Unveiling and Engaging with the Humans of Networking Research

Networking research often treats the Internet as an abstract system to be measured, modeled, and optimized. However, those measurements are always taken in specific regions, under particular economic and political constraints, and for specific populations. Ignoring this context does not make research more objective; it simply hides whose experiences are being represented and whose are being excluded.

These limitations prompted my colleagues and I to reflect on how previous approaches align (or fail to align) with the norms of networking research, and how we could potentially argue for a change.

In our HotNets 2025 paper, we discuss the need to consider individuals' and communities' participation in the network research process, as well as the researcher’s positionality, to improve the social context and rigor of the research.

While Internet research itself began in relatively structured and institutionally driven contexts (for example, DARPA and the NSF), the field has since expanded to encompass a much broader range of methods and perspectives.

Action research (AR), Participatory Action Research (PAR), and Participatory Design (PD) approaches are employed in various studies to integrate community participation directly into the research output, often including those from socially, culturally, or financially marginalized backgrounds.

Allowing participants to express their concerns helps better understand how things are done or perceived. For example:

• Through a series of PD workshops, researchers could understand how communities balance their privacy and security concerns when using various sensors in their homes, such as audio or video surveillance systems, installed in shelters in the Pacific Northwest of the USA.
• The Seattle Community Network (SCN) uses AR to bridge the digital infrastructure gap by leveraging community-owned, community-operated wireless Internet infrastructure in urban settings.
• Rural women’s concerns and opinions, which are often neglected, were incorporated when agricultural and weather-related mobile applications were developed in Bangladesh. This work eventually elevated the social status of women who were able to access and disseminate adequate information using technology.

In this broader landscape, incorporating AR/PAR into networking research offers a way to reason about impact—not as an afterthought, but as part of the research process itself.

This shift also motivates the use of ethnographic methods alongside traditional network measurement methodscould access and disseminate adequate information through. Rather than asking only whether a system “improves” outcomes, ethnography allows researchers to examine how networks are actually experienced, adapted to, and worked around by individuals and communities in practice.

Explicitly acknowledging the researchers' positionality can add significant value to understanding the problem and solution spaces, keeping the doors open to alternative perspectives from diverse backgrounds. For example, a researcher based in the Global North with abundant infrastructure and institutional support may approach networking challenges very differently from a researcher working in a resource-constrained setting in the Global South.

These differences can be further shaped by intersecting factors such as gender, race, religion, and social position, even within a shared research agenda. Eliminating hierarchy at the data level across nations can enable the dissemination of adequate networking information, which might provide a more transparent way of connectivity.

We offer these as initial recommendations, meant to be suggestive rather than prescriptive, and to be refined by future work:

• Make partnerships explicit: Document how collaborations are structured, including power asymmetries, economic differences, and institutional roles—not to constrain the research, but to provide necessary context for interpreting its outcomes.
• Treat informal knowledge as data: Many insights emerge through observations, conversations, and day-to-day interactions that never make it into formal datasets. These qualitative signals should be recorded and discussed alongside quantitative measurements, not relegated to anecdote.
• Reflect on researcher positionality: Researchers’ backgrounds, resources, and lived experiences shape how problems are framed, studied, and interpreted. Making these perspectives explicit strengthens, rather than weakens, the rigor of the work.
• Engage beyond the networking silo: Questions about participation, power, and impact are actively studied in adjacent communities, such as Science and Technology Studies (STS), Human-Computer Interaction (HCI), and social science, but remain largely disconnected from traditional networking and system research. Creating space for sustained dialogue with these fields can surface blind spots and broaden the kinds of questions networking research is willing to ask.

There may be arguments for placing this kind of work exclusively in venues centered on human or societal concerns. We argue instead that these discussions belong squarely within systems and networking venues, where technical decisions and humane values are inseparable and where their interaction must be made explicit.

Read our HotNets 2025 paper 'Unveiling and Engaging with the Humans of Networking Research’ to learn more.

Nova Ahmed is a Professor at North South University, who returned to Bangladesh after completing her PhD at the Georgia Institute of Technology, where she works on ways to make systems more human-centric.

Contributors: Loqman Salamatian.

The views expressed by the authors of this blog are their own and do not necessarily reflect the views of the Internet Society.