Questions for Question Box
I recently started volunteering for a remarkable nonprofit called Question Box, after following (and talking incessantly about) them for nearly a year. Their aim is to bring the web to poor, often illiterate people in remote areas. On the weekend I caught up for a chat with the project’s founder, Rose Shuman.
Andy: To kick things off, could you please tell us a little about your background?
Rose: Certainly. I am founder of Open Mind and the Question Box Project. I am a consultant specializing in the intersection of business and international development. On the side, I consult for a social enterprise, Adlens, in Oxford, UK. Previously, I managed corporate relations for one of the largest nonprofits in the U.S., Direct Relief International. I hold two BAs and a Masters from Brown University, in Rhode Island.
Andy: So, Question Box – what’s it all about?
Rose: Question Box brings the internet to people who will likely never get online in their lives, or at least in the near future. It’s a local language telephone hotline that connects people to Operators. They connect to the Operator through Question Boxes, which were designed especially for the project. Question Boxes are metal boxes with a hacked mobile phone inside that calls only the Operator. The users can ask just about any question imaginable, and the Operators look the answer up for them, usually translating the question into English first. It allows one internet-connected person to service the needs of thousands, without needing to bring expensive infrastructure into unserved places.
Andy: Are the boxes always part of the loop? Can people call the Operators directly?
Rose: That’s the next step in the project. We want to make the service available from several portals – physical Question Boxes, users’ cell phones, plus we are building in SMS and web query functions as well. We are about to pilot a variation on the concept in Uganda, in which community workers with cell phones will introduce the service to villagers.
Andy: What inspired the project?
Rose: A very long day at a desk in a Mountain View, California hotel room. I was meeting someone at Google.org the next day for a job interview, and I wanted to say something original. So, drawing up my years of experience in development, I started brainstorming in a notebook, which I still have. Eventually, I came to Question Box, fully formed. I didn’t get the job but the concept still tugged at my sleeve.
Andy: So where are the boxes currently running?
Rose: Right now, we have one in a village called Loni outside of Pune. Until recently, we had a second one in a Pune slum, but it was experiencing a technical difficulty. It’s going back into the trenches in a couple of weeks. There are also plans to pilot one in Orissa in partnership with an ICT organization, eKutir.
Andy: What kind of questions to end-users typically ask? Is it a challenge to convey the extent of the resources available on the web?
Rose: Currently, users are asking a lot about train times and produce prices. In the previous pilot, there was more cricket and university scores. We haven’t tried to explain the web or what’s on it yet to people, still observing what they come up with. However, the Operator is going to start prompting people with suggestions, to see if we plant some seeds.
Andy: How are Operators selected, and what kind of training are they given?
Rose: Operators need to be internet-savvy, fluently speak the local language, and read and write English. They also must be able to improvise. There is a training handbook, as well as tester and practice questions.
Andy: Has there been a case study yet showing how a question box has raised a person’s quality of life?
Rose: Not yet. The main evidence we have is by looking at the questions and figuring out the implied value. A researcher would be a welcome addition.
Andy: What do you do with the data after it’s logged?
Rose: Right now it’s being saved for analysis. We’re interested in what is being asked, by whom, and if they are coming back or not. Also, it is useful to understand how long it takes the Operator to search a query. It also helps us track the Operator’s performance.
Andy: Have there been concerns about user privacy?
Rose: Right now, no. Users are identified by first name only.
Andy: So how many people are working on the project right now? Could you give us a breakdown of the roles involved?
Rose: Let’s see – me, our CTO, 3 volunteers, 3 Indian MBAs in Pune, plus their professor; 2 engineers in Pune, plus a few people in Uganda who are with a new partner organization. Then there’s the Operators, a Board of Advisors, plus a slew of informal advisors. Not to mention our lawyer… It’s quite a collection, honestly.
Andy: In terms of infrastructure, what does the project look like from the organizational end? What kind of technologies do you use to collaborate with remote volunteers, workers and administrators?
Rose: That’s the least developed part of our org. Right now, it is all email, chat and calls. We would love to be introduced to a better system that allowed the team to cross-pollinate, a place, that was free or at least affordable.
Andy: Have you ever encountered resistance to a Question Box deployment from local authorities or community figures?
Rose: None so far. Before going into a place, we get permissions to make sure it has the proper benediction.
Andy: In a perfect world, what will the Question Box project look like in 5 years?
Rose: Question Box would be widely available to people in the developing world, both through their phones and via Question Boxes in storefronts. Millions of questions would be answered annually, and it would be the place to go for information. Basically, that sounds like a lot of fun.
Andy: Given it’s a free service, how do you envision the project sustaining itself monetarily moving forward?
Rose: It will be a mix of user payment per call, sponsorships, and possibly underwriting from foundations or the government.
Andy: Where the service isn’t pay-per-call, can you see potential dangers to the freedom and impartiality of the information given?
Rose: I think in a government case, it would either have to be an unfettered Question Box, or else a truncated QB service that was dealing only with government inquiries. Generally, if the QB is sponsored, I don’t think sponsorship ought to be allowed to affect what answers come up.
