[an error occurred while processing this directive]

learning.now: at the crossroads of Internet culture & education with host Andy Carvin

[an error occurred while processing this directive]
April182007

Virginia Tech: Yet Another Wake-Up Call for Better Emergency Preparedness

Eleven years ago, Readers Digest proclaimed Blacksburg, Virginia as “The Most Wired Town in America.” So why couldn’t this wealth of technology be used to save lives during the horrific massacre at Virginia Tech? Perhaps the wrong tools were in place.

“Communication is usually the weakest link in a school emergency.”

-School security expert Ken Trump, talking with CNN’s David Mattingly, April 17, 2007

If you had asked me 10 or 12 years ago to cite an example of what it means to bridge the digital divide, I’d bet that my unequivocal answer would have been Blacksburg, Virginia. The city of Blacksburg, in conjunction with Virginia Tech, was one of the first municipalities in the country to make a commitment to developing Internet infrastructure for the public good. They even had a phenomenal online community, Blacksburg Electronic Village, a website that did a fine job at representing the community’s diverse institutions.

So as I watched the events unfold on Monday as police realized the magnitude of the shooting that had taken place on the Virginia Tech campus, one of my first thoughts was, “Of all the campuses in the country, how could this happen here?” Of course, there’s no way any institution can be 100 percent prepared for such a random, terrifying event.

As a high-tech university, VT’s students are about as wired as any other school, with laptops everywhere and cell phones close to ubiquitous. Over the coming weeks, we’ll surely see more hand-wringing and outrage over the two-hour delay that took place between the time the first shootings occurred and the first mass emails informing students and staff that an incident had taken place. I don’t want to second-guess why the university waited, or why they chose not to lock down the campus immediately. Instead, I’d like to focus on a broader technological issue - why email probably wasn’t the best way to notify everyone on campus.

There’s no shortage of email access on Virginia Tech, and the university’s telecommunication office even supports Treo smartphones for those who choose to buy them. But the reality of the situation is that many, if not most students still rely on their computers for checking email, using their phones for talking and text messaging instead. Even if a mass email had gone out earlier, would it have reached everyone in time, particularly on a campus where the majority of students are commuters?

One telling insight into what happened that morning is an email that was circulated by a university librarian. The email was posted anonymously on MySpace and elsewhere soon after the shootings, but I managed to track down the author as Bruce Pencek, who gave me permission to identify him as the author. In his email, he wrote:

I watched from upstairs, marveling at how many people were still outside, and some in the building who hadn’t heard anything. Communication of the event was very much a case of who had a cell phone or wireless device before the bandwidth got constipated. (The campus web server soon crashed, taking down a big chunk of web-based email capacity with it.) I don’t know how instructors in class were informed of the events, since the emergency phone message system goes to office phones, not personal ones. With MP3 players taking the place of car radios, I’m not surprised that people in transit to o across campus would not have heard the newscasts.

I should note that the library’s PA system is used only to warn people when the fire alarms are being tested. In the event of an actual emergency, the semi-audible closing bell rings and workers from the circulation desk walk through each floor to give that happy news to whomever they happen to see.

Fortunately, the police soon started warning people through car-mounted loudspeakers every few minutes; I think this was a useful innovation since the town shooter incident at the beginning of the academic year. Even after the announcements, there were still people gawking outside.

Bruce hits upon several points worth noting. First of all, the telecommunications network was swamped: email and Web servers crashed, making it harder for those people who did have Internet access at the time to find out what was going on. Emergency messaging services that were in place for faculty went to their office phones rather than their mobile phones. And the PA system wasn’t used, though thankfully they were superceded by car-mounted loudspeakers.

Reading Bruce’s account got me interested in learning more about what systems the university had in place to alert people on campus. Last July, you may recall, I wrote about how some universities were bucking the trend seen in K-12 schools, actually making cell phones mandatory for cases of emergency. Why? Because almost every cell phone available today is able to send and receive SMS text messages. SMS infrastructure generally holds up better in times of crisis than email, and it automatically appears on your phone’s screen when you receive one. Here’s a quote from the article, discussing how Montclair State University made their decision:

After surveying students to identify specific features they’d like to see in a campus cell phone network, they partnered with Rave Wireless and Sprint/Nextel to develop the service, MSU Connect. They rolled out the program last fall, when they provided freshman living in residential halls with one semester of free service; students were then expected to pay $186 for subsequent semesters. This coming school year, all freshmen and on-campus sophomores will be required to participate in the program.


