By Matt Flick | November 6, 2008
I have had many discussions this year regarding the future of the application security industry and even more about its current state. It’s interesting how people of such varying backgrounds will have similarly varying views; this short article is designed to capture those views and hopefully drive some productive discussion as a result.
Where are we now? Should be a simple question, right? Let me summarize in three main categories:
Maturity – Purists aside, “Application Security” today means the security of web applications and related things, such as web services. Application Security started to grow up as older information security areas–like network security and vulnerability management–matured quite well. I am quite sure it seemed at the time a great idea to port the mature security ideas and technologies over to the application security world. However, as usual we have learned the hard way that the proverbial square peg does not fit into a round hole. Result? Application Security is in the early stages of being adopted into organization’s information security programs. **Golf Clap**
Technology – I remember from childhood riding long distances (>1000 miles) with my family of five in a compact car to visit the in-laws. Of the many shortcuts we tried using our trusted 10-year old map, I remember one that ended with a missing bridge over the Mississippi River. The recent development of app sec tools eerily reminds me of this trip: choosing the wrong path and driving insistently down that path until you end up with very wet socks. Luckily for my family the car brakes worked…unfortunately I don’t see anyone trying to even find the app sec technology brakes. So who is driving us down the wrong path, and why is it the wrong path? The typical response would be the vendors, but I disagree. Runtime scanners, source code scanners, application layer firewalls–they all perform as designed (in most circumstances). The problem lies in how these tools are sold and used as a method to secure the vulnerable applications. To a slightly lesser degree is the nearly insurmountable problem of these security tools keeping pace with the fast growing arena of application technology and subsequent vulnerabilities. Both of these problems illustrate why organizations needs to focus more on the root cause of the vulnerabilities rather than on the detection and prevention of attack vectors. The unfortunate fact is that application security technologies can not—and may never be able to—keep pace with vulnerability and attack research. This is the wrong path. This is why we need to hit the brakes and find a better route.
Approach – Is that light? Are we in a tunnel? Yes and no. The application security world has witnessed several of its citizens make wonderful presentations on why we need to…drumroll…incorporate security into the SDLC! Or at least it was witnessed by the five of us not watching the more exciting and entertaining presentations on the latest and greatest XSS and CSRF attacks. Whereas this statement of incorporating security into the world of development is theoretically valid, when put into practice the wheels tend to fall off (or in worse cases, explode). After assessing so many different environments and working with clients to build practical and effective application security programs, I’ve all but killed the nerdy theorist inside of me. The old “one size fits all” adage really starts to become annoying! I am not arguing the approach nor will I espouse a new theory of my own–remember I said it is valid. Instead I will just note that we are, as an industry, still at the theoretical stage. At least I can cross off “job security” from my list of worries.
Where are we going? If the first half of this post was not depressing enough for you, go back and read it again more thoroughly. Then read on. Here is a sampling of quotes I collected this past year on the question of where the application security industry will be in 3-5 years:
“More of the same. New technology maybe, same attacks against both the new and old technology.” –security consultant
“I don’t think much will change. With Flash and AJAX growing there will be new opportunities there, but not much else will be different.” –application security manager
“SaaS is the future of app sec. The tools are quickly getting smart enough to attack application business logic and will revolutionize the industry.” –vendor marketing guru (shocker)
“More companies will realize application security is something they have to deal with.” –security guy
Job security…oh right, I already crossed that off. Is this the best we can do? Inch our way further down the wrong path? Marginally better tools, more of the same attacks against current technology, and bigger budgets for the same insufficient solutions? I cannot be the only one thinking that this is a problem. Sadly, I have to somewhat agree with the comments above. In dealing with people from all sides of the aisle, it does appear that the application security community is settled comfortably with the status quo. With the massive increase in application security spending over the last few years, can anyone blame them? That being said, maybe I am the only one plotting and scheming of a better way…and maybe that’s not such a bad thing.