Author: Benjamin Mako Hill

More than a billion people visit Wikipedia each month and millions have contributed as volunteers. Although Wikipedia exists in 300+ language editions, more than 90% of Wikipedia language editions have fewer than one hundred thousand articles. Many small editions are in languages spoken by small numbers of people, but the relationship between the size of a Wikipedia language edition and that language’s number of speakers—or even the number of viewers of the Wikipedia language editions—varies enormously. Why do some Wikipedias engage more potential contributors than others? We attempted to answer this question in a study of three Indian language Wikipedias that will be published and presented at the ACM Conference on Human Factors in Computing (CHI 2022).

To conduct our study, we selected 3 Wikipedia language communities that correspond to the official languages of 3 neighboring states of India: Marathi (MR) from the state of Maharashtra, Kannada (KN) from the state of Karnataka, and Malayalam (ML) from the state of Kerala (see the map in right panel of the figure above). While the three projects share goals, technological infrastructure, and a similar set of challenges, Malayalam Wikipedia’s community engaged its language speakers in contributing to Wikipedia at a much higher rate than the others. The graph above (left panel) shows that although MR Wikipedia has twice as many viewers as ML Wikipedia, ML has more than double the number of articles on MR.

Our study focused on identifying differentiating factors between the three Wikipedias that could explain these differences. Through a grounded theory analysis of interviews with 18 community participants from the three projects, we identified two broad explanations of a “positive participation cycle” in Malayalam Wikipedia and a “negative participation cycle” in Marathi and Kannada Wikipedias. 

As the first step of our study, we conducted semistructured interviews with active participants of all three projects to understand their personal experiences and motivation; their perceptions of dynamics, challenges, and goals within their primary language community; and their perceptions of other language Wikipedia. 

We found that MR and KN contributors experience more day-to-day barriers to participation than ML, and that these barriers hinder contributors’ day-to-day activity and impede engagement. For example, both MR and KN members reported a large number of content disputes that they felt reduced their desire to contribute.

But why do some Wikipedias like MR or KN have more day-to-day barriers to contribution like content disputes and low social support than others? Our interviews pointed to a series of higher-level explanations. For example, our interviewees reported important differences in the norms and rules used within each community as well as higher levels of territoriality and concentrated power structures in MR and KN.

Once again, though: why do the MR and KN Wikipedias have these issues with territoriality and centralized authority structures? Here we identify a third, even higher-level set of differences in the social and cultural contexts of the three language-speaking communities. For example, MR and KN community members attributed low engagement to broad cultural attitudes toward volunteerism and differences in their language community’s engagement with free software and free culture.

The two flow charts above visualize the explanatory mapping of divergent feedback loops we describe.  The top part of the figure illustrates how the relatively supportive macro-level social environment in Kerala led to a larger group of potential contributors to ML as well as a chain reaction of processes that led to a Wikipedia better able to engage potential contributors. The process is an example of a positive feedback cycle. The second, bottom part of the figure shows the parallel, negative feedback cycle that emerged in MR and KN Wikipedias. In these settings, features of the macro-level social environment led to a reliance on a relatively small group of people for community leadership and governance. This led, in turn, to barriers to entry that reduced contributions. 

One final difference between the three Wikipedias was the role that paid labor from NGOs played. Because the MR and KN Wikipedias struggled to recruit and engage volunteers, NGOs and foundations deployed financial resources to support the development of content in Marathi and Kannada, but not in ML to the same degree. Our work suggested this tended to further concentrate power among a small group of paid editors in ways that aggravated the meso-level community struggles. This is shown in the red box in the second (bottom) row of the figure.

The results from our study provide a conceptual framework for understanding how the embeddedness of social computing systems within particular social and cultural contexts shape various aspects of the systems. We found that experience with participatory governance and free/open-source software in the Malayalam community supported high engagement of contributors. Counterintuitively, we found that financial resources intended to increase participation in the Marathi and Kannada communities hindered the growth of these communities. Our findings underscore the importance of social and cultural context in the trajectories of peer production communities. These contextual factors help explain patterns of knowledge inequity and engagement on the internet. 

