Much has been made of how students used social media -- Instant Messaging, LiveJournal, blogs etc -- to update one another during the mass campus shootings at Virginia Tech last month. But despite all this networking, and the ubiquity of mobile phones, social media did as badly as official communication channels in issuing warnings in the two-hour gap between the first two shootings and the later rampage.

I was surprised by this. I thought I had spotted a trend for phones, especially mobiles, playing a growing role in emergencies. I had this neat thesis that growing access to communications technology would make it increasingly possible to issue alerts and so limit the impact of humanitarian emergencies.

In my first month at AlertNet I'd seen how in Goma, Democratic Republic of Congo, January 2002, a mass exodus prompted by the eruption of Mt Nyiragongo had theatened renewed tensions with neighbouring Rwanda before it was abruptly reversed by those left in the town phoned their friends to tell them the lava flows were slowing.

Later, when an earthquake struck Bam in December 2003, an Iranian Red Crescent volunteer saved a hundred lives when he woke people from their sleep by simply ringing them up before dying by his phone as a second devastating tremor killed him and demolished his town.

And last year, the Swedish government used SMS text messages to evacuate its citizens from Lebanon during the war with Israel.

At Virginia Tech, the warning from the University authorities was eventually sent via e-mail.

Sending an e-mail may have been the simplest thing for the authorities to do but, given that students don't all use Blackberrrys, it was not the most effective, according to Liz Gasster of the Cyber Industry Security Alliance quoted in Technology Daily.

In an insightful piece for Wired magazine, Ryan Singel surveying developments in the U.S. since 9/11 concluded bluntly that it's now possible to send 'geographically targeted messages' via mobiles. This being so, who should send the message? The two-hour delay by University officials suggests that any top-down system must be considered fallible. Even so, there's been a surge of interest in mobile alerting with many schools investigating commercial services.

So is it possible to combine geographically targeted messaging with social networking? I asked a couple of leading voices in social media for their views.

Howard Rheingold is the author of Social Mobs, a book articulating the huge potential a mobile-equipped, socially networked population could have. He remains optimistic despite Virginia Tech:

"I believe it's a matter of the lag in literacy regarding the use of smart mob technologies in the US. When the rest of the world was sending text messages by the billions, just 5 years ago, when Smart Mobs was published, you saw very few people using their phones for email text messages. Now, of course, every other person on the street is looking at their phone screen. But it takes a while for a culture to develop. I think you are correct that social network sites like Facebook (and its use of RSS) and instant access presence services like Twitter offer potential for emergency warning systems. Technically, it's not a problem to set it up. What to do about false alarms is a genuine concern. But also, it's a matter of general literacy about the use of these devices and techniques. I think the US population is ready for such media now in ways that they weren't ready before texting, social network services, RSS, and presence services started to become popular."

Howard mentioned Twitter -- a cross between SMS texting and blogging which can be done via your mobile. So I asked Biz Stone, one of Twitter's co-founders, to comment on whether it could be used as a social warning system.

"Messages can spread quickly in real time across multiple devices to groups of people on a system like Twitter. This seems to be key in any emergency. Additionally, if a person is part of a social network that operates in near real time, that person's friends may be of assistance to an emergency worker. Where was this person last? Are they allergic? Etc.

"With regard to false alarms, that's an interesting question. My hunch is that false alarms might be detected similarly to the way rumors are dispelled online. The community itself fact-checks and self organizes around the truth eventually. Not sure if that would be true or not or if it would take too long.

"On the subject of earthquakes, I'll say there's something there at least socially. I'm from Boston, MA originally but I live in Berkeley, CA now. Whenever there is an earthquake, my phone starts buzzing with Twitter updates from friends saying, "feeling an earthquake!" So within seconds of feeling an earthquake I know others are too and that they're safe. It makes me feel less anxious."

I was optimistic about mobiles and social networks coming together to create effective emergency alerting, particularly in the developing world, but now I'm not so sure. The issue of false alarms just seems so critical. Am I right to be less hopeful or was Virginia Tech not a good guide?

Reuters AlertNet is not responsible for the content of external websites.