See that image above? When I posted it, I wrote some alt-text for it — i.e. a description of what's in the image.

Here's what I wrote:

A group of alphabetic dice spelling out the words "ALT TEXT". They are on a white surface and surrounded on all sides by perhaps two dozen similar dice that have random letters that spell nothing. Each die has a small hole going through the center; they're the type of dice you see used to make bracelets that spell out messages

This is new behavior for me — but one that has opened some interesting writerly doors.

Lemme unpack this …

For years, I rarely included alt-text when I posted images online.

I knew I was supposed to. There are many excellent things that alt-text does — the chief of which is that it makes images accessible to anyone who's sight-impaired. If there's alt-text, then their screen-reader software can tell them what's going on in the images. If not …

In the last year, though, I finally began to mend my ways. I now put alt-text in nearly every image when I'm blogging or posting on social media.

What changed? Well, it was mostly the effect of being on Mastodon.

I'd been using Mastodon on and off for years, mostly to hang out in various open-source software communities. But when Elon Musk took over Twitter last fall, I decided — for a whole host of reasons, too many to list here — that I'd be happier moving most of my social-networking over to Mastodon. I liked its vibe; I liked the cool things its decentralization made possible. It seemed full of nerds, techies, and artistic folks, a scene I enjoy. So I joined a friend's server, saturation.social, and set up my tent there, almost full-time.

The thing is, Mastodon has a really strong ethos of accessibility. I quickly noticed, spending a lot more time there, that the majority of folks added alt-text. If people forgot to include alt-text, other folks would (usually very politely) remind them to include it. If an image that didn't have alt-text was going a little bit viral (things rarely go very viral on Mastodon, since it doesn't have a rich-get-richer attention-juking algorithm), a Mastodon user would often re-post it with a good alt-text description.

Influenced by this culture, and by the lovely examples of people's everyday behavior, I started getting much more attentive about adding alt-text to my images.

That behavior quickly spread beyond Mastodon, too! I'd recently launched a newsletter — my "Linkfest" collections of the best stuff I encounter online (subscribe here folks!) — and I made sure to put alt-text on images there. The same goes with images on my Medium posts; I'd been neglecting to put in alt-text for years, but now I try to do it regularly.

As I've leaned into alt-texting, I've discovered it's a super interesting literary practice.

There's the old saying that "a picture is worth a thousand words". But here the goal is to reverse that proposition: To take a picture and distill its essence into language.

How exactly does one do this — and do it well?

Accessibility literature generally tells us to write alt-text as pithily and straightforwardly as possible. Adding a bit of style is good, too: I've spoken to people who use screen-readers, and most often they'll tell me they're looking for a sort of just-the-facts description, with enough humanity in it that it doesn't sound robotic. The blind Mastodon user Christy described the goal quite neatly …

The way I've always described to them people is, imagine you're talking to someone on the phone with just voice, no video. How would you explain the image so it made sense to them and got the point across? If there is text in the image, putting that text in the description is probably important, either a summary or exact depending on what the image is trying to convey.

There can be something quite striking about even a fairly straightforward description. I recently saw a Mastodon post of a picture of a tree — you can see it here, it's lovely — with this alt-text …

A horse chestnut in full leaf on a sunny late summer day, although there are a few leaves on the ground to suggest autumn. The trunk is broad and with twisted bark, as mature horse chestnut trunks usually are. Halfway up, a growth of four comparatively new branches with smooth bark looks a little spider-ish.

I love this style — the mix of clear description, studded with bits of natural knowledge ("as mature horse chestnuts usually are"), and little lexical flourishes ("a little spider-ish").

The more I wrote alt-text, the more I realized there was something meditative and poetic about it. I was no longer merely looking at images; I was noticing them, examining their features. Trying to give a nuanced description of an image has the effect of gently alienating you from the image — and I mean that in the positive, useful sense of "alienating". You have to look at it with fresh eyes.

For example, last month I wrote a blog post about nearly being hit by an SUV while cycling. As the main image, I picked this one …

And at first I just wrote a simple description, something like "a cyclist going down the street." Then I added more details, including that the cyclist is in the center of the picture, and behind them are a bus stop and a brightly lit store, and that the cyclist is moving quickly.

But as I was writing that last clause — "the cyclist is moving quickly" — I realized something curious about the composition of the photo: The cyclist is crisp, while the background is blurred.

That's not an easy effect for the photographer to achieve! And it's precisely what gives the image its power.

So I wound up writing the alt-text thusly …

A cyclist going rapidly down a city street. The photo is taken from the side, and the cyclist is in the center of the image, heading towards our right. In behind the cyclist we see a bus stop and a brightly lit store. Interestingly, despite the fact that the cyclist is moving quickly, they are unblurred and crisp in the photo — while the background is blurred. The ultimate effect is curious: It's as if the store were moving quickly, while the cyclist was standing still

One could critique this alt-text for being too damn long. Fair enough! And there's something a bit narcissistic about me focusing on the internal experience I have while puzzling over how best to describe this image.

Nonetheless, I regard this as a very cool moment in online culture: The act of stopping to carefully scrutinize an image, to notice things that are worth noticing, and put them into words. It is also a curiously still moment. I'm buzzing along, reading and replying to posts on Mastodon, flitting here and there, zipping from tab to tab. Then I want to post an image and suddenly, boom: I have to stop, pause, focus on a picture, and find the right words.

Many other people have noticed this: Alt-text captions can be oddly literary.

As the Mastodon user Gambolloch notes …

While alt text image descriptions aren't poems, writing them has felt like practice for writing poetry. I wasn't expecting that.

And as Francois Heinderyckx, another denizen of Mastodon, writes …

I am floored and fascinated by the near-literary quality of many of the descriptions of images that people carefully craft when appending pictures to their posts on Mastodon. Some of them will reveal a detail that had escaped my attention. Others will help me understand a subtle joke I had missed. Others still are true poems. ALT are their own microblogging world that reveals itself as you hover on a picture. And I hover, and hover.

I endorse both these sentiments! (And BTW I found both those comments on a terrific site run by Stefan Bohacek called the "Alt Text Hall of Fame". On the site, he highlights different ways people write their alt-text — from the just-the-facts style to ones that are filled with wit and humor.)

So, in summation: If you aren't currently writing alt-text for your images? Consider doing it! You may discover the writerly joys of this activity. And you'll make the Internet a better place.

As the Mastodon user StormyNight wrote, not long after joining the social network this summer …

(Enjoyed this one? Then grab that trackpad — or mouse — and hie thee to the "clap" button! It's good for up to 50 claps, per reader.)

I publish on Medium two times a week; follow me here to get each post in your email.

I'm also a contributing writer for the New York Times Magazine and Wired, and author of "Coders". Follow me on Mastodon or Instagram, or sign up for my pay-what-you-want newsletter "The Linkfest": It's "the opposite of doomscrolling".