---

Please refer to the preprint of the paper for more details on the study and our design suggestions for localized peer production projects. We’re excited that this paper has been accepted to CHI 2022 and received the Best Paper Honorable Mention Award! It will be published in the Proceedings of the ACM on Human-Computer Interaction and presented at the conference in May. The full citation for this paper is:

Sejal Khatri, Aaron Shaw, Sayamindu Dasgupta, and Benjamin Mako Hill. 2022. The social embeddedness of peer production: A comparative qualitative analysis of three Indian language Wikipedia editions. In CHI Conference on Human Factors in Computing Systems (CHI ’22), April 29-May 5, 2022, New Orleans, LA, USA. ACM, New York, NY, USA, 18 pages. https://doi.org/10.1145/3491102.3501832

If you have any questions about this research, please feel free to reach out to one of the authors: Sejal Khatri, Benjamin Mako Hill, Sayamindu Dasgupta, and Aaron Shaw. 

Attending the conference in New Orleans? Come attend our live presentation on May 3 at 3 pm at the CHI program venue, where you can discuss the paper with all the authors. 

On November 5, folks from the Community Data Science Collective held an open session to allow folks to answer questions about our PhD programs. We’ve posted a 30 minutes video of the first half of the event on our group YouTube channel that includes an introduction to the group and answers some basic questions that attendees had submitted in advance.

If you have additional questions, feel free to reach out to the CDSC faculty or to anybody else in the group. Contact information for everybody should be online.

Keep in mind that the first due date (University of Washington Department of Communication MA/PhD program) is November 15, 2021.

The rest of the deadlines at Purdue (Brian Lamb School of Communication), Northwestern (Media, Technology & Society; Technology & Social Behavior), and UW (Human-Centered Design & Engineering; Computer Science & Engineering; Information School) are in December—mostly on December 1st.

Should online communities require people to create accounts before participating?

This question has been a source of disagreement among people who start or manage online communities for decades. Requiring accounts makes some sense since users contributing without accounts are a common source of vandalism, harassment, and low quality content. In theory, creating an account can deter these kinds of attacks while still making it pretty quick and easy for newcomers to join. Also, an account requirement seems unlikely to affect contributors who already have accounts and are typically the source of most valuable contributions. Creating accounts might even help community members build deeper relationships and commitments to the group in ways that lead them to stick around longer and contribute more.

In a new paper published in Communication Research, Benjamin Mako Hill and Aaron Shaw provide an answer. We analyze data from “natural experiments” that occurred when 136 wikis on Fandom.com started requiring user accounts. Although we find strong evidence that the account requirements deterred low quality contributions, this came at a substantial (and usually hidden) cost: a much larger decrease in high quality contributions. Surprisingly, the cost includes “lost” contributions from community members who had accounts already, but whose activity appears to have been catalyzed by the (often low quality) contributions from those without accounts.

---

The full citation for the paper is: Hill, Benjamin Mako, and Aaron Shaw. 2020. “The Hidden Costs of Requiring Accounts: Quasi-Experimental Evidence from Peer Production.” Communication Research, 48 (6): 771–95. https://doi.org/10.1177/0093650220910345.

If you do not have access to the paywalled journal, please check out this pre-print or get in touch with us. We have also released replication materials for the paper, including all the data and code used to conduct the analysis and compile the paper itself.

In exciting collective news, the US National Science Foundation announced that Benjamin Mako Hill has received of one of this year’s CAREER awards. The CAREER is the most prestigious grant that the NSF gives to early career scientists in all fields.

You can read lots more about the award in a detailed announcement that the University of Washington Department of Communication put out, on Mako’s personal blog (or in this Twitter thread and this Fediverse thread), or on the NSF website itself. The grant itself—about $550,000 over five years—will support a ton of Community Data Science Collective research and outreach work over the next half-decade. Congratulations, Mako!

Like everyone else, Internet users who protect their privacy by using the anonymous browsing software Tor are welcome to read Wikipedia. However, when Tor users try to contribute to the self-described “encyclopedia that anybody can edit,” they typically come face-to-face with a notice explaining that their participation is not welcome.

Our new paper—led by Chau Tran at NYU and authored by a group of researchers from the University of Washington, the Community Data Science Collective, Drexel, and New York University—was published and presented this week at the IEEE Symposium on Security & Privacy and provides insight into what Wikipedia might be missing out on by blocking Tor. By comparing contributions from Tor that slip past Wikipedia’s ban to edits made by other types of contributors, we find that Tor users make contributions to Wikipedia that are just as valuable as those made by new and unregistered Wikipedia editors. We also found that Tor users are more likely to engage with certain controversial topics.

