Editor’s Note: Global health research is not easy to coordinate. Publicly shared global health research is even more complex. That is why last month, Ethnography Matters was so excited to feature Erik Bigras‘s and Kim Fortun‘s innovative research methods for The Asthma Files, a project at Rensselaer Polytechnic Institute where ethnographers gather and publicly share data about asthma. We believe their work signals to an important turn in policy oriented and public ethnography.
In Erik’s and Kim’s first post, Innovation in Asthma Research: Using Ethnography to Study a Global Health Problem (1 of 3), they focused on why The Asthma Files is necessary and introduced some of the technical logistics for creating a crowd-sourced qualitative data health gathering project.
In this month’s Ethnozine, Erik and Kim’s second post details the exciting process of choosing the best data sharing platform for their project, Plone. We learn about how the Tehran Asthma Files was born out of a close collaboration with the Samuel Jordan Center for Persian Studies and Culture at the University of California, Irvine.
We look forward to their final post in this series that will discuss how other researchers from social scientists to epidemiologists and global health experts can participate in the research project and make use of the data.
Check out past posts from guest bloggers.
_________________________________________
Choosing the Right Platform
Collaborating with other disciplines (here, data science) allows us to better understand the ways in which scientific knowledge is able to cross particular boundaries.
Our ethnographic experiments are made possible partly because of the choice of online platform that The Asthma Files uses. Choosing the right platform is anything but simple. Each platform has its own capabilities, and these don’t necessarily align with the goals of the project. For The Asthma Files, we’ve so far been through three different platforms. We eventually settled on Plone because it was the one most suited to our needs.
As we said in our previous post, one of the goals of The Asthma Files is to rethink the everyday work of ethnography. In order words, we’re trying to understand how digital environments can transform the everyday, mundane, things that ethnographers do. As such, one of our primary audiences is ourselves as ethnographers and researchers. For this purposes, we needed an online platform that would be more than a delivery mechanism. We needed something that would act as a fully developed workspace where we could share, store, and create material.
The first two platforms that we used were a wiki platform and Ning. However, it quickly became apparent to us that neither of these would be adequate. As social networking platforms, they offered many advantages when it came to content delivery. Both were relatively easy and intuitive to use. However, neither really allowed us to store and share material easily, nor to experiment with new types of workflows. We therefore needed a new option.
Cue Plone, an open-source content management system. The fact that it is open-source is important to us. It means that we’re able to modify it to suit our needs, and then share these modifications with other social scientists and humanities scholars who also want to experiment with online work spaces.
Despite it’s advantages, Plone does not possess extensive documentation. As such, the Plone community organizes itself in order to be able to alleviate this issue. Here, Erik Bigras logs into IRC, one of the main media through which the Plone community becomes accessible and tangible.
Also important to us is that Plone is optimized for more than simply content delivery. It allows us to store and share files easily and securely. This is especially important when it comes to IRB-protected material. Social networking platform security mainly relies on anonymity to create security: something is secure if no one knows that it is there.
For example, such platforms differentiate between public (unsecured) and private (secured) pages. Public pages are those pages available for everyone to see. Private pages, on the other hand, are available in a different folder on the server, and are only available if 1) one knows the URL to the page, or 2) one has FTP access to the server. While Plone does differentiate between public and private, it also takes a different approach. It allows individual Plone folders to be password-protected so that the material inside can only be accessed by the users with the appropriate permissions. Plone security is not contingent on the existence of a space beyond the platform. It is created within the platform.
So with most social networking platforms, if the audio files from interlocutor A and interlocutor B are both in the private space, interlocutor A will potentially be able to see interlocutor B’s files when he or she goes to access his or her materials for review. With Plone, this is impossible unless interlocutor A also knows the password to interlocutor B’s folder. Also, because Plone’s security is handled at the level of the platform, interlocutors are not required to install any kinds of software on their own system (such as FTP clients).
Another important consideration is the way in which Plone organizes itself. As we explained in our previous post, we try to use a logic of juxtaposition to create new understandings of the world. Most social networking platforms rely on the creation of new elements that can then be added to the overall collection of material. With Plone, however, the overall goal of users is to bring pages together through juxtaposition in order to create a sense of play where the newness is not in the information being displayed but in the ways in which the juxtapositions were created in the first place. This particular feature allows us to put material together in news ways in order to provide ourselves and our interlocutors with a different understanding of asthma research.
