Every month, you click through thousands of pictures, perhaps even more. You keep up with friends, shop for important or beautiful things, gather information from snazzy infographics. It makes sense. The internet is visual, pages on pages spreading before you in the interconnected, brightly colored smorgasbord of content inherent to its very nature. And it should be. Pictures are gorgeous, and our class text makes very clear how important it is that content catches the eye. But for some people, a picture alone doesn’t work, no matter how fancy the page. Some people need words. Thankfully, the internet has ways to make that happen. But content creators aren’t using them.
When people use words to describe a picture, its called alt, (alternative,) text. Not only does it help Google sort your images into better search results, it also makes images accessible to people who use screen readers, software that speaks everything on the screen and makes devices accessible without sight. More information on this technology can be found here. Without alt texts, screen readers might announce the word, “Image,” or read the file name such as “078GUU.jpeg, according to this thorough article. Such extraneous words when trying to look through a page are at best unhelpful and at worst cluttering as a totally blind user trying to scroll through a web page or a social media feed. But with good alt text, the pictures become relevant, clear and often interesting. Some platforms are beginning to attempt to bridge this gap with AI, but as this page so clearly illustrates, it frankly does a terribly inadequate job, often spouting things like, “May be an image of cup.” Thankfully, humans do much better, though results are still varied.
Good descriptions differ a lot in length and focus, perfectly natural when there are so many types of images. They might be brief. “Model wearing a red floral dress with a sweetheart neckline and short sleeves,” for example, to accompany a listing in a shop that might have a wider description below. It might be long, “Two toddler boys, one with red hair and the other with blond, demolish a chocolate cake. The blonde-haired boy has frosting all over his face,” for something like a social media post or perhaps a description in a picture book. The above article suggests making sure you’re brief, no more than 125 characters, and clear in detail. Think about it like you’re trying to describe the image over the phone, and don’t include too much extraneous content. I’ve seen those same standards reflected in other research and real-world examples, but it’s very flexible, depending on the needs and preferences of an author. Its also an incredibly interesting art, because you get to decide what’s relevant, and that will very much depend on what kind of context or message your visual carries. Take the example of the two boys above. If they’re wearing yellow shirts, one is white, one is black, and they’re in a kitchen with blue walls, how much of that matters? All of it? None? If it’s a housing advertisement, perhaps the kitchen needs some love. Skin color is generally important these days, so an author might want to add that regardless. But so much around what gets highlighted has to do with purpose, so there’s no one right answer. This might make some people feel daunted, which might mean it gets neglected altogether, but that can’t continue to stand. Leaving images unlabeled is shutting a whole group out of the beauty, uniqueness and wisdom the Internet has to offer.
As of February 2021, a New York Times article states that 60.6% of home pages have no alt text, and a study in 2019 of 1.09 million tweets with images found that only 0.1 percent included it, as well. Please, help put an end to babbled file names and unhelpful descriptions. Here is a full guide on how to add alt text to some of the top social media sites. And if you want even more advice on writing descriptions, you can check out Alt Text is Poetry, a really unique project with a great mission to humanize the process while making it better for all involved. Shine a little bit of light into the darkened jumble at the heart of the net, one described picture at a time. Your visually impaired viewers will thank you.
MDST 485 Class Blog 