To conduct our study, we first identified more than 11,000 Wikipedia edits made by Tor users who were able to bypass Wikipedia’s ban on contributions from Tor between 2007 and 2018. We then used a series of quantitative techniques to evaluate the quality of these contributions. We found that Tor users made contributions that were similar in quality to, and in some senses even better than, contributions made by other users without accounts and newcomers making their first edits.

We used a range of analytical techniques including direct parsing of article histories, manual inspections of article changes, and a machine learning platform called ORES to analyze contributions. We also used a machine learning technique called topic modeling to analyze Tor users’ areas of interest by checking their edits against clusters of keywords. We found that Tor-based editors are more likely than other users to focus on topics that may be considered controversial, such as politics, technology, and religion.

In a closely connected study led by Kaylea Champion and published several months ago in the Proceedings of the ACM on Human Computer Interaction (CSCW), we conducted a forensic qualitative analysis of contributions of the same dataset. Our results in that study are described in a separate blog post about that project and paint a complementary picture of Tor users engaged—in large part—in uncontroversial and quotidian types of editing behavior.

Across the two papers, our results are similar to other work that suggests that Tor users are very similar to other internet users. For example, one previous study has shown that Tor users frequently visit websites in the Alexa top one million.

Much of the discourse about anonymity online tends toward extreme claims backed up by very little in the way of empirical evidence or systematic study. Our work is a step toward remedying this gap and has implications for many websites that limit participation by users of anonymous browsing software like Tor. In the future, we hope to conduct similar systematic studies in contexts beyond Wikipedia.

In terms of Wikipedia’s own policy decisions about anonymous participation, we believe that our paper suggests that the benefits of a “pathway to legitimacy” for Tor contributors to Wikipedia might exceed the potential harm due to the value of their contributions. We are particularly excited about exploring ways to allow contributors from anonymity-seeking users under certain conditions: for example, requiring review prior to changes going live. Of course, these are questions for the Wikipedia community to decide but it’s a conversation that we hope our research can inform and that we look forward to participating in.

---

Authors of the paper, “Are anonymity-seekers just like everybody else? An analysis of contributions to Wikipedia from Tor,” include Chau Tran (NYU), Kaylea Champion (UW & CDSC), Andrea Forte (Drexel), Benjamin Mako Hill (UW & CDSC), and Rachel Greenstadt (NYU). The paper was published at the 2020 IEEE Symposium on Security & Privacy between May 18 and 20. Originally to be held in San Francisco, the event was held digitally due to the COVID-19 pandemic. This blog post borrows with permission from this news release by Andrew Laurent at NYU.

Paper Citation: Tran, Chau, Kaylea Champion, Andrea Forte, Benjamin Mako Hill, and Rachel Greenstadt. “Are Anonymity-Seekers Just like Everybody Else? An Analysis of Contributions to Wikipedia from Tor.” In 2020 IEEE Symposium on Security and Privacy (SP), 1:974–90. San Francisco, California: IEEE Computer Society, 2020. https://doi.org/10.1109/SP40000.2020.00053.

The research was funded by the National Science Foundation.

In May 2019, we were invited to give short remarks on the impact of Janet Fulk and Peter Monge at the International Communication Association‘s annual meeting as part of a session called “Igniting a TON (Technology, Organizing, and Networks) of Insights: Recognizing the Contributions of Janet Fulk and Peter Monge in Shaping the Future of Communication Research.”

Mako Hill gave a four-minute talk on Janet and Peter’s impact to the work of the Community Data Science Collective. Mako unpacked some of the cryptic acronyms on the CDSC-UW lab’s whiteboard as well as explaining that our group has a home in the academic field of communication, in no small part, because of the pioneering scholarship of Janet and Peter. You can view the talk in WebM or on Youtube.

A good chunk of the Community Data Science Collective is in Austin, Texas this week for the 2019 ACM Conference on Computer-supported Cooperative Work and Social Computing (CSCW).

The conference marks the official publication of four papers by collective students and faculty. All four papers were published in the journal Proceedings of the ACM on Human-Computer Interaction: CSCW.