Along with providing students with local and long distance telephony, MSU Connect offers mobile classroom management tools, campus event listings, community updates and real-time public transportation information. The mobile phones are also equipped with a personal safety alert system called Rave Guardian. Utilizing the phone’s GPS capabilities, Rave Guardian allows university officials to keep track of the exact geographic location of every student at all times. It also lets students trigger a distress beacon that dispatches campus police to their location. “Police can track a student’s trip on a large screen in a bread-crumb sort of way until the student deactivates the service,” says Edward W. Chapel, who helped launch the program.

Ironically, a rumor began spreading yesterday that Rave Wireless had actually partnered with Virginia Tech, raising questions as to why the system hadn’t been used. I contacted the company, and a spokesman denied the report, saying that Rave had no relationship with VT. Perhaps part of the confusion came from a now-chilling newspaper article posted last September talking about how VT was considering text messaging broadcast applications, following another shooting incident nearby:

When Virginia Tech wanted to alert students to developments in a recent campus manhunt for an accused double murderer it relied on e-mail, the Web and messages sent to dorm phones. One method that was not available: sending text messages to cellphones. That could change.

University officials are considering following the lead of Penn State University and other schools that use text messaging to stay in contact with students for whom even e-mail is becoming passe.

The manhunt in Blacksburg for William Morva, which led to the evacuation of Squires Student Center and the first day of classes being canceled, is an extreme example of the type of situation that Tech officials would like to be able to notify students about on their cellphones.

“We will certainly be investigating other kinds of communications vehicles,” Tech spokesman Larry Hincker said, though he cautioned that a text-messaging system was still a “blue sky idea” that would take investigation to implement.

The article also mentions several companies offering these types of services, including Rave Wireless, which might be why some people thought that they were in business with Virginia Tech.

It’s not just universities that are trying to figure out how text messaging can play a role in emergency response. Some K-12 schools are beginning to make the investment as well, though the target audience for the emergency notifications are generally staff and parents, not students. One company, Roam Secure, offers an alert network that can be used to send mass broadcasts receivable via email, SMS, voicemail, desktop, even XM satellite radio. Ironically, some of their clients are Virginia counties and their school districts, including King George County and Fairfax County.

It’s not just local-level entities getting in on the act. Last year, the US Department of Homeland Security partnered with the Association of Public Television Stations to expand the current emergency alert system, formerly known as the emergency broadcast system. The system was first developed to broadcast emergency announcements over TV and radio - I’m sure you’ve heard “This is a test of the emergency broadcast system” on air, followed by some annoying beeps - and now they want to expand it to digital television, the Internet and, yes, text messaging. These messages would go out on a presidential order, though, so it’s designed for national emergencies more than local ones.

So where does that leave us? It leaves us with a lot of questions, some potential answers, and perhaps a few more questions after that. I have no doubt that universities that don’t have mandatory cell phone requirements or SMS alert systems are going to take the idea a lot more seriously now. But will K-12 schools? I’m skeptical. There is enormous opposition to allowing students to possess phones on campus, even though many parents argue they’re necessary for emergency communications. Some administrations will respond by saying the chances of a real emergency are slim, and students can’t be trusted to use them responsibly. Yes, an emergency on the scale of Virginia Tech are few and far between, but smaller-scale emergencies do happen from time to time. When more school shootings happen - and they will happen - it’s likely that more parents will be outraged by the fact their schools made it difficult or impossible to communicate with their children. And there will come a time when we will have no choice but to allow our students to carry communications devices. It may not be this month or even this year, but it will happen.

We’re simply not prepared, in any sense of the word. Perhaps we could at least begin a more serious conversation on the topic. That would be a start. -andy

Note: Speaking of text messages, you can now receive alerts from learning.now whenever new blog entries are posted. We’re using the messaging tool Twitter.com, which lets you subscribe to our feed and receive post announcements either online, via text messaging or instant messaging. Simply register a free account on Twitter then click the “add” button on the right side of the Twitter learning.now page. You can then set up your account preferences to receive announcements whichever way you choose.

