The Interledger Community 🌱: Yotam Liel

Yotam Liel — ILP Summit Reflections

Yotam Liel — Tue, 20 Dec 2022 15:06:20 +0000

Introduction

Hi! My name is Yotam Liel, and I am a research scientist from Tel Aviv University. It was a pleasure to have the opportunity to attend the ILP summit and present the work Lior Zalmanson and I have been working on. It was also amazing to have the chance to meet the team and connect with other members of the community.

Key highlights about your participation at the Summit

One of the Summit's highlights for me was the opportunity to hear from the Interledger team about their vision for the protocol and how it is being used in the real world. It was particularly interesting to learn from Stefan Thomas's presentation about future use cases, such as marketplaces for training data.

Another highlight of the summit was the opportunity to network with other members of the Interledger community. I had the chance to participate in interesting discussions and breakout sessions, which gave me valuable feedback about our research and ideas for future work.

Key takeaways from the Summit

For me, it was very interesting to learn about the many applications of the Interledger Protocol and how it is used to solve a wide range of real-world problems, from enabling cross-border payments to new monetization models for content creators in a variety of verticals.

Overall, my experience at the ILP Summit was incredibly valuable, and I feel grateful to have had the opportunity to participate. I had a lot of fun, and I am excited to continue learning more about ILP and to be a part of the community as it continues to grow and evolve.

Web Monetization encourages users’ to make value-driven content consumption choices (experiments results)

Yotam Liel — Tue, 01 Nov 2022 10:27:21 +0000

The results of the first half of our research agenda have provided us with positive initial signs about the potential effects of the Web Monetization model on users’ mindful and socially conscious web browsing (see here, and here). Building on these results, we have designed and developed a setting to experimentally test the relationships between web monetization models, peoples’ values, and online content consumption choices. Our experiments demonstrate the effects of users’ awareness of online content monetization across multiple contexts, scenarios, and factors. We found that when users are aware that their online browsing provides monetary compensation to content creators, they prefer to consume content from non-profit organizations or independent creators and activists over that of mainstream media outlets.

Background and Theory

One of the most prominent characteristics of the web monetization model is, in our view, the fact that it increases users’ awareness of the monetization of their online browsing - for example, think of the “Coil is paying” badge. Our studies have, therefore, focused on examining the effects of increased awareness of online content monetization on users’ consumption choices. Our conceptual theoretical framework builds on theories from social and cognitive psychology, which have recognized people’s desire to maintain consistency with their values and behaviors (i.e., self-consistency (Lecky, 1945), self-perception (Bem, 1972), and cognitive dissonance (Festinger, 1957) theories). Our hypothesis, generally stated, is that increasing users’ awareness of the monetary support their online browsing provides to content creators will drive them to make content consumption choices that align with their values and beliefs. Our studies also contribute to the important ongoing debate on whether online activism is a complementary or rather a substitute for meaningful and impactful in-person activism (i.e., activism vs. “clicktivism” or “slacktivism”; Cabrera et al., 2017; Freelon et al., 2020; Gladwell, 2010; Morozov, 2011). We suggest that the Web Monetization protocol may be a solution that puts an end to this dilemma and be a welcome addition to any activist’s digital toolbox. Enabling creators to leverage their quality content to raise funds to promote their causes seamlessly and without using advertisement; and allowing users to contribute their time, clicks, and funds to creators who rally around specific causes. We assert that this could be closely related to the foot-in-the-door effect (Burger, 1999; Freedman & Fraser, 1966); a consequence of the desire to maintain self-consistency by which individuals are significantly more likely to comply with a substantial request after first agreeing to a minor request. We suggest that when browsing their favorite sites, Web Monetization / Coil users could be seen as already agreeing to comply with a small initial request, as their time on the site translates into monetary compensation to the creators. Based on the foot-in-the-door effect, we hypothesize that this makes them more willing to give out additional support to the creators or the causes being promoted. Furthermore, this effect is expected to amplify when users consume content by creators who are activists and promote social causes as they are driven by the motivation to avoid the cognitive dissonance between their existing beliefs and self-image and actions that may conflict with them - disagreeing to participate or provide additional support.