Information on the talks as well as links to the papers are available here (CSCW members are listed in italics):

• Mon, Nov 11 11:00 – 12:30: Technological Frames and User Innovation: Exploring Technological Change in Community Moderation Teams by Charles Kiene (UW), Aaron Jiang (UC Boulder), and Benjamin Mako Hill (UW). Charlie will be presenting. [Blog post] [Paper]
• Mon, Nov 11 14:30 – 16:00: All Talk: How Increasing Interpersonal Communication on Wikis May Not Enhance Productivity by Sneha Narayan (Carleton College), Nathan TeBlunthuis (UW), Wm Salt Hale (UW), Benjamin Mako Hill (UW), and Aaron Shaw (Northwestern). Sneha will present! [Paper]
• Mon, Nov 11 14:30 – 16:00: A Forensic Qualitative Analysis of Contributions to Wikipedia from Anonymity-Seeking Users by Kaylea Champion (UW), Nora McDonald (Drexel), Stephanie E Bankes (Drexel), Joseph Zhang (Drexel), Rachel Greenstadt (NYU), Andrea Forte (Drexel), Benjamin Mako Hill (UW). Kaylea will present! [Paper]
• Tue, Nov 12 11:00 – 12:30: Challenges in Voice-based Online Communities On Discord by Aaron Jiang (UC Boulder), Charles Kiene (UW), Skyler Middler (UC Boulder), Jed R. Brubaker (UC Boulder), and Casey Fiesler (UC Boulder). Aaron will present! [Blog post] [Paper]

Additionally, Charlie Kiene is co-organizing a workshop on moderation:

• Sun, Nov 10 09:00 – 17:00: Volunteer Work: Mapping the Future of Moderation Research

And Sneha Narayan and Andrés Monroy-Hernández is participating in a panel:

• Mon, Nov 11 11:00 – 12:30: Patron or Poison? Industry Funding of HCI Research

Finally, Mako Hill is chairing a session:

• Mon, Nov 11 14:30 – 16:00: Wikipedia and Wiki Research

Salt, Kaylea, Charlie, Regina, and Kaylea will all be at the conference as will affiliate Andrés Monroy-Hernández and tons of our social computing friends. Please come and say “Hello” to any of us, introduce yourself if you don’t already know us, and pick up a CDSC sticker!

Online anonymity often gets a bad rap and complaints about antisocial behavior from anonymous Internet users are as old as the Internet itself. On the other hand, research has shown that many Internet users seek out anonymity to protect their privacy while contributing things of value. Should people seeking to contribute to open collaboration projects like open source software and citizen science projects be required to give up identifying information in order to participate?

We conducted a two-part study to better understand how open collaboration projects balance the threats of bad behavior with the goal of respecting contributors’ expectations of privacy. First, we interviewed eleven people from five different open collaboration “service providers” to understand what threats they perceive to their projects’ mission and how these threats shape privacy and security decisions when it comes to anonymous contributions. Second, we analyzed discussions about anonymous contributors on publicly available logs of the English language Wikipedia mailing list from 2010 to 2017.

In the interview study, we identified three themes that pervaded discussions of perceived threats. These included threats to:

1. community norms, such as harrassment;
2. sustaining participation, such as loss of or failure to attract volunteers; and
3. contribution quality, low-quality contributions drain community resources.

We found that open collaboration providers were most concerned with lowering barriers to participation to attract new contributors. This makes sense given that newbies are the lifeblood of open collaboration communities. We also found that service providers thought of anonymous contributions as a way of offering low barriers to participation, not as a way of helping contributors manage their privacy. They imagined that anonymous contributors who wanted to remain in the community would eventually become full participants by registering for an account and creating an identity on the site. This assumption was evident in policies and technical features of collaboration platforms that barred anonymous contributors from participating in discussions, receiving customized suggestions, or from contributing at all in some circumstances. In our second study of the English language Wikipedia public email listserv, we discovered that the perspectives we encountered in interviews also dominated discussions of anonymity on Wikipedia. In both studies, we found that anonymous contributors were seen as “second-class citizens.”

This is not the way anonymous contributors see themselves. In a study we published two years ago, we interviewed people who sought out privacy when contributing to open collaboration projects. Our subjects expressed fears like being doxed, shot at, losing their job, or harassed. Some were worried about doing or viewing things online that violated censorship laws in their home country. The difference between the way that anonymity seekers see themselves and the way they are seen by service providers was striking.

One cause of this divergence in perceptions around anonymous contributors uncovered by our new paper is that people who seek out anonymity are not able to participate fully in the process of discussing and articulating norms and policies around anonymous contribution. People whose anonymity needs means they cannot participate in general cannot participate in the discussions that determine who can participate.

