Do You Really Know Your GIFs?

Aaron Jiang
ACM CSCW
Published in
5 min readSep 26, 2018

This blog post summarizes a paper about how people communicate with animated GIFs that will be presented at the 21st ACM Conference on Computer-Supported Cooperative Work and Social Computing.

Let me start with a story of Jane.

The SMS notification on Jane’s phone chimes almost constantly. Throughout the day she keeps in touch with dozens of friends, sharing messages, images, and more recently, animated GIFs. Jane just got a job offer from a local company and wants to share how happy she is. She opens her messaging app, finds one of her friends, types “me right now.” She then brings up the built-in GIF keyboard, searches “happy,” and attaches this GIF:

Image credit: Giphy.

She hits the “Send” button, and her friend’s confused reply arrives almost immediately: “Brian and Stewie? 4:20 much? I didn’t know you smoked pot.”

Jane stares at her phone in horror, as she has just sent the same GIF to her mother.

(If you also don’t understand what this GIF is about, this might give you a hint.)

Animated GIFs have long been an important part of Internet culture. While during the 90s GIFs mainly existed as flashing “under construction” signs and spinning email icons on personal webpages, they are now used in very different ways that focus on their abilities to convey emotions. These silent, short, usually low-resolution clips now pervade online communication, and have become part of the standard functionality of many social media sites and instant messaging applications.

But do these moving pictures always get across the emotion the way you intended? The answer is no — people misinterpret animated GIFs. I misinterpret GIFs all the time, and I’m sure you weren’t sure about what a GIF meant at some point or another. To understand how miscommunications happen with GIFs, I, with Dr. Casey Fiesler and Dr. Jed Brubaker, interviewed 24 people across the US, and we heard many stories of miscommunication like Jane’s.

Our interview participants talked to us about the difficulty of using GIFs on platforms with little or no GIF support, which is understandable, but they also said platforms with GIF search engines or GIF keyboards weren’t ideal either. Participants told us that the results yielded from GIF search engines were usually low-quality, and that it would take them a long time to find the perfect GIF that they wanted.

There’s a shit load of GIFs, and like, you know, I’ll go through something like 200 GIFs, before I find like the perfect one.

Many people told us it was hard to find the perfect GIF, but what does it mean for a GIF to be “perfect”? What is essential for a GIF to get the right message across? We realized it was the complex context surrounding a GIF that could make or break the GIF.

Whether one has watched the TV show or the movie which a GIF comes from — the source material of a GIF — has a significant impact on their interpretation of that GIF. If Jane had watched that particular episode of Family Guy, she probably wouldn’t have used that GIF at all. But familiarity with the source material really goes both ways in communication: It’s not only about the sender; it’s also about the recipient. If the sender had watched that TV show but the recipient hadn’t, the sender would not be able to convey the point of the GIF. But if the recipient had watched that TV show but the sender hadn’t, the sender would have sent some unintentional messages, which could be good or bad.

At this point, it should be pretty obvious that the person one is sending the GIF to — the communication partner — also makes a difference on whether the communication would be successful. If the communication partner doesn’t have the same context as the sender does, whether it be the source material or some other context, miscommunication is likely to happen.

Participants also talked about using GIFs more on one platform, and less on another. In other words, platform culture is important. Participants generally used GIFs more on platforms that celebrated and promoted GIF use, or simply is easier to use GIFs, and less on platforms where they believed GIFs didn’t fit the platforms’ communication styles.

Regardless of where one uses GIFs, or whether one knows that GIF is alluding to some esoteric Family Guy joke, the prerequisite of using GIFs is that both communication partners know what GIFs are, how they are used, and why they are a legitimate form of communication — what we call the media norms. While you might think “of course I know what a GIF is,” participants did not think the same of their parents. They avoided sending GIFs to their parents because they speculated their parents would ask: What is that? Why are you sending me that? Who is that guy in the GIF? And what does that guy have to do with what we are talking about? In more extreme cases, participants believed hat their parents thought using GIFs were wrong:

They would view it as the continued destruction of the English language or something. Like they would see it as not a proper way to communicate. … [My mom] would probably tell me to stop. She sees it as an unintelligent way of communicating.

These findings provided us with insights into how to design technology for GIFs, a novel form of communication. When people use GIFs in their communication, they are not just communicating the emotion on the surface, but also the complex contexts behind those GIFs. A GIF contains many layers of meanings and contexts, and miscommunication would happen if any one of the layers misaligns between the communication partners. On the other hand, when all these layers align, GIFs can really take communication to another level. We suggest technology designers carefully think about how to support these different contexts, and make GIFs that we all love more enjoyable to use.

Citation:

Jialun “Aaron” Jiang, Casey Fiesler, and Jed R. Brubaker. 2018. “The Perfect One”: Understanding Communication Practices and Challenges with Animated GIFs. Proc. ACM Hum.-Comput. Interact. 2, CSCW, Article 80 (November 2018), 20 pages. https://doi.org/10.1145/3274349

If you have questions or comments about this study, email Aaron Jiang at aaron [dot] jiang [at] colorado [dot] edu.

--

--

Aaron Jiang
ACM CSCW

Computational social scientist. Gamer. Have some opinions about online content moderation. https://aaronjiang.me