Perimeter

7/7/2011
02:30 PM
50%
50%

Simple Isn't Simple

It's time to admit security is hard, and to stop blaming the victims for being human

We hear it time and again in various articles and reports: "The breach could have been prevented by following common practices."

Sounds simple, doesn't it? All we have to do is focus on the basics, and the bad guys will move on to a softer target. All we need to do is train our developers to avoid SQL injection and cross-site scripting, and the world will be a better place. We just need to upgrade to the latest version of XYZ, keep our systems patched, and our children will sleep better at night.

I mean really, it's only the idiots and fools who are getting hacked -- the ones who reused a password, didn't scan their Web site code, or let their users click on a bad link in an email.

It isn't like we lack for checklists, standards, or frameworks. From ISO 27001 to PCI to BSIMM, we have no shortage of guides to security nirvana. Just follow these, and you should be able to stop all but the most determined attackers. Simple, isn't it?

Except simple doesn't scale, for any given value of the scale. Are you a small shop with only 10 employees and systems to manage? Woe on you if you stupidly spend more on the GUI for your product than the security within it. A large enterprise? How DARE you miss a single targeted phishing email and let a user open the attachment.

We security pundits, researchers, and vendors tend to forget how hard real-world operational IT is. If you're small, you can control more, but you have fewer resources at your disposal. If you're large, you still struggle for resources, but now at an enormous scale. It's a no-win situation because no one can be perfect all the time. Or even some of the time.

Take Web applications. In small companies, you can't really afford third-party secure development tools, and at best might be able to pay for a scan or two. Yes, there are freeware tools, but nothing is free when you have to pay someone to learn it. Train your developers? Yes, you should invest there, but developers are humans and prone to error, and even minimal training hurts the development cycles you need to sell the product and feed your families.

Are you larger? You might be able to afford every tool in the book, but now you have more applications and more developers to manage. Even something as simple as weeding out SQL injection is nearly impossible for a multinational with hoards of developers, contractors, and outsources. Nothing is ever simple at scale.

I'm not claiming security is impossible. I'm not saying we should give up. And, believe me, I'm not out defending mediocrity. But I am saying we need to recognize that it's hard at all levels. That even the easy parts are nearly universally difficult in practice.

This isn't one of those articles with answers. Sure, I can talk all day about how users need to operationalize security more, and vendors need to simplify, consolidate, and improve functionality. But in the end those problems are every bit as hard as everything else I'm talking about and won't be solved anytime soon. Especially since the economics aren't overly favorable.

But we can recognize that we rely on complex solutions to difficult problems, and blaming every victim for getting hacked isn't productive. Especially since you're next.

Security is hard. It's even harder at scale. And we need to stop pretending that even the most basic of practices are always simple, and start focusing on how to make them more effective and easier to manage in a messy, ugly, real world.

Rich Mogull is is founder of Securosis LLC and a former security industry analyst for Gartner Inc. Rich has twenty years experience in information security, physical security, and risk management. He specializes in cloud security, data security, application security, emerging security technologies, and security management. He is also the principle course designer of the ... View Full Bio

Comment  | 
Print  | 
More Insights
Comments
Newest First  |  Oldest First  |  Threaded View
Four Faces of Fraud: Identity, 'Fake' Identity, Ransomware & Digital
David Shefter, Chief Technology Officer at Ziften Technologies,  6/14/2018
Meet 'Bro': The Best-Kept Secret of Network Security
Greg Bell, CEO, Corelight,  6/14/2018
Containerized Apps: An 8-Point Security Checklist
Jai Vijayan, Freelance writer,  6/14/2018
Register for Dark Reading Newsletters
White Papers
Video
Cartoon
Current Issue
Flash Poll
Twitter Feed
Dark Reading - Bug Report
Bug Report
Enterprise Vulnerabilities
From DHS/US-CERT's National Vulnerability Database
CVE-2018-9036
PUBLISHED: 2018-06-20
CheckSec Canopy 3.x before 3.0.7 has stored XSS via the Login Page Disclaimer, allowing attacks by low-privileged users against higher-privileged users.
CVE-2018-12327
PUBLISHED: 2018-06-20
Stack-based buffer overflow in ntpq and ntpdc of NTP version 4.2.8p11 allows an attacker to achieve code execution or escalate to higher privileges via a long string as the argument for an IPv4 or IPv6 command-line parameter. NOTE: It is unclear whether there are any common situations in which ntpq ...
CVE-2018-12558
PUBLISHED: 2018-06-20
The parse() method in the Email::Address module through 1.909 for Perl is vulnerable to Algorithmic complexity on specially prepared input, leading to Denial of Service. Prepared special input that caused this problem contained 30 form-field characters ("\f").
CVE-2018-6563
PUBLISHED: 2018-06-20
Multiple cross-site request forgery (CSRF) vulnerabilities in totemomail Encryption Gateway before 6.0.0_Build_371 allow remote attackers to hijack the authentication of users for requests that (1) change user settings, (2) send emails, or (3) change contact information by leveraging lack of an anti...
CVE-2018-1120
PUBLISHED: 2018-06-20
A flaw was found affecting the Linux kernel before version 4.17. By mmap()ing a FUSE-backed file onto a process's memory containing command line arguments (or environment strings), an attacker can cause utilities from psutils or procps (such as ps, w) or any other program which makes a read() call t...