We conclude our paper with the observation that, although social norms have played an important role in HCI research, relying on them as a yardstick for measuring privacy expectations may leave out important minority experiences whose privacy concerns keep them from participating in the first place. In online communities like open collaboration projects, social norms may best reflect the most privileged and central users of a system while ignoring the most vulnerable

---

Both this blog post and the paper, Privacy, Anonymity, and Perceived Risk in Open Collaboration: A Study of Service Providers, was written by Nora McDonald, Benjamin Mako Hill, Rachel Greenstadt, and Andrea Forte and will be published in the Proceedings of the 2019 ACM CHI Conference on Human Factors in Computing Systems next week. The paper will be presented at the CHI conference in Glasgow, UK on Wednesday May 8, 2019. The work was supported by the National Science Foundation (awards CNS-1703736 and CNS-1703049).

I’ve heard a surprising “fact” repeated in the CHI and CSCW communities that receiving a best paper award at a conference is uncorrelated with future citations. Although it’s surprising and counterintuitive, it’s a nice thing to think about when you don’t get an award and its a nice thing to say to others when you do. I’ve thought it and said it myself.

It also seems to be untrue. When I tried to check the “fact” recently, I found a body of evidence that suggests that computing papers that receive best paper awards are, in fact, cited more often than papers that do not.

The source of the original “fact” seems to be a CHI 2009 study by Christoph Bartneck and Jun Hu titled “Scientometric Analysis of the CHI Proceedings.” Among many other things, the paper presents a null result for a test of a difference in the distribution of citations across best papers awardees, nominees, and a random sample of non-nominees.

Although the award analysis is only a small part of Bartneck and Hu’s paper, there have been at least two papers have have subsequently brought more attention, more data, and more sophisticated analyses to the question.  In 2015, the question was asked by Jaques Wainer, Michael Eckmann, and Anderson Rocha in their paper “Peer-Selected ‘Best Papers’—Are They Really That ‘Good’?“

Wainer et al. build two datasets: one of papers from 12 computer science conferences with citation data from Scopus and another papers from 17 different conferences with citation data from Google Scholar. Because of parametric concerns, Wainer et al. used a non-parametric rank-based technique to compare awardees to non-awardees.  Wainer et al. summarize their results as follows:

The probability that a best paper will receive more citations than a non best paper is 0.72 (95% CI = 0.66, 0.77) for the Scopus data, and 0.78 (95% CI = 0.74, 0.81) for the Scholar data. There are no significant changes in the probabilities for different years. Also, 51% of the best papers are among the top 10% most cited papers in each conference/year, and 64% of them are among the top 20% most cited.

The question was also recently explored in a different way by Danielle H. Lee in her paper on “Predictive power of conference‐related factors on citation rates of conference papers” published in June 2018.

Lee looked at 43,000 papers from 81 conferences and built a regression model to predict citations. Taking into an account a number of controls not considered in previous analyses, Lee finds that the marginal effect of receiving a best paper award on citations is positive, well-estimated, and large.

Why did Bartneck and Hu come to such a different conclusions than later work?

My first thought was that perhaps CHI is different than the rest of computing. However, when I looked at the data from Bartneck and Hu’s 2009 study—conveniently included as a figure in their original study—you can see that they did find a higher mean among the award recipients compared to both nominees and non-nominees. The entire distribution of citations among award winners appears to be pushed upwards. Although Bartneck and Hu found an effect, they did not find a statistically significant effect.

Given the more recent work by Wainer et al. and Lee, I’d be willing to venture that the original null finding was a function of the fact that citations is a very noisy measure—especially over a 2-5 post-publication period—and that the Bartneck and Hu dataset was small with only 12 awardees out of 152 papers total. This might have caused problems because the statistical test the authors used was an omnibus test for differences in a three-group sample that was imbalanced heavily toward the two groups (nominees and non-nominees) in which their appears to be little difference. My bet is that the paper’s conclusions on awards is simply an example of how a null effect is not evidence of a non-effect—especially in an underpowered dataset.

Of course, none of this means that award winning papers are better. Despite Wainer et al.’s claim that they are showing that award winning papers are “good,” none of the analyses presented can disentangle the signalling value of an award from differences in underlying paper quality. The packed rooms one routinely finds at best paper sessions at conferences suggest that at least some additional citations received by award winners might be caused by extra exposure caused by the awards themselves. In the future, perhaps people can say something along these lines instead of repeating the “fact” of the non-relationship.

---

This post was originally posted on Benjamin Mako Hill’s blog Copyrighteous.