Andy: You’re probably aware of the One Laptop per Child initiative, and the controversy that erupted when it encountered competition from corporate interests. Would Question Box welcome a similar for-profit competitor?
Rose: Ideally, it would be a for-profit collaborator.
Andy: Less than ideally?
Rose: Less than ideally, it would be direct competition.
Andy: If a similar for-profit initiative leapfrogged Question Box entirely, wouldn’t Question Box’s aims be served either way?
Rose: Of course. But it’s always nice to see something to completion. That’s part of the adventure.
Andy: So what are your key goals for the coming year, and how can people help?
Rose: Key goals are to get funding to drive the service to the next level, both in terms of reach and sophistication. Help is welcome across the board – engineers, programmers, project managers, researchers, interns – they are all going to be needed in spades, and actually could be useful even now. We also have a PayPal “Donate Now” button on our website, www.questionbox.org
It’s an interesting concept, but the fundamental problem with it is that it doesn’t contribute to people’s own capacity to find, analyse and utilise information effectively. If there’s somebody else answering their questions, they aren’t necessarily being provided with the information that they want. It’s all well and good saying that the service providers are English literate, but they aren’t necessarily information literate – a key factor in such a service. If these villagers are relying on somebody else to provide their information, they don’t acquire the skills to find it themselves – that’s not smart, because they don’t empower themselves to deal with an increasingly information oriented world. Perhaps the project would be more effective, if it functioned instead as a tool to teach people how to find their own information and installed computers with Internet access and Skype.
I think it’s an interesting concept.
Question #1: If you’re recieving government funding, who’s to say that the government isn’t influencing the answers that you get? Or corporations? Say you ask where you can get a Coke, and although you can get one in the store across the street the operator tells you that s/he cannot find one near you, but there is a Pepsi vendor down the block?
Question #2: If one day voice recognition software gets to the level of ability and availability, would they do away with the operators and have a completely self-sustaining system?
@Trishna
The reliance of the users on the Operator is of course less desirable than giving everyone a net connection and a thorough education. But such a goal is long-term, the villages where Question Boxes are deployed have underdeveloped infrastructure – far from being able to support broadband. Illiteracy is also a factor at work here. Ultimately these obstacles *need* to be overcome, but they are of a much larger scope than what Question Box aims to address.
@Zach
Q1 – See Rose’s answer near the bottom. I’m speculating but I’d think the key to retaining impartiality in this case would be to keep the Operators and their responsibilities insulated from the sponsoring body.
Q2 – That’s a very interesting scenario, but I don’t think it’s likely. You would need a combination of some powerful artificial intelligence, voice recognition, and voice generation – tailored for each local language. This is a lot of effort for a project aiming to cover the interim transition of web access to remote areas. As Trishna mentioned, education is a much higher priority in the long term.
@Trishna
@ Trisha, Andy –
Yes, Andy’s right here. In India, over 40% of people can’t read. For a population of 1.2 billion, there’s only 5 million or so Internet connections. Longterm, yes, everyone searching on their own is a great vision. However, Question Box begins solving people’s needs now. Most queries are quite straightforward – ex., the price of cauliflower in town X – and the Operators handle them ably.
@ Zack, Andy –
1. Re: Government, corporate funding – An interesting question, and we will learn by doing. Regarding corporate sponsorships, I think it would be odd to curtail the results to sell more Pepsi, but sponsors should have a right to play a message while people are waiting for search results. For the Government, I’d presume it would be branded as a government service powered by Question Box, meaning there’s full disclosure, and probably a limited set of topics for that service (like how to get land titles, etc.)
2. Re: voice recognition, Andy pretty much sums it up. People in India, even middle class ones, still aren’t using voice mail, let alone voice recognition. It’s going to be awhile before that works for mass audiences.
Thanks to everyone for reading Andy’s interview! Please feel free to get in touch with me via http://www.questionbox.org.
It interesting to see such projects meant to overcome the physical infrasructural and financial difficulties to the poorest of poor masses. People may have many hypothetical ot hitherto unforeseen questions in mind about the successfulness of the project. But what I think at the moment is Rose has done a marvellous job which requires an awful kind of courage to serve back the very mass/public purposefully from which you have risen with some skill. When something is inititaed people see it with much skeptics, instead of admiring and encouraging the efforts.
The project is indeed meant for the poors who even struggle to meet both the ends, forget skilling and equipping them with computers and other such gadgets. In such circumstances, when they are able to know about the health problems, domestic problems ar any sort of information they might need in day to day life, at least Question Box can be ideal to equip them with some sort of information which they are deprived of otherwise. Even this can help them transmit the information from their remote village during some calamitous emergency, some attacks by bandits or so, like in naxalite infested areas and likes. Congratulations Rose for your novel efforts using your IT skills to serve the poor people. How many IT professionals we see even think of poor people leave alone serving them with their skills. Good Bavo Buck up.
Interview with Question Box and Open Mind founder.