Experiments Design and Results

Our experiments were set to examine how increased awareness of online content monetization affects users’ consumption choices, and especially their preference between different types of creators (i.e., mainstream media outlets vs. non-profit organizations vs. independent bloggers and activists) and with relation to their own values, beliefs, and personal connections. Specifically, study one examined the effects of monetization awareness on preference for co-partisan sources (i.e., sources that are aligned with the viewer’s own political preferences) and for non-profit organizations vs. mainstream media, and study two examined the effects of monetization on the preference for independent video creators and activists (“vloggers”) vs. mainstream media.
Both studies included two groups: a treatment group for which we included questions designed to increase participants' awareness of the monetization of content on the YouTube platform; and a control group in which participants answered similar questions minus the ones intended to increase monetization awareness. We measured the effects of the treatment by observing participants’ choice of video to watch from a simulated YouTube search results page, and given a specific scenario (e.g., to learn more about “gun regulation” (study one; see Figure 1) or about “disability rights” (study two)). We also measured peoples’ willingness to provide additional support by asking whether or not they would be interested in learning how they can support a non-profit organization (relevant to the topic of the video they chose to watch) (see “methodology” section below for further details on the experiments’ design).

Study 1: politically polarizing context - preference for mainstream media vs. non-profit organizations

Overall, the results of study one show that increasing users’ awareness of the monetization of online content indeed affects their consumption choices. We first examined the effects in relation to users’ political preferences. We found that users’ choice of co-partisan sources was almost identical across the two conditions (78.1% in the control group and 77.9% in the treatment; see Figure 2a). A reasonable explanation for the lack of effect in this measure is that the initial rate of co-partisan preference is so high that what we observed here is a ceiling effect. Next, we looked at the effect of monetization awareness on the choice between mainstream media outlets and non-profit organizations. As shown in Figure 2b, the results clearly show that the percentage of people who chose to watch a video from a non-profit organization doubled when users’ awareness of monetization was raised (from 24.8% in the control group to 48.4% in the treatment group). This difference is also highly statistically significant (p < .001). Overall, we learn that monetization awareness drives people to support organizations that match their political stance. Thus supporting our hypothesis that increasing users’ awareness of the monetary support their online browsing provides to content creators, will encourage them to make content consumption choices aligned with their values and beliefs.
Additionally, we examine the effects of monetization awareness and content consumption choices on users’ willingness to provide additional support. The results here were mixed. While there was no direct effect of the treatment on participants’ compliance with the request for additional support (28.6% of positive responses in the control group and 29.5% in the treatment group), we did find that those who chose to watch a video from one of the non-profit organizations were also more likely to agree with the request for additional support (37.5%) as compared with those who watched one of the mainstream media videos (24.2%). This difference was statistically significant (p < .05), providing some support to our hypothesis on the self-consistency effect.

Study 2: neutral context - preference for mainstream media vs. independent creators / activist

The second study was intended to build on the results of the first and extend them. We were interested in studying these questions in a neutral, non-political context and with additional types of online content creators. The experimental design was similar to the one used in study one. The main difference was the use of a different scenario and a different set of videos in the YouTube search results simulation. For the neutral context, we used the topic of “disability rights and living with disability”. Instead of comparing the choice between mainstream media and non-profit organizations, we had participants choose from a pool of mainstream media videos and videos created by independent vloggers (video bloggers) and activists. Additionally, we had participants answer a short open-text question explaining their choice of video.
The results show again how increasing people’s awareness of monetization changes their online content consumption choices. Preference for the videos by independent vloggers and activists increased from 45.9% in the control condition to 60.6% in the treatment condition (p < .05). Furthermore, this effect was also reflected in participants' explanations for their choices. For example: “I think she [the vlogger] will benefit more from me watching her video, rather than ABC.”; “I chose the video that interested me the most and I also chose the video from a smaller creator because I would like for them to get the money from the ad, if possible.”; and “Initially I chose CNN but figured the individual channel needs my views”.
With regards to the effects on users’ willingness for additional support, the results were again inconclusive. We observed a slight (not statistically significant) negative effect for the treatment, and an increase in compliance (also not significant) among the people who chose to watch the independent creators' videos.