Filed under : Mobile Devices, Policy, Safety

Responses

Hi, Andy,

I can’t help thinking that this is a case where old-fashioned tech would have worked best:

== Air-raid style sirens and PA
— Truck mounted PA systems
— Blocking entrances to campus to turn away people coming to school/work

Andy, Thanks for this very useful posting. I wanted to share some resources from the American Academy of Pediatrics to help with dealing with the aftermath of this tragedy http://www.aap.org/featured/resourcepage.htm

Andy,

Thanks for a wonderfully informative article. Keep up the great work.

Frank

My question is, what would you be telling all 26,000 people on your campus to do whenever someone is killed on campus? Send everyone home for the day? What if they get shot when everyone pours out into the quad at the same time? What about the students that stay on campus? Should they in their rooms? Where are they going to go to eat? You can’t order a pizza because the pizza guy might be the killer. When is it safe to come back? What about if someone is assaulted but not killed? Do you send out an instant warning about that?

Schools need to go through gaming exercises to anticipate different scenarios. Of course, no single message will ever be guaranteed to prevent further casualties. So the question is, are there ways to warn people to mitigate the number of casualties? That’s what security experts do for a living - teach institutions and enterprises about worst-case scenarios and how to deal with them. Meanwhile, as these best practices are developed, perhaps they should even be shared in an open source way so all schools can benefit from the knowledge. (Sorry, security experts, I’m not trying to put you out of business - I just want to get appropriate knowledge into the hands of the right people as equitably as possible.) -andy

This is a terrific article — thank you so much for posting it. Since we’re right in the middle of college decision-making, this has now become a factor that is being considered.

To Tom Hoffman above — I think it’s worth brainstorming different ways and levels of alerts. In this case, a message after the first shooting saying that there has been criminal activity on campus and to be aware might have sufficed, with a followup message saying to remain in their dorms at the moment that the second 911 call was received.

I can’t say with certainty that lives would have been saved. no one expects campus officials to be clairvoyant, but in my opinion, more information is better than less.

This is a thoughtful and informative article — thanks for posting it.

I think the first commenter has a good point. We ought to be careful not to put our faith in more complex and potentially more fragile technologies without first having solid low-tech plans in place.

Even then it seems that while text-messaging, email and other technologies are a useful supplement, they’re not reliable enough to depend on as a first or primary means of contact (as suggested by the story about the September Virginia Tech incident). Networks and servers go down, instant delivery isn’t guaranteed, software is buggy, batteries wear out, and people don’t generally have these devices on them at all times.

As I scrolled down the article, I anticipated your every comment - or at least the ones at forefront of my mind. I’m both a school board member and a PR person for a large telecom concern. I understand exactly what happens to the network the minute everyone tries to place a call or access a server simultaneously.

There is no question that SMS text messaging is the most efficient to reach the largest number of students. What needs to occur in advance, however, is a modicum of training/instruction for students, staff & faculty on what a particular message means and how to respond in event of such.

Yes, K-12 is reluctant to change and lose control of its charges. Technology is disruptive but it also is enabling. Better for educational institutions at all levels to get on board using its capabilities effectively for emergencies and disasters.

One concern, however, is legitimate - IF parents are allowed to receive communication from their children in time of crisis, the possibility that some would come to the school (perhaps armed), can further exacerbate a life threatening situation. Again, good communication and planning within a school community in advance can mitigate these concerns.

Should we extend an SMS type warning system to cities and neighborhoods?

Hi Steve,

The SMS warning systems I referred to in Fairfax and King George Counties here in the DC area are just that - county-wide systems which could be used municipally as well. They’re just working with their local school district to make it available for their needs as well. There have been some experiments to use SMS warning systems for coastal US cities because of hurricane threats. So yeah, as far as K-12 schools are concerned, I’d expect such systems to be locally based, but perhaps part of a county, state and national hierarchy depending on the emergency. Private universities would probably want to partner with their communities as well as making it campus-only when needed, while public universities are natural partners with state and local govts.

There’s already some precedent for national-level SMS broadcasts. After the tsunami two years ago, Sweden got their local mobile companies to ID every subscriber whose signals had recently originated in southeast Asia, then sent them a text message to report their whereabouts so they’d know if they were missing or not. If I remember correctly, Sweden previously used the same systems for blizzard alerts at the local level. -andy

I’ve been blogging about the failure communications at Virginia Tech. I did have the opportunity yesterday to speak to Rodger Desai, CEO of Rave Wireless, in order to understand what parameters a university administration must consider to be prepared for emergencies as well as to have a mobile communications system that the students and faculty could use for daily contact.

