Update: This project is now called UN Global Pulse.
I’m in Bellagio on Lake Como this week for a Blue Sky Thinkers Workshop on the UN’s new Global Pulse Initiative. When I first blogged about GIVAS as it was called back in July 2009, the Pulse Team in the UN Secretary General’s Office actually commented on my blog post. The fact that the UNSG’s Office was taking the time to read blogs and comment on them was the first sign that something about this project was very different from my previous experience with the UN.
We’re under Chatham House Rules here so I’ll just stick to my own thoughts on what I think Global Pulse should be. First, I don’t think Global Pulse should be for the UN or nation states. The global alert system should directly empower vulnerable communities to prevent or mitigate the impact of crises on their own livelihoods. In other words, Global Pulse should be a self-help system for vulnerable communities. The development and maintenance of this system should be the responsibility of the UN and governments.
So here’s an idea (still under development): why not use the QuestionBox technology and approach to create “call in” centers for information on tactics for resilience.
Question Box helps people find answers to everyday questions like health, agriculture, business, education and entertainment. It provides easy access to information in hard-to-reach areas and breaks through technology, language and literacy barriers. We do this through:
- Live telephone hotlines connected to live operators
- SMS (Text Messaging)
- Mobile and solar technologies that operate off the grid
- Open Question – a simple software to start your own Question Box project
The group behind Question Box also help several organizations start their own Question Box-inspired services. So lets turn Global Pulse into a Global Resilience Information Service for vulnerable communities. There are four key reasons I find this approach compelling. First, this approach provides a demand-driven direct service to vulnerable communities as opposed to just “watching” them. Databases of resilience tactics can be (continually) developed by either local communities themselves or government sponsored projects. This information can then be shared across towns and regions. Think of this as a “Resilience Wiki”.
The second reason I want to continue exploring this system is because the queries made using a Resilience Question Box approach are in and of themselves important indicators. Think of Google’s Flu trends project. The team “found a close relationship between the number of people who search for flu-related topics and the number of people who actually have flu symptoms.” In other words, the queries made to Resilience Question Box could serve as proxy indicators for local vulnerability. This data would then be analyzed for trends and policy making at the UN.
The third reason this approach appeals to me is because it serves the interest of vulnerability communities first and foremost. The rhetoric behind global alert systems is that they are for vulnerable communities but the reality is that these communities rarely know that such systems actually exist. The first indicator of success for Global Pulse will therefore be whether vulnerable communities are aware of Pulse. A second indicator will be whether they actually use the system.
The fourth and final reason I’m keen about a Question Box approach is because of the focus on information for vulnerable communities. In my opinion, most crises are ultimately crises in information, which is another reason why information is power. Imagine if Global Pulse could work with Member States to set up hundreds of thousands of Resilience Boxes. Think of the One Laptop per Child (OLPC). Global Pulse could provide a Resilience Box per Vulnerable Community, enabling the latter to make more informed decisions to increase their own resilience in near real-time.
Now how does one fund all this? I got a possible answer over lunch while talking to one of the participants (whose identity I can’t reveal because of Chatham House Rules). How about a Kiva for Resilience Boxes? Instead of donating money to one (or more) specific project, you could donate money for 50 QuestionBox answers to a specific community in Bangladesh. I personally find that very compelling, i.e., knowing that my money provided 50 answers for vulnerable communities. In return for my donation, perhaps I could also get a copy of the 50 questions/answers and learn something new in the process.
Ok, I’ll spare you the long post and just give you the highlights.
This is a really good basic idea but perhaps could be improved with a few re-alignments in the fundamentals of the system.
1. Why limit yourself to only calling a centralized location for answers? Perhaps there is a really good reason for this- but by the same token I see some really good reasons to create a networked approach. Such a mainframe style system must have the same questions asked -> translated -> researched/answered -> re-translated many times, creating an amazing amount of redundancy, which must make this whole thing very expensive to setup and maintain compared to the alternative:
What if, instead of only having a single centralized or regional location to call, you could call any of the villages in your same linguistic region?
The mainframe style system also suffers from the ‘single point of failure’ scenario- what happens if there’s a fire, flood, or power loss to the central office? What if the line is cut for whatever reason, or there’s no more funding to pay people to translate… does the whole system need to collapse? The mainframe model is vulnerable to this, but a networked model has some real resiliency.
2. If you’re looking for a funding model I might consider a two-pronged approach; separate out the capital investments from the operational tasks/expenses.
No one wants to buy transaction/translation credits just like no one wants to buy carbon offset certificates… because the “limbic loop” just isn’t tight enough to see the satisfaction of what you’re paying into- people aren’t generally altruistic in a vacuum, but they’re slaphappy when it comes to well balanced limbic feedback. The brilliance of microfinance organizations like Kiva (or even better Okio Credit) is that they close the limbic loop just enough to create a good satisfying feeling in the consumer.
Everyday people like me are far more likely to put money/time into capital investments/operations things if you separate them out. And I’m more likely to invest in a capital investment like a school or answer box (or both together) if I can see it, track it… and best of all, interact with it!
