As long as you can fool some of the users some of the time, we're going to have problems like the latest Android malware. And since you're always going to be able to fool some of the people -- that's just a part of human nature -- we need to stop grousing about stupid users and instead find more effective ways to stop more kinds of malware.
Yes, we can do that.
What got me thinking about this was a story that caught my eye about a self-propagating worm that infects Android devices. The story tells of malware that spreads via a URL link in an SMS text. According to the article, "Once installed on a device, the malware, which was dubbed Selfmite, sends a text message to 20 contacts from the device owner's address book." Further, "The text message sent by Selfmite contains the contact's name and reads: 'Dear [NAME], Look the Self-time,' followed by a goo.gl shortened URL." That URL, in turn, contains the Android malware itself. If the victim clicks on the URL and allows the malware to install, the cycle will continue.
What caught my eye was that this all sounds painfully familiar. Back in the late 1990s and early 2000s, there were many PC viruses that behaved in a very similar manner, only they propagated via email rather than SMS. This is the same old story, with some updates.
Now, I'm not saying we should be freaked out about this so-called Selfmite. It's unlikely to spread far (though I'll bet it will hit a fair share of victims).
But I do have a message for my security colleagues: Let's not be contemptuous of those users who do fall victim to Selfmite and its ilk. I know, I know; it's hard not to roll your eyes and say something such as, "This sort of thing will only spread if users are dumb enough to click on the link and then allow the software to install on their device." That's perfectly true, but it misses the point somewhat.
Let's give users some slack. After all, the SMS messages look as if they are from friends, or at least acquaintances. We're all more likely to trust things that come from people we know. Sure, this particular come-on has some glaringly obvious red flags. (Who uses "Dear" when addressing someone in a text? What kind of sense does "Look the Self-time" make?) But clicks happen. Some people will just automatically click on something sent by certain people -- their boss, maybe, or their spouse. You cannot get the number of people who will not click down to zero. That much is guaranteed.
What's needed is more secure software. I'm not talking about security software, although I'm sure many a mobile security software vendor will use this and similar malware incidents to try to sell us their wares. I'm talking about software that is more secure to begin with. Software with security built in.
Think about it. Why on earth should the SMS program on a mobile device be allowed to install software? For that matter, why should unvetted software be allowed to install at all? We should be asking ourselves these sorts of basic questions at every stage of development. Think about the things that we let software do because it's always been that way, and then ask yourself, "Is it really a good idea? Wouldn't the software be more secure if we didn't do that?"
We simply have to do a better job of building software that doesn't betray users when they make unwise choices.
In fact, we have to design and build software that assumes users will make unwise choices. Because they will. They need us to make some wise choices before they ever get their hands on our software.
With more than 20 years in the information security field, Kenneth van Wyk has worked at Carnegie Mellon University's CERT/CC, the U.S. Deptartment of Defense, Para-Protect and others. He has published two books on information security and is working on a third. He is the president and principal consultant at KRvW Associates LLC in Alexandria, Va.