Meta & Children's Online SafetyEmail: Daisy experience and thoughts
Raw text
Case 4:22-md-03047-YGR Document 2887-44 Filed 03/27/26 Page 1 of 3 AMENDED Exhibit 386 PLAINTIFFS' OMNIBUS OPPOSITION TO DEFENDANTS' MOTIONS FOR SUMMARY JUDGMENT Case No.: 4:22-md-03047-YGR MDL No. 3047 In Re: Social Media Adolescent Addiction/Personal Injury Products Liability Litigation Case 4:22-md-03047-YGR Document 2887-44 Filed 03/27/26 Page 2 of 3 Message From: Mark Zuckerberg [/O=THEFACEBOOK/OU=EXTERNAL (FYDIBOHF25SPDLT)/CN=RECIPIENTS/CN=7C603B7604634528AF59E0313BAD293E] Sent: 2/16/2020 10:33:32 PM To: Fidji Simo ; Adam Mosseri Subject: Daisy experience and thoughts I wanted to share some thoughts on my experience with Daisy in FB and IG so far. I'm also curious how much time you've spent in the tests and what your personal experience has been with it. At first I really didn't like it. My eyes kept scanning to see the like counts that weren't there so I just felt like I was getting less signal and it felt strictly worse. After a week or so I've acclimated in a way I didn't expect. Now that I don't expect to see like counts, I just don't pay attention to the UFI and focus more on the content itself. It feels good in a way -- although I'm still not sure the good outweighs the bad. My plan is to swap back out of the test in another week or so and see how I feel about the like counts once they're back. To be clear, this reaction has nothing to do with well-being or social comparison at all. I don't feel better or worse about myself, and I don't think it has had a meaningful effect on my engagement -- although it's quite possible that by not emphasizing likes I am less likely to like content myself and that may contribute to a compound negative engagement effect across the network. I get that some people, especially teens, may feel self-conscious about posting content that doesn't get enough likes. I'm not sure if the tests showed this was actually helpful at all for those use cases, but that's not really relevant to my own use. Instead, my experience has been more about the quantification of social interactions. There's something nice about not quantifying them. The app feels like people are just interacting rather than playing a game. Even though other people are still seeing like counts in their apps, the fact that I don't see them on other people's posts still makes me feel like they're not in a game. I can see why some people said they thought this was better for others -- although in more of a product experience sense and not in a well-being sense. Of course, if the net result is that this leads to less connection then that's a real substantive issue, but I was surprised by how it feels. It hasn't been as bad for navigating group discussions as I'd expected. I feel the same experience there around more natural interactions and less quantification. The experience around public content hasn't bothered me as much as I expected either, although it was certainly very jarring at first and still is on occasion when I could use the extra signal. But since my reaction is primarily about quantification, I wonder if we could make up some of the issues with the public content experience by showing a temperature meter or some other qualitative measure of the post's popularity rather than an order of magnitude count (which still feels quantified, just inaccurately and therefore unsatisfyingly). The experience of having to tap twice to see the people who liked a post on FB hasn't bothered me much either, but I don't do that frequently so I may not be experiencing the regular amount of pain others would. The most jarring part for me has been seeing random people's names in the UFI. On one hand, it's nice to be reminded there are other people out there interacting. There's something nicer about that than a number. But on the other hand, it makes the experience feel less personal to me and it has trained me not to look closer when I see a name because I no longer expect it to be a person I actually know. So there are pros and cons of this aspect, but overall I think it's a bit negative. Like I mentioned, my plan is to switch back in a week or so and see how I feel about the like counts. Intellectually, after talking to more of the management team (with both of you in Product and Brand) and quickly discussing this with the board in our private session, I think it's more important to do something similar HIGHLY CONFIDENTIAL (COMPETITOR) META3047MDL-004-00012022 Case 4:22-md-03047-YGR Document 2887-44 Filed 03/27/26 Page 3 of 3 across FB and IG than I did when we had our review. I also continue to think it would be very painful to roll this back completely, but I'm still unconvinced rolling it out completely across both would be positive. So I'm very interested in seeing the alternative steps you're thinking through now. Of course this means we're on a longer time frame than March. Extending the time frame to get this dialed correctly feels right to me, but I wanted to call it out since I know the IG team felt some urgency. In the realm of intermediate steps, one more thing we may want to consider is publishing the research now and trying to correct the misperception that this change has any impact on well-being. If this had the tenor of us simply evaluating a product change rather than addressing a social issue -- and, to be clear, the former is what we're doing -- then we might feel more flexibility in exploring different outcomes here. HIGHLY CONFIDENTIAL (COMPETITOR) META3047MDL-004-00012023