liz writes stuff down
22Sep/160

Triggering videos, thoughtful content warnings, and responsible feature release policies

Content warning: police murder of black people

We need to talk about potentially triggering videos and social media.

Social media is nearly unavoidable. There's a lot of upsides to using it, such as keeping up with family and friends you can't see frequently and reaching out to your customer base in more personalized ways, but it's also evolving very, very fast. One day we're posting our latest relationship update; the next we're getting news as soon as it breaks. It empowers marginalized voices to tell their stories - stories that often go untold.

One second we were uploading photographs; the next gifs and videos - videos of our cats, then videos of police shooting black people. Maybe people are posting them to draw attention to issues the mainsteam media doesn't see fit to focus on. Maybe police shootings are being posted as videos because people don't believe black victims didn't pose a threat. I don't personally share these videos, so I can only guess at the motivations.

If we listen to black voices, we hear them pleading to keep videos depicting police shootings of minorities out of their view.

Black lives matter. Properly respecting black lives involves understanding that videos depicting a death of a human can be very triggering and respecting viewers by allowing them the chance to opt out of seeing them.

We need to use thoughtful content/trigger warnings.

Content warnings allow people to engage with content how and when they feel comfortable doing so. They are not censorship.

Placing content warnings for videos is a lot less straightforward than placing content warnings for written pieces without even photographs like this one. A written content warning before a video in a social media post is a good start. They work well for posts that link to articles with videos in them when those videos don't sneak into the previews social media like to embed.

The reader is able to see this tweet without having to see a potentially triggering graphic video.

We need tools that help us place content warnings specifically for video formats.

If someone comes across an article or social media post with a video directly in it, text-based content warnings aren't enough.

Preview frames might include a triggering image. We can replace the default frame selected from a video with one that contains only the text of relevant content warnings. We need tools that make this process easy to do. Preferably, social media would include these tools in their upload interfaces themselves. Their inclusion would provide not just ease of use but also a convenient reminder to use them.

Also, the video may start playing anyway - maybe it autoplayed, maybe the reader accidentally clicked on it. Tools to supply text-only frames containing the relevant content warnings for the first ten seconds of a video would give someone the opportunity to pause or scroll past it before triggering content is forced upon them. This would provide a buffer even when video autoplay is turned on.

We need to release new features responsibly.

When Facebook and Twitter released their video autoplay features, they enabled it by default. Their decisions forced users to unnecessarily engage with content they didn't desire to see.[1] We should not turn on new features like video autoplay by default.

Facebook does allow users to disable autoplay for videos, but that still doesn't make it okay to turn it on by default. Twitter allows you to turn off autoplay for some videos[2], but you're stuck with autoplay while you scroll through a Moment, regardless of your video autoplay settings. It is absolutely irrepsonsible of Twitter to force this feature on you in any situation.

Content warnings, tools, and responsible feature release policies will never be a complete solution, but maybe these tools and feature releases would allow for us to benefit from having triggering videos available without causing as much unnecessary harm.

[1] It's also a huge mobile data hog.
[2] This appears to now be in the "Data" section of settings on iOS, and I am told it is a similar process for Android.

5Nov/152

Hearts, stars, and trumpets: the things I “favorite” but don’t “like”

Twitter recently changed "favorites" to "likes".

We want to make Twitter easier and more rewarding to use, and we know that at times the star could be confusing, especially to newcomers. You might like a lot of things, but not everything can be your favorite.

It's true - not everything can be my favorite. For me, there are three categories for tweets I "favorited": tweets talking about something I like, tweets about something important, and tweets that make me want to offer the author support. I probably didn't favorite anywhere near all of the tweets that talked about something I liked, but I find there's only so many times I'm interested in actively clicking to like a gif of a cat rolling around.

The other two categories - tweets that are about something important and tweets that make me want to offer the author support - are not always things I categorize as things I "like". In fact, the ones that I think are most important to favorite are definitely not things I like, such as commentary on Sandra Bland's death or a friend struggling with her workplace's sexism.

By favoriting, I could help highlight something important, especially since favorites/likes are one of the "engagements" that influence Twitter's "While you were away" selections, and I could highlight anything I found important, not just the easier topics I found "like"able. There are a lot of important stories I want to tell that social media algorithms overlook because they are not "like"able, and in my circles, Twitter was one of the places that was less affected than average by this phenomenon. By favoriting, I could increment a counter. That counter makes it easier for the author to see that myself and 54 other people found their thoughts important and supported them. As the type of person who can internalize metrics better than scattered - though thoughtful and appreciated - notes, I can attest to the positive impact a high (or even just non-zero) favorite count can have. By favoriting, I could express support with a mere click. I enjoyed being able to provide that support when I didn't have the energy to do more than that or couldn't find the time to phrase my support correctly, and I felt a little bit better when friends did the same for me. These nuances are lost when I have to "heart" someone's comments instead of "star" them.

More granularity for our reactions can be helpful. Slack employs a very liberal "do whatever you want with an emoji" reaction, and thanks to Emily Price's creativity, my favorite channel - XOXO's #ladies - uses the trumpet emoji to express emotional support. It works well because we've agreed that "the emotional support trumpet" is sounding our support and because we also agree that we do not need to say anything more to show meaningful support. This is awesome beacuse it's much easier to give support when you don't have to turn often complex feelings into words first. I feel intrinsically empowered to give emotional support trumpets to others, and I feel warm and fuzzy when I receive emotional support trumpets, too.

Maybe adding something like the emotional support trumpet to Twitter's responses could replace the support usage of "favorite"; maybe adding another different thing could replace using "favorite" to highlight something important. But that's just covering how I used the less specific "favorite" where I won't use "like". With an audience as large as Twitter's, we might not even be able agree on a finite set of specific reactions. Whatever the next steps, I miss "favorite" a lot and feel like we've lost something important.