Let’s just lay out one possible scenario real quick. What if there was a real question asked through this translation pipeline that I could answer and actually have it translated back to Hindi, Creole or whatever? I would be infinitely more likely volunteer my time to research the question and give it a proper well thought out answer if I knew where it was going. Similarly if I knew English and Hindi I would be a lot more willing to volunteer my time to translate if I could see where that information was going. – Furthermore (as my wife who does fundraising and development will attest) I am far more likely to also invest real cash in the capital expenses and maintenance of a system that I was an integral part of, and had a real connection to people on the other end.
3. Why is this project re-inventing the wheel here, with a one-size-fits-all solution? What’s the real problem behind doling out satellite/cell phones? Is it carrier expenses? Recharging stations? Robustness? Theft? I’m actually very curious to know the driving reasons… but whatever they are I suspect that if you look at the full lifecycle it’s probably cheaper to create a robust backend (not unlike the brilliance of Ushahidi & Swift) that can interact/interface with the regionally available technology to create regionally relevant scalable mini-information networks.
4. Lastly- partnerships, partnerships, partnerships. Why not partner with people like Greg Mortenson and put one of these in every school to further leverage the value of the answers? Why not partner with Nokia,
5. Ok, this is the real last one. What’s the real value being created by these questions being asked? (Like you touched on with the google search terms and flu.) Perhaps the value is meta-data, micro-trend spotting, global trend spotting, crisis interruption, or breaking news? Without even getting into the ethics – this information will have real value to someone, and someone will be willing and able to pay into the system to know this information at some level.
As always, many many thanks for your feedback, Sean!
A few quick thoughts in response:
* I don’t see the need for translation at all (never brought up the issue in my blog post). The “call/sms” centers would all be local, and there would absolutely be more than one. So yes, the call/sms centers would be in the same linguistic region, absolutely.
* Well put re capital investments vs op tasks/expenses.
* Not sure I follow your comment re re-inventing the wheel. It’s not an either/or at all. Keep dolling out cell phones. QuestionBox is simply one way for rural/marginalized communities to get access to information. Think ecosystem deployment.
* Partnerships, absolutely! This idea of QuestionBox/FluTrends/Kiva just sprung up a few days ago. I’m hoping the GIVAS team will be interested in exploring this further in the coming weeks.
* On value creation, the real value created is two-fold: (1) providing vulnerable communities with a source of information to enable them to make the most informed decisions possible on their immediate security and livelihoods, thus increasing their resilience in times of crises; (2) have the questions themselves serve as meta data (FluTrend style) that can be analyzed using geospatial statistics and visualized in compelling ways for policy makers, donors etc. While most early warning systems I’ve seen start out focused on the former and latter equally, they almost always end up ditching the information service for vulnerable communities and go instead with the fancy sophisticated technical platforms for policy makers. Lets hope this won’t be a repeat.
*Re translation… oh, yes I see now… I sometimes leave out entire trains of thought and absentmindedly assume that people were either present (or perhaps are omnipresent, at a conversation that sometimes only happens in my head)… I catch it most of the time when I go back and re-read what I’ve written, but sometimes things get through the filter. I think being jet-lagged and without coffee increases those odds significantly. It’s a result of my Asperger’s, a source of great strength and sometimes confusion… my apologies, now let me see if I can backfill.
So regarding the translation step, I should “set the table” as my wife likes to remind me (a term she uses to remind me to give people I’m talking with the necessary background to follow me down a path and not get lost/bored.)
Ok, so without revealing too much unrelated to this post, we’ve also been doing a lot of deep-thinking on HITs and I guess I was thinking ahead somewhat strategically on how one would socialize the AnswerBox, thereby creating that strong limbic connection to the “prosuming” everyday person like myself. In essence I was attempting to think strategically about a couple things at once: scalability, funding, robustness and flex-capacity (i.e. the ability to respond in real-time to the massive information spike in a crisis, a la Haiti.) But perhaps that’s beyond the desire and scope of the AnswerBox team’s goals- I don’t know I’m just waxing here.
In that context the translation step would become necessary to really connect people like myself (with discretionary time, money, and knowledge) to the target region or event. Think crowd/turk-sourcing, combined with ‘adopt a highway’ on steroids, combined with fundraising, prosuming, and the robustness of de-centralized network architecture.
*You know I’m such a slow typist and non-linear thinker that I that this conversation would probably take up a lot less time and be a lot more productive if it was in fact a face-to-face conversation… and probably make more sense because you could ask real-time what the heck I was going on about.
Maybe we could connect in San Francisco, or if you wanted to stop in Seattle ahead of schedule? I’ve never actually been to San Fran but am told that I would love it. Email me.
Hiya Sean, no worries at all! One classic example of Mechanical Turk Services is the application to translation, so perfectly in order to bring this up, not least since–as you noted–the translation step would become necessary to really connect people with discretionary time, money, knowledge etc. Thanks!
Pingback: The Best of iRevolution: Four Years of Blogging | iRevolution
Pingback: Question Box: Catching Up with an Old Grantee | Indigo Trust