Day two was harder for those of us who are non-techies, but seeing the process that goes into the design of these things is quite interesting. The mix of people was really key today.
Sara Farmer kicked off the day on exactly this point. Innovations, she said, is not about ideas, but a thing that comes out of a process that happens after you have your idea. Yesterday was the start of the engineering, figuring out what gap in the market we were filling for each of our user personae, and (crucially) finding a usable solution that would fill that gap. The landscape for coming up with solutions, however, is rapidly expanding, for the types of solutions we can build and analysis of the inputs we have to work with. For example, there are a lot of challenges that one could envision working with the “crowd” could help with, but crowd sourced data is a relatively recent tool for analysts, policy makers, and emergency responders. In addition, while Global Pulse is premised on the idea that we can get data in real time, we are also living in an era where some apps can be developed in 24 hours, so solutions to information problems can also potentially be developed in close to real time as well.
The potential sources of data and ability to harness that data brings with it a whole load of design challenges. First and foremost, when looking at data that is pulled in from disparate sources, who is ultimately responsible for the veracity of that data? In our case, it will inevitably be the Global Pulse itself. This speaks also to the issue of trust. It’s important that users have the right balance between trusting the system too much and not enough.
In addition, this “new data” requires new types of analysis, and these methodologies may take longer to develop than the actual ability to acquire “new data”. For example, spatial reasoning: how does one work with data that moves in space? This is about more than working with panel data: when you are able to get images tracking migration movements in real time, how does that change how you approach patterns in behavior? Your analysis may revolve not around statistics, but perhaps around images. How does this change not only your analysis, but the types of questions you are able to ask?
Standards?
Much of the morning discussion from there revolved around data standards. The current thinking among the Global Pulse team is that the Global Pulse is a catalyst which sets up a framework in which people can test out ideas and innovate. As such, it does not start with standards.
There was much debate around this. It was pointed out that it is perhaps one of the most clear-cut roles of the UN to set global standards which can promote a public good, for example standards on international calls. The UN doesn’t manage telecommunication systems, but it does set the standards so that the system works. It was suggested that the UN could set the standards for impact monitoring data and invite people and governments to participate. Robert Kirkpatrick responded, noting that setting standards from the top-down is tricky, especially in an arena which relies as heavily as this on innovation.
From here, the rest of the day was spent in workshops. We first revisited our user personae from Day One, and developed our product in a box. The purpose here was to market the system to our different users, to distill the advantages for our users into a few easy slogans. To revisit the users I described yesterday:
Su-Yin, PhD Student in Sustainable Development
Su-Yin will primarily be tapping into the Global Pulse as a way to access data. The Global Pulse was marketed to Su-Yin with the slogans:
- Do what’s possible squared
- Effortlessly tag and manage data
- Itunes of data management: easy exchange between users
- Integrated platform for all your data needs
Sonia, Community Health Worker in Tazania
Sonia will primarily be a supplier of information to the Global Pulse operating system, although the system into which she inputs data should also feed back information. Sonia may never directly access the Global Pulse on-line; her interaction will be through an intermediary, most likely in the Ministry of Health, and most likely through her mobile phone. The tag-lines for Sonia were:
- Easy standardized reporting
- Tell us what you see, we’ll tell you how to use it and what others are seeing
- Access a stream of actionable information, training opportunities, alerts, stock information, and more
Zina, UN Worker (future Secretary General!)
The UN worker (who since yesterday was upgraded to the future Secretary General) was interested in evidence based advocacy, an ability to spot crisis, and an ability to garner political will from relevant parties. The key features were:
- Effective early warning
- New! Crowd sourced information from the most vulnerable!
- Holistic understanding by integrating UN sectoral analysis
- Compatibility with National Sovereignty
- Warning: not a decision maker
From our exploration of users, which in total took up almost a full day’s session, we were able to pull out key commonalities among users. We were also able to eke out thing that the Global Pulse is not.
The Global Pulse is (key features):
- Operable between different systems
- Control of ownership and sharing
- Headless (ie. it is a bunch of services that other people overlay a user interface on to)
- Standardizing data sets
- Translation of data
- Collaboration tools
- Connecting communities
- Social data, sharing
The Global Pulse is not:
- Statistical Evidence
- New Data Collection
- Bounded
- Composite index (like the Human Development Index, or a new poverty measurement)
From this, we split into new groups to do “feature decomposition”. We each took one of the “Global Pulse is. . .” points and broke it into particular features, being as specific as possible and developing a user interface with each of our points. We filled in the two statements (1) As a PERSONA, I want FULLFILL ACTION, so that I can DO X; and (2) This feature initiates XXX and is limited to XXX. We then were to fill our acceptance criteria, pulling out all the things that might be needed to effectively accomplish our goal. I was part of the social data and connections between communities group. At first, it seemed like these would have an overlap, but in reality they had almost no overlap at all. Social data involved, in essence, applying tools that are common on sites like Facebook, Ebay, and Netflix for data sharing. Connections between communities was much more difficult, since many of the communities we would like to connect may be interacting with the Global Pulse through an intermediary, like Sonia, our community health worker, who transmits and receives information through the Ministry of Health. The connections between communities here is primarily dependent on the national system (and the Pulse Labs), not the operating system.
And that was our day. The outputs were interesting, and for me, so was the process. In particular, coming from the social sciences, this user- centered approach is quite different. In essence, at Pulse Camp we’ve been trying to distill the value of the system for a group of individuals from very different populations. It’s interesting that in the world of international development, the methods are almost the opposite—interventions are designed around the aggregate needs of a specific population.
Day Three throws open the doors to discussion even wider than our user-centered process thus far. Throughout the two days, people have been posting notes with potential topics for further discussion on a board. Tomorrow we will convene around those topics. Will report back.