It requires that university contract with a carrier (I don’t think competition would lend itself to contracts with multiple carriers) to ensure adequate coverage capacity. Campuses seem to be moving away from the landline paradigm anyway since most of the students are carrying cell phones.

What happened at Virginia Tech serves as an important case study to show what options are needed to be considered for updating campus communications as well as student protection.

Andy,

Thank you for the informative and thought provoking article. I am the CEO of Roam Secure (thanks for the unsolicited plug). We started the emergency text alerting business in the aftermath of 9/11. If you lived in Washington or New York during 9/11, you were not able to call anyone on the phone for several days. However, text messaging on cell phones, email, pagers, Blackberry, instant messaging, etc. worked very well.

Subsequently, we launched the first public alerting system in the country with Arlington County, VA, the first major city (Washington, DC), the first region (National Capital Region) and the first statewide contract award from Virginia a few weeks ago. We have learned a lot over the years while deploying and managing these systems for routine incidents (traffic, crime, severe weather) and major emergencies (DC Sniper, Hurricanes Katrina and Rita, Blacksburg tragedy).

One thing is for certain, people want information and they want it quickly. The media is a great resource, but the information is often filtered and may be inaccruate. The government and emergency officials on the ground often have the most accurate and updated information during an incident. They need a way to communicate with their first responders, leaders, employees, residents, neighbors, the media and more. Text alerting has proven to work time and time again and it can deliver targeted, real-time information.

We also believe the more ways you can reach people, the better. Its not just about an SMS, or an email, but layering multiple means of communication. For example, Arlington County, VA has a text alerting system (Arlingtonalert.com), a voice dialing system (Reverse911), an AM radio station, XM Satellite Radio, amplified voice notification, the county website, a cable TV channel and they continue to strive for more. They learned the lessons of interoperability and public information from their experiences with the Pentagon on 9/11 and continually look for ways to improve.

All of this is an evolving field. We learn something new every day. The current systems and technology in place today across the country will evolve as more and more schools, universities, governments and businesses deploy and use them and as the vendors listen and make continual improvements. It is most unfortunate it took a tragic event such as the Virginia Tech shootings to bring this issue to the forefront. Last week was a relatively quiet week, this week the phones are ringing off the hook across the country from Universities, governments, the media and more. We appreciate the current interest and can only hope people don’t forget the lessons learned and they take proactive measures to continually improve their communication capabilities.

Our thoughts and prayers go out to the students and families of Virgina Tech.

Cheers, David.

Sound The Siren

When life gets complicated, technologically speaking, it is good to go back to the traditions. Sound the alarm. The human touch (a voice over the school alarm speaker) is far superior than thumb or finger tapping on a computer. If a voice came across the speaker air waves, it would have been far more effective at notifying people than assumed access to a computer would be.

I saw a text service for schools on CNN last night called “e2Campus” and checked out their web site. It looks like they send messages to students cell phone phones, emails, pages, PDA’s, and other places where students can be.

Also- in reference to Bill Koslosky’s comment, unlike Rave, e2Campus works will all cellular phone carriers so students can use whatever phone and phone service they came to school with. You can confirm this on their web site www.e2campus.com.

You are correct Bill- I do not think a university should be required to contract with a specific carrier in order to send messages students.

Since I’m quoted (typo and all), I want to clarify. My library’s PA system was not used. (This is a discussion point within the library’s assessment of what we could do better.) I’ve no idea if there is a PA system in all Virginia Tech buildings, but I rather doubt it.

Mr Grecco is on target. The mobile loudspeakers (in vans, not patrol cars, I hear) were very effective: they gave a consistent, authoritative message that was so loud and intrusive you couldn’t not hear them.

Even so, some people continued to around outdoors, and many clustered at windows and doorways despite explicit instructions not to. As with the notion that the campus perimeter can be sealed, there simply aren’t enough officers both to compel people to take protective action and also deal with the manifest threat.

In an emergency, I can’t trust a communications system that depends on the recipients’ being willing and able to get the message in time.

It is understandable but unfortunate that discussion of the crisis here has centered on notifying students (who are a pretty differentiated lot beneath our generalizations). Emergency communications also have to reach faculty and staff, whose circumstances make keep them out of a high-tech loop (even if we set aside volition and generational patterns of media preference).