Ethnography from Up Close and from Afar
Having a digital platform for ethnographic work has changed and enabled our practice in a number of ways. First, it has changed the way individual ethnographers work, giving each of involved a place to archive and visualize our material within a structure that situates our individual work within the larger project; shared questions stimulate without over constraining our work; material contributed by others provides context and comparisons.
The digital platform has also enabled us to work together, sharing material and analysis at an early stage of development, even if we aren’t co-located physically. This enables a type of peer review and collective deliberation that is worth nurturing. It also allows us to bring in junior scholars (including undergraduates and high school students) who work alongside us, in something of an apprenticeship mode; they have their own focus and space to work, with guiding structures and examples.
A snippet of a conversation in the Tehran Asthma Files forum. Here, participants are discussing the pollution sources found in Tehran, in Farsi and English.
The digital platform has also allowed new kinds of collaborations, which bring really differently focused researchers together. For the last year, for example, we have had a sustained collaboration with the Samuel Jordan Center for Persian Studies and Culture at the University of California, Irvine, to build a set of Tehran Asthma Files. The Iran Studies specialists are interested in our project because it provides structure for a new kind of interdisciplinary collaboration within area studies.
We are interested in Tehran’s dramatic air quality problem, and how it is being governed, but couldn’t get very far trying to explicate this without area studies expertise. Again, The Asthma Files has provided the structure and space for a new kind of work to play out. The digital platform has also allowed us to work organically with Tehran Asthma Files, customizing how we proceed in a way that makes sense to all involved. We began the Tehran-focused collaboration with a face-to-face workshop in Irvine last May, for example, followed by monthly Skype meetings to figure out next steps.
Through this, we decided to experiment with an online discussion to jumpstart our understanding of how variously positioned people understand air quality and environmental health in Tehran. This web forum – rich with images and videos to think with – is what we came up with. In the future, we will further experiment with ways to use The Asthma Files web platform to engage with our colleagues and interlocutors in new ways, using the platform to both generate and share ethnographic understanding.
Erik Bigras in the process of customizing the python code of a web forum add-on that he just installed to permit online conversations.
These new engagements have allowed us to cross disciplinary boundaries both personally and collaboratively. Personally, ever since we’ve started using Plone, we’ve had to act as occasional programmers and full-time system administrators. While some of the researchers on the project do possess some computer-related expertise, these were nonetheless types of work that we, as ethnographers, were not originally trained to do. However, we’re now successfully able to install and manage our own platform.
Collaboratively, we’ve had to seek out different kinds of expertise whenever our own knowledges were incomplete. Using Plone has put us in contact with web designers, computer scientists, and database managers, all of whom have had some impact in the ways in which The Asthma Files is shaped. In fact, due to some of Plone’s intricacies, we’ve had repeated interactions with the Plone community of developers and users itself. Each of these encounters becomes a micro-ethnographic moment that allows us to try to understand why we would want to think of The Asthma Files in the ways that we do.
So far, our series of posts have dealt with what the project is, and how it helps us do our everyday ethnographic work. The final post in this series will detail how other researchers can become involved with The Asthma Files.
IN THIS SERIES
Really intrigued by this project. One thing I’m curious to hear more about is some of the nitty gritty involved in sharing data. Are there any processes that help to smooth the path to sharing, or has that not really been necessary (or am I misunderstanding how this works?)
When I think of trying to get a collaboration like this through IRBs at some institutions I have worked with, it seems pretty daunting. Especially when you’re dealing with health-related data, data from kids, data describing illegal substance use, data from groups such as Native Americans that have been burned repeatedly by researchers, etc.
Tends to be easier for the researchers and less worrisome for communities to handle data in the ways people have become comfortable with over time. Which totally makes sense, but on the other hand keeping ethnographic data locked away can contribute to the invisibility of communities that are the most marginalized (+ often the most at risk).
Anyway, not sure if any of that actually makes any sense. Just wondering I guess of you run into any of those concerns. Or if you have any sort of templates or best practices for things like streamlining sharing, managing confidentiality, putting IRBs and communities at ease?
