Security Crowdsourcing is Here

In an October post I speculated about "crowdsourcing" security to spread mundane, manpower-intensive tasks away from soldiers and onto willing members of the American public. I gave one speculative, half-baked example: what if we put thousands of video cameras along the border between Pakistan and Afghanistan, plugged them into the Internet, and let anyone in America watch the feeds? The Army could create a website allowing Americans to self-organize into surveillance teams; if these teams saw anything suspicious (i.e. insurgents trafficking arms over the border), they could report it immediately to a US military nerve center that would analyze the event and determine an appropriate response. Mom and pop could do their part to keep Johnny safe in Afghanistan.

I gave my example only to demonstrate possibilities. I never actually expected to see anything like this, at least for a long time. But it turns out the State of Texas is doing exactly what I suggested on the US-Mexico border. Check out www.texasborderwatch.com. Jeff Howe reports briefly on the subject at his Crowdsourcing blog.


According to Howe, the first incarnation of the Virtual Border Watch program failed because of lack of interest. Texas is trying again at a cost of $2 million. Howe is skeptical the program will work. I agree. While the right ideas are there, the project doesn't seem to have the right business model behind it. Just because an idea is "wiki-powered" doesn't mean it will be successful or even remotely interesting. The Wikinomics book and blog frequently discuss the need for business leaders to find creative, innovative ways to squeeze profit out of wikinomic ideas.

Could a crowd-sourced border patrol work? I still think so, but in my original post, I argued the idea needs to be incentivized. I say that with some hesitation, because I'm nervous about the idea of living in a world where someone sitting at home in his underwear gets paid to bag illegal Mexican immigrants or Taliban insurgents. But we live in a crazy world, and that may be a component of future security. Incentives don't necessarily have to be monetary, either. Many web forums award various "flair" like rank pips and medals for active involvement. I've never understood the appeal myself, but this fame/recognition might motivate some people.

A crowdsourced security effort also needs to be communal. Sitting at a computer staring at a camera feed is boring as hell, but people might find purpose, enjoyment, and esprit de corps in being part of an online community dedicated to the task. When I first went to the Air Force Academy my mom found an e-mail group for parents of cadets. I couldn't believe how much time and energy these parents dedicated to talking about the Academy. Right before I started Basic Training my mom printed out 20 pages of e-mails from the group on shoe shining technique. If you can find a crowd willing to do that, you can probably find a crowd willing to help enforce border security. But you have to make a crowd, not just a collection of isolated individuals. A website for virtual border patrol should have communications tools built in, like message forums, user profiles, instant chat, etc.

Technology can also help users target their efforts on vital areas. Because these video feeds are digitized, it should be a simple task to identify motion and flag cameras that are particularly active. Significant events could be recorded in logs associated with each camera (wiki-style of course, maintained by users). Local officials could also intervene to suggest hot areas in need of surveillance. What you want to avoid is the presentation Texas Virtual Border Watch gives you: a long, indiscriminate list of cameras with no effort to steer your surveillance to needed areas.

Finally, the system should be easy. Virtual Border Watch requires users to create an account before they can view a camera. I suspect that leads a large number of users (like me) to simply leave the site. Some form of user identification is probably necessary, but maybe these sites could have limited guest logins to draw in users.

Will this idea ever work? I have no idea. I certainly think it's worth a try, but it has to be implemented smartly. The Texas Virtual Border Watch project has a long way to go. For $2M I expect a lot more. Maybe Texas should crowdsource its R&D. It could save its $2M and give a few $1,000 grants to 19 year-old World of Warcraft players who know how to build websites. Texas would probably get a lot more for its money.