Methodology

Study 1

Participants were recruited from Amazon Mechanical Turk and told that the objective of the task was a survey about online media usage and preferences. They were randomly assigned to either a treatment (“High Awareness”) or a control group. The survey began with questions regarding participants' usage and familiarity with various online news and media platforms. Importantly, one of the questions was designed to increase participants’ awareness of the monetization of content on YouTube. This question was presented only to those in the treatment group. In contrast, participants in the control group answered a similar question that did not deal with monetization.
Next, participants were directed to a page that simulates a search results page on YouTube for the query “gun regulation” (see Figure 1). The page included four relevant videos whose sources were either right-leaning (Fox News, Gun Owners of America) or left-leaning (NBC News, March For Our Lives). The videos were also differentiated based on whether they were from a mainstream media outlet or a non-profit organization. Participants were instructed to imagine they were interested in learning more about gun regulation. Then, they were asked to choose the video they would like to watch and instructed that they would be asked about the content of the video. Participants’ choice of video to watch is our primary measure of the effect of monetization awareness on content consumption. Participants then watched the video they chose and answered questions regarding its content.
Next, we presented each participant with information regarding one of the two non-profit organizations - Gun Owners of America for those who chose to watch one of the right-leaning videos and March For Our Lives for those who watched the left-leaning videos. Participants were asked if they would like to learn how they could help support that organization. Their answers to this question allowed us to measure the effect on their willingness to provide additional support.

Study 2

The design of study two was very similar to that of study one. It differed mainly in the content of the simulated YouTube search page. In study two, we presented a page that simulates a search for “disability rights and living with disability”. We, again, presented four videos as the results of the search. However, instead of mainstream media and non-profit organizations, we had mainstream media and independent creators and activists (“vloggers”). The mainstream media outlets included PBS News, CNN, The Wall Street Journal, and ABC News (a random selection of two for each participant). The vloggers included videos by Annie Elainey, Sitting Pretty Lolo, Zack Collie, and Freddo the Wheelchair Guy (YouTube channel names; again, two of the four were randomly selected to be presented). Another difference from study one was the inclusion of the open-ended question “Why did you choose this video” directly following the participants’ choice of video watch. Lastly, the non-profit organization used in the request for additional support was the Disability Rights Fund.

References

Bem, D. J. (1972). Self-Perception Theory. In Advances in Experimental Social Psychology (Vol. 6, pp. 1–62). Elsevier.
Burger, J. M. (1999). The Foot-in-the-Door Compliance Procedure: A Multiple-Process Analysis and Review. Personality and Social Psychology Review, 3(4), 303–325.
Cabrera, N. L., Matias, C. E., & Montoya, R. (2017). Activism or slacktivism? The potential and pitfalls of social media in contemporary student activism. Journal of Diversity in Higher Education, 10(4), 400–415.
Festinger, L. (1957). A theory of cognitive dissonance (Vol. 2). Stanford university press.
Freedman, J. L., & Fraser, S. C. (1966). Compliance without pressure: The foot-in-the-door technique. Journal of Personality and Social Psychology, 4(2), 195–202.
Freelon, D., Marwick, A., & Kreiss, D. (2020). False equivalencies: Online activism from left to right. Science, 369(6508), 1197–1201.
Gladwell, M. (2010, September 27). Small Change. The New Yorker. http://www.newyorker.com/magazine/2010/10/04/small-change-malcolm-gladwell
Lecky, P. (1945). Self-consistency; a theory of personality. Island Press.
Morozov, E. (2011). The net delusion: The dark side of Internet freedom. PublicAffairs.

Survey results: Part two - Ideological considerations in content consumption

Yotam Liel — Mon, 08 Aug 2022 19:04:00 +0000

After reporting our findings on people’s preferences and views on online content monetization models in the first part of our report, we now turn to discuss the topic of value-related considerations in content consumption choices.

Do people make ideological and value-related considerations in their online content consumption choices?