Couchsurfing and Airbnb are websites that connect people with an extra guest room or couch with random strangers on the Internet who are looking for a place to stay. Although Couchsurfing predates Airbnb by about five years, the two sites are designed to help people do the same basic thing and they work in extremely similar ways. They differ, however, in one crucial respect. On Couchsurfing, the exchange of money in return for hosting is explicitly banned. In other words, couchsurfing only supports the social exchange of hospitality. On Airbnb, users must use money: the website is a market on which people can buy and sell hospitality.

The figure above compares the number of people with at least some trust or verification on both Couchsurfing and Airbnb based on when each user signed up. The picture, as I have argued elsewhere, reflects a broader pattern that has occurred on the web over the last 15 years. Increasingly, social-based systems of production and exchange, many like Couchsurfing created during the first decade of the Internet boom, are being supplanted and eclipsed by similar market-based players like Airbnb.

In a paper led by Max Klein that was recently published and will be presented at the ACM Conference on Computer-supported Cooperative Work and Social Computing (CSCW) which will be held in Jersey City in early November 2018, we sought to provide a window into what this change means and what might be at stake. At the core of our research were a set of interviews we conducted with “dual-users” (i.e. users experienced on both Couchsurfing and Airbnb). Analyses of these interviews pointed to three major differences, which we explored quantitatively from public data on the two sites.

First, we found that users felt that hosting on Airbnb appears to require higher quality services than Couchsurfing. For example, we found that people who at some point only hosted on Couchsurfing often said that they did not host on Airbnb because they felt that their homes weren’t of sufficient quality. One participant explained that:

“I always wanted to host on Airbnb but I didn’t actually have a bedroom that I felt would be sufficient for guests who are paying for it.”

An another interviewee said:

“If I were to be paying for it, I’d expect a nice stay. This is why I never Airbnb-hosted before, because recently I couldn’t enable that [kind of hosting].”

We conducted a quantitative analysis of rates of Airbnb and Couchsurfing in different cities in the United States and found that median home prices are positively related to number of per capita Airbnb hosts and a negatively related to the number of Couchsurfing hosts. Our exploratory models predicted that for each $100,000 increase in median house price in a city, there will be about 43.4 more Airbnb hosts per 100,000 citizens, and 3.8 fewer hosts on Couchsurfing.

A second major theme we identified was that, while Couchsurfing emphasizes people, Airbnb places more emphasis on places. One of our participants explained:

“People who go on Airbnb, they are looking for a specific goal, a specific service, expecting the place is going to be clean […] the water isn’t leaking from the sink. I know people who do Couchsurfing even though they could definitely afford to use Airbnb every time they travel, because they want that human experience.”

In a follow-up quantitative analysis we conducted of the profile text from hosts on the two websites with a commonly-used system for text analysis called LIWC, we found that, compared to Couchsurfing, a lower proportion of words in Airbnb profiles were classified as being about people while a larger proportion of words were classified as being about places.

Finally, our research suggested that although hosts are the powerful parties in exchange on Couchsurfing, social power shifts from hosts to guests on Airbnb. Reflecting a much broader theme in our interviews, one of our participants expressed this concisely, saying:

“On Airbnb the host is trying to attract the guest, whereas on Couchsurfing, it works the other way round. It’s the guest that has to make an effort for the host to accept them.”

Previous research on Airbnb has shown that guests tend to give their hosts lower ratings than vice versa. Sociologists have suggested that this asymmetry in ratings will tend to reflect the direction of underlying social power balances.

We both replicated this finding from previous work and found that, as suggested in our interviews, the relationship is reversed on Couchsurfing. As shown in the figure above, we found Airbnb guests will typically give a less positive review to their host than vice-versa while in Couchsurfing guests will typically a more positive review to the host.

As Internet-based hospitality shifts from social systems to the market, we hope that our paper can point to some of what is changing and some of what is lost. For example, our first result suggests that less wealthy participants may be cut out by market-based platforms. Our second theme suggests a shift toward less human-focused modes of interaction brought on by increased “marketization.” We see the third theme as providing somewhat of a silver-lining in that shifting power toward guests was seen by some of our participants as a positive change in terms of safety and trust in that guests. Travelers in unfamiliar places often are often vulnerable and shifting power toward guests can be helpful.

Although our study is only of Couchsurfing and Airbnb, we believe that the shift away from social exchange and toward markets has broad implications across the sharing economy. We end our paper by speculating a little about the generalizability of our results. I have recently spoken at much more length about the underlying dynamics driving the shift we describe in my recent LibrePlanet keynote address.

More details are available in our paper which we have made available as a preprint on our website. The final version is behind a paywall in the ACM digital library.

---