I meant to reply to this sooner, my apologies.
I’ll begin by addressing the IRB. I’ve forwarded your comment to Kim as she’ll have more experience with the existing IRB arrangement, but I can offer a preliminary answer from the perspective of my own dissertation, given that I’m dealing with data sharing as well.
For my own dissertation project (I focus on data models and scientific communities), I proposed to make my data accessible online as a public resource. Kim (she’s my advisor) was very supportive of this decision, but we both knew that making it happen would be tricky, precisely because of IRB concerns for privacy and confidentiality. However, we managed to pull it off.
I think that the main way to avoid any major issues is to open up negotiations early with your IRB and potential interlocutors. I submitted a first proposal to the IRB and they quickly returned a list of what can only be called major concerns. However, reading through the list, Kim and I quickly realized a few things.
First, they weren’t very knowledgeable about the existing debates in academia concerning open access and data sharing. One member told us that, before reading my proposal, he had never heard about copyleft. So part of my job wasn’t revising my proposal to meet stringent privacy rules so much as it was putting it into the right context.
Second, many of the concerns revolved around the data being online. However, reading the comments more closely, we realized that the concerns really revolved around the platform. Most IRB members, if they had any experience with online publication, were mostly familiar with blogs. However, once I explained how Plone worked, the concerns largely disappeared. Choosing the right platform, and being knowledgeable about it, goes a long way.
Last, negotiating with your interlocutors is equally important. For precisely the reasons you mention, many of your interlocutors will agree to have the data posted online as a public resource. In fact, I’ve had interlocutors that were very eager to see that happen. It’s important to let the IRB know this, and it can be accomplished partly through the consent form. Ensure that it explains how the data will be stored and give your interlocutors the option to opt in or out.
We’re using this experience as a model by which to negotiate The Asthma Files’ next IRB renewal, and we’re optimistic that it’ll allow us to greatly facilitate our data sharing efforts.
Now, about the nitty gritty of data sharing. Part of the answer to this question is a technical one. One of the reason that we chose Plone over other platforms (such as Drupal or WordPress), is that it allows for files to be very easily centralized, curated, and shared. So, on a technical level, file sharing is not a problem.
Concerning best practices, however, that’s where we hit a snag. Anyone who attempts a collaborative data sharing anthropological project quickly realizes that there is no set of best practices out there. Common practices often are limited to uploading a file somewhere, and then making it visible through a hyperlink. Often, questions of metadata and relationality aren’t really addressed. There are, of course, exceptions to this. For example, Open Folklore is a project that pays great attention to its metadata.
Part of what The Asthma Files is trying to do is coming up with a set of best practices for data sharing (caption your images, enter your metadata, relate your uploads to other files in the database, etc). We’re still struggling with this, partly for technical reasons, but also because it isn’t something that, as anthropologists, we’re usually trained to do.
However, we’re slowly coming up with a set of practices that, we think, can be share-able across platforms and projects. Part of my job in the upcoming months will be to actually write those down and make them available on the site. One of the issues we’re encountering is that, in order for best practices to be share-able, the platform’s shell itself also needs to be share-able in some capacity. That, however, is easier said than done. We have a team member who is currently looking at that particular issue, but we don’t yet know how difficult it will be.
This is *fantastic*. (And I’m sorry that I am just coming across your reply now — don’t know how I missed it). I have been exploring Plone a bit. Sadly all the FOSS options have been a stumbling block with one institution I work with, but so far we’ve managed to get around it (*crosses fingers*).
I don’t know if you have come across this digital anthropology interest group (https://01anthropology.wordpress.com/). When they met for the first time at the American Anthro Assoc meeting this year, there was talk of trying to draw up best practices and ethical guidelines for digital research. The Asthma Files looks like a great example to explore and learn from.
As you mentioned, there is a lot of communication with IRBs that needs to happen for things to change. And communication with interlocutors. The “standard” path faces the least resistance, but I’m not sure it’s really the best way at this point to respect and care for the interests of research participants.
Anyway, lots to think about. I really appreciate your detailed response, and am looking forward to hearing more of your thoughts on best practices.