak about cyber security. Perhaps the up-tick in invitations is because of the allegations by Edward Snowden and their implications for cyber security. Or maybe it is because news of my recent awards has caught their attention. It could be it is simply to hear about something other than the (latest) puerile behavior by too many of our representatives in Congress and I'm an alternative chosen at random. Whatever the cause, I am tempted to accept many of these invitations on the theory that if I refuse too many invitations, people will stop asking, and then I wouldn't get to meet as many interesting people.
As I've been thinking about what topics I might speak about, I've been looking back though the archive of talks I've given over the last few decades. It's a reminder of how many things we, as a field, knew about a long time ago but have been ignored by the vendors and authorities. It's also depressing to realize how little impact I, personally, have had on the practice of information security during my career. But, it has also led me to reflect on some anniversaries this year (that happens to us old folk). I'll mention three in particular here, and may use others in some future blogs.
In early November of 1988 the world awoke to news of the first major, large-scale Internet incident. Some self-propagating software had spread around the nascent Internet, causing system crashes, slow-downs, and massive uncertainty. It was really big news. Dubbed the "Internet Worm," it served as an inspiration for many malware authors and vandals, and a wake-up call for security professionals. I recall very well giving talks on the topic for the next few years to many diverse audiences about how we must begin to think about structuring systems to be resistant to such attacks.
Flash forward to today. We don't see the flashy, widespread damage of worm programs any more, such as what Nimda and Code Red caused. Instead, we have more stealthy botnets that infiltrate millions of machines and use them for spam, DDOS, and harassment. The problem has gotten larger and worse, although in a manner that hides some of its magnitude from the casual observer. However, the damage is there; don't try to tell the folks at Saudi Aramaco or Qatar's Rasgas that network malware isn't a concern any more! Worrisomely, experts working with SCADA systems around the world are increasingly warning how vulnerable they might be to similar attacks in the future.
Computer viruses and malware of all sorts first notably appeared "in the wild" in 1982. By 1988 there were about a dozen in circulation. Those of us advocating for more care in design, programming and use of computers were not heeded in the head-long rush to get computing available on every desktop (and more) at the lowest possible cost. Thus, we now have (literally) tens of millions of distinct versions of malware known to security companies, with millions more appearing every year. And unsafe practices are still commonplace -- 25 years after that Internet Worm.
For the second anniversary, consider 10 years ago. The Computing Research Association, with support from the NSF, convened a workshop of experts in security to consider some Grand Challenges in information security. It took a full 3 days, but we came up with four solid Grand Challenges (it is worth reading the full report and (possibly) watching the video):
- Eliminate epidemic-style attacks within 10 years
- Viruses and worms
- Denial of Service attacks (DOS)
- Develop tools and principles that allow construction of large-scale systems for important societal applications that are highly trustworthy despite being attractive targets.
- Within 10 years, quantitative information-systems risk management will be at least as good as quantitative financial risk management.
- For the dynamic, pervasive computing environments of the future, give endusers security they can understand and privacy they can control.
I would argue -- without much opposition from anyone knowledgeable, I daresay -- that we have not made any measurable progress against any of these goals, and have probably lost ground in at least two.
Why is that? Largely economics, and bad understanding of what good security involves. The economics aspect is that no one really cares about security -- enough. If security was important, companies would really invest in it. However, they don't want to part with all the legacy software and systems they have, so instead they keep stumbling forward and hope someone comes up with magic fairy dust they can buy to make everything better.
The government doesn't really care about good security, either. We've seen that the government is allegedly spending quite a bit on intercepting communications and implanting backdoors into systems, which is certainly not making our systems safer. And the DOD has a history of huge investment into information warfare resources, including buying and building weapons based on unpatched, undisclosed vulnerabilities. That's offense, not defense. Funding for education and advanced research is probably two orders of magnitude below what it really should be if there was a national intent to develop a secure infrastructure.
As far as understanding security goes, too many people still think that the ability to patch systems quickly is somehow the approach to security nirvana, and that constructing layers and layers of add-on security measures is the path to enlightenment. I no longer cringe when I hear someone who is adept at crafting system exploits referred to as a "cyber security expert," but so long as that is accepted as what the field is all about there is little hope of real progress. As J.R.R. Tolkien once wrote, "He that breaks a thing to find out what it is has left the path of wisdom." So long as people think that system penetration is a necessary skill for cyber security, we will stay on that wrong path.
And that is a great segue into the last of my three anniversary recognitions. Consider this quote (one of my favorite) from 1973 -- 40 years ago -- from a USAF report, Preliminary Notes on the Design of Secure Military Computer Systems, by a then-young Roger Schell:
…From a practical standpoint the security problem will remain as long as manufacturers remain committed to current system architectures, produced without a firm requirement for security. As long as there is support for ad hoc fixes and security packages for these inadequate designs and as long as the illusory results of penetration teams are accepted as demonstrations of a computer system security, proper security will not be a reality.
That was something we knew 40 years ago. To read it today is to realize that the field of practice hasn't progressed in any appreciable way in three decades, except we are now also stressing the wrong skills in developing the next generation of expertise.
Maybe I'll rethink that whole idea of going to give a talks on security and simply send them each a video loop of me banging my head against a wall.
PS -- happy 10th annual National Cyber Security Awareness Month -- a freebie fourth anniversary! But consider: if cyber security were really important, wouldn't we be aware of that every month? The fact that we need to promote awareness of it is proof it isn't taken seriously. Thanks, DHS!
Now, where can I find I good wall that doesn't already have dents from my forehead....?