Personal example: In order for me to get my various tasks — library instruction, reference, outreach, service, my own research agenda — completed I communicate asynchronously. A voicemail signal on my office phone or my mobile does not connote urgency to me, merely one more damned thing demanding my attention. (Of course, in the public spaces of a library and in classrooms, protocol and courtesy require silencing it or switching off entirely.) Ditto emails, which I may disable for hours on end in order to avoid distraction. If I’m teaching a class, there’s no way I’d answer a call on my mobile, and I’ll continue to discourage students’ answering their phones or checking email.

Broadcast media add middlemen to the problem of recipient volition.

Various, simultaneous media, from the lowest tech to the highest, have to be used. My campus has gotten better in my six years here.

Many thanks for the expressions of sympathy and support for Virginia Tech.

Thanks for the info Andy.

I checked the Rave website and their Guardian application sis pretty interesting. Prevention is better than cure for sure.

Thx, Patrick.

VA Tech officials are being criticized for not ordering a lockdown after the first shooting. What does “lockdown” mean if the classrooms had no locking device?

Suggested Ideas for increased campus security:

Each classroom should have one escape route either directly to the outdoors or to a different part of the building.

Each door should be hardened and should have a deadbolt on the interior.

Each building should have a central office facility with access to video feeds and audio communication from multiple locations in the building. This central facility would coordinate responses and give information on the shooters’ location to police, faculty and staff.

Having been raised in Europe, until now I have always felt that the answers to gun violence in the U.S. are early intervention and tougher gun-control laws. However, given the reality of the Second Amendment, we need to seriously consider the additional solution of recruiting University faculty and staff members to carry handguns. Candidates would be subjected to a rigorous screening process, would agree to extensive training, and would participate in regular drills simulating these shooting events. In addition to their handguns they could store other weapons and equipment in their offices under suitably secure conditions. They could quickly don easily-identifiable Kevlar vests that would simultaneously protect them and prevent accidental shooting by police.

I have been recommending a two prong approach which would include the deployment of Blackberry phones to the first team of responders - Staff, Campus Security and Police - followed by a secondary which would then include email and SMS. Blackberry would also provide a carrier agnostic approach. Using the METAmessage EC product, there is the ability to instantly ‘Blast’ out a PIN message [native to each Blackberry and automatically updated with EC] to set distribution lists - which will be received instantly - and without a 120 - 160 character limitation either. The sender will have notification via a read receipt to know that the message was received and read, by whom and when. The entire event can be archived for the record. The product also features the ability to create custom alerts - overriding the user’s device and ensuring that the device notifies of an urgent message - the policy settings are very flexible, so that something as simple as a rule that has any email or PIN message with the word ‘Urgent’ in the subject, for example, can set off a special ring, vibration and pop-up text box - or all three. You can even override the volume, and set it as loud as possible. The architecture is superior to the SMS in that it doesn’t rely on a webserver, an email server or gateway - not even the BES is required. Another feature is a ‘Panic’ button, which can have a preset distribution list, and with an 8800, 8100, or the new 8300 model, can also include GPS coordinates. One more very appealing item is the document push. You can set it up so that critical Business Continuity documentation [word files, etc.] are routinely pushed out - directly to the handset so that all users are prepared in the event of an emergency with the tools they need - up-to-date and at their fingertips. Additionally, since it’s a Blackberry, you can create your urgent notification and Blast it from just about anywhere. SMS is planned to be added to email and PIN at some point in the future, however, no one should forget that it was designed as a two-way paging service that makes use of the control plane. Used as the second prong for redundancy, as the first responder teams receive first notification and act accordingly.

SMS for redundancy - not first line defense. I was selling this before the incident occurred at Virginia Tech and will continue to sell it. While everyone sincerely hopes there is never a repeat of what transpired there - I would caution any school trying to take some Web hosted SMS service and call it an ‘emergency communications’ solution - because it is not. There are simply too many points of vulnerability. And at VT, as per what I remember reading, the website was unavailable, students were having trouble sending text messages, and the email and webservers went down.

I have a question.

Can 60,000 SMS messages be sent out in 15 minutes? Can a university campus covered by 3 or 4 carriers actually handel that number of sms messages with out crashing?

[an error occurred while processing this directive]