First, we examined to what extent people are mindful of the values and beliefs of the content creators whose content they consume and of the sites they browse. We asked participants to rate their agreement with the statements: "I consume content by creators who share my own values and beliefs", "I choose to go to sites that reflect my values and beliefs", and "I do not go to sites that support values that I disagree with". The results show that people indeed pay attention to the alignment between their values and beliefs and the values promoted by the sites they go to and the creators whose creations they read or watch. Above 75% of the participants reported that they choose to consume content by creators who share their own values and on websites that reflect those values. Additionally, approximately 50% of the participants agreed that they avoid visiting sites that support values they disagree with. Interestingly, those who politically identify as centrists were less likely to agree with these statements than those identifying as liberals or conservatives.
To study this question from another angle, we also measured how people’s political preferences affect their likelihood of watching videos from certain creators. The results here strengthen our previous findings. As shown in Figure 1, when asked if they would watch a video about climate change published by the DNC (Democratic National Committee) official YouTube channel, democrats replied positively and republicans negatively. However, as expected, when the topic was gun regulation and the publisher the NRA (National Rifle Association), republicans responded more favorably and democrats negatively.

Are people looking for content by people from historically marginalized groups?

Next, we wanted to learn if people are sensitive to the identity of content creators and whether they belong to a historically marginalized group. To study this, we measured their rate of agreement with the statement: "It is important to me to read articles and watch videos created by people from historically marginalized groups (groups excluded due to race, gender identity, sexual orientation, age, physical ability, language, and/or immigration status)".
A majority (51%) of the people agreed with the statement, while only 18% disagreed. However, we saw significant differences between people as a factor of their political preferences and identity. Specifically, we found that those who identify as republicans were significantly less likely to agree with the statement as compared with democrats and centrists (see Figure 2). Additionally, people from the underrepresented groups sample agreed with the statement significantly more than people from the general population sample (for more details about the samples see the first part of this report).

Do people see online content consumption as a way of showing support to causes and creators?

Most people in our survey see online content consumption as a way of supporting creators and causes they care about. Over 75% of the participants agreed with the statement: "By reading articles and watching videos, I show my support to the creators and the causes they promote". Figure 3 shows this was true regardless of people’s political preferences.

Overall our survey results are encouraging and provide initial support to our hypotheses regarding the possible link between the web monetization model and people’s choices of online content. We found that people are, in fact, mindful of the values and beliefs of the creators and outlets of the content they consume. We also learned that people see online content consumption as a means to support causes and creators they care about. These could be great news for activists and content creators overall, and especially for those belonging to underrepresented groups.

Survey results: People's views, preferences, and familiarity with online content monetization (part one)

Yotam Liel — Wed, 27 Jul 2022 12:30:00 +0000

How do people view online monetization, and how open are they to accepting the web-monetization model? Do people consider their “shared values” with the content creators and ideological and value-related topics when making online content consumption decision processes? Is there a difference in the consumption habits and preferences of people from underrepresented groups compared to the general population? In this post, we share fresh from the oven results from our first survey that aimed to answer these questions.

TL;DR (because we know you are busy :))

We surveyed 300 people about their online media consumption, preferences for monetization models, and value-based and ideological considerations in online content consumption. We found that people think content creators deserve a more significant part of the income than they estimate that they receive. But also that they tend to underestimate the amount the creators are receiving. We also found that most people like monetization models by which their subscription fees are divided between the creators whose content they consume. Interestingly, people from underrepresented groups preferred a model by which funds are equally divided between the creators rather than a distribution based on the time spent on each item. However, we also found that when it comes to choosing between paying for content or seeing ads, people, perhaps unsurprisingly, generally prefer the latter.

Who were the participants in our survey?

We drafted participants for our survey from the crowd-source platform Amazon Mechanical Turk. Participants were all US residents and were compensated for their time filling out the survey. We drafted the participants in two waives, each including 150 people. In the first wave, we aimed to recruit a sample representative of the general population. We will refer to it as the “general” sample. In the second wave, we recruited a sample of people from underrepresented groups (based on race and ethnicity). We will refer to it as the “underrepresented” sample. Table 1 and Figure 1 below provide information on the demographics in each sample.

Political preferences

In addition to their demographic information, we were interested in learning about our participants’ political preferences. From the results shown in Figure 2, we can learn that, on average, both samples lean toward the liberal end of the political space. Comparing the two samples, it appears that the underrepresented groups sample is less conservative and more centrist (and not so much more liberal). Additionally, people from the underrepresented groups were much less likely to identify themself with the more extremist options of “very liberal” or “very conservative”.

Why do we need a sample from underrepresented groups?

Comparing people from underrepresented groups to the general population is important since we are interested in studying the potential effectiveness of web monetization for activists who are coming from, and working with, people from traditionally underrepresented groups.

People’s views on online content monetization

In this section of the study, we examined people’s knowledge of online content monetization and their preference for various monetization models.

How much of the income from advertisements goes to the creator?

We first asked participants to estimate what percent of YouTube’s income from showing ads goes to the video creators (“estimation”). Immediately afterward, we asked what they think this percentage should be (“opinion”). As it stands today, 55% of the income from ads on YouTube goes to the creator (conditioned on the creator being in the YouTube Partner program). As shown in Figure 3, people in both samples significantly under-estimated this amount. The average estimations were 35% in the general sample and 37% in the underrepresented groups sample. Another commonality between the samples is the opinion that creators deserve a larger share of the income (in comparison to their own estimations). However, the average percentage that people in the underrepresented groups thought creators should receive (55%) was significantly lower in contrast to the opinion of people in the general sample (63%).

Which model do people prefer for the distribution of funds between content creators?

Next, we wished to learn about peoples’ preferences regarding how the subscription fees they pay are divided between the content creators. Again, we stayed within the context of YouTube and asked the participants to imagine that they have a Premium subscription to the platform. We then told them that out of the total $11.99 that YouTube charges, approximately $7 goes to the content creators, and asked them to choose between the following four models for the distribution of the $7 between the creators on the platform: (1) An even distribution between all of the content creators on the website, (2) Only the creators whose videos you watch are paid; the money is evenly distributed between them, (3) Only the creators whose videos you watch are paid; based on the time you spent on each video, and (4) You choose who gets how much at the end of each month. Notably, the third model was meant to represent the web-monetization/Coil model.
The results show that overall the “Coil compatible” model is supported in both samples (see Figure 4). It is the most popular model in the general sample and the second most popular in the underrepresented groups sample. Interestingly, people in the underrepresented groups sample show greater support for the models in which the money is distributed evenly between the creators. This difference might reflect a more significant appreciation for equality and a dislike of the potentially discriminating models.

Do people prefer to see ads or pay for subscriptions?

An essential aspect of the success of the web-monetization project is people’s willingness to pay for online content. To learn more about this, we asked people to choose between three levels of subscription-fee - ads ratios across four types of online content platforms. Specifically, they could choose between (1) Free content and many ads, (2) Low subscription fee and some ads, and (3) Higher subscription fee and no ads; for each of the content types: (1) Video streaming (Netflix, Hulu, etc.), (2) Online news (foxnews.com, nytimes.com, etc.), (3) Video platforms (YouTube, TikTok, etc.), and (4) Social networks (Facebook, Instagram, Twitter, etc.).
Figure 5 shows that people dislike paying and are willing to see ads in exchange for content. However, it is important to note that seeing ads in exchange for content (and more) is what people are generally used to. Moreover, if we look at the differences between the content types, we find that people’s preferences generally reflect the current status quo - a state where the model for video streaming is a subscription fee and no ads, and social networks are free with many ads. Nonetheless, it is interesting to learn that a non-negligible number of people (21% in the underrepresented groups sample) are willing to pay to see fewer ads (or none at all) on social networks. Furthermore, people from the underrepresented groups sample were also generally more positive towards the option of a hybrid model of a low subscription fee and some ads. To sum, we suggest that these results highlight that any attempt to shift people from the models they got used to (i.e., from free ad-based models to subscription fees and vice versa) might require time and re-education of users. This seems to be especially true for the shift from free content and ads models to those requiring paying for subscriptions.

Do people make ideological and value-related considerations in their online content consumption choices?

The short answer is “yes, they do''. We also found some differences between the two types of samples, and as a factor of people’s political preferences. We will tell you all about it in the second part of this report which will be published soon.
Stay tuned.