Quick question! I'd like to add ID to my posts but dont know how much information is too much? Is there a minimum that you would recommend?
Purpose of image descriptions
Oh, is this my cue to write an extremely comprehensive guide on how to write useful IDs??
(thank you anon for your question!)
I think a good rule of thumb is to think about what you'd want to know if you couldn't see the image, and just describe that.
I'll use the images from this guide (very imperfect written by a sighted person, as most are) as an example, pointing out the good and bad.
I like its idea of describing images as object-action-context. If you want to read through the article, I really like its way of explaining different ways to describe a photo of a dog, depending on its purpose.
Essentially, when you're describing an image, you want to figure out the purpose of an image, and explain that.
I'll use my icon as another example:
[Image description: Mangacap of Envy edited over the bi pride flag. They're looking off to the side. End ID]
Notice how I didn't describe the bi flag itself, or go into detail about Envy's position (they're curled up, they have a worried facial expression, etc). If I were making a post explaining the bi pride flag, I would explain the colors of the different stripes. And if I were describing an art piece of Envy, I'd go into more detail about their pose or their facial expression. But since this is an icon, the main thing you want to know is that it's a tumblr icon of a character over a pride flag.
Something that the above guide did badly:
This is the ID that the author used:
Description: Graphic titled “Protesting safely” at the top in lavender text on dark purple background. It includes 3 sections: “What to wear,” “What to bring,” and “Don’t bring” all with graphic illustrations. In the light purple “What to wear” section, it says, “nondescript, solid color, layered clothing; cover identifying tattoos, goggles & mask, emergency contacts written down, heat resistant gloves, and tie your hair up.” In the medium purple “What to bring” section, it says, “water for drinking & tear gas, snacks, cash/change & ID, washcloth, bandages & first aid supplies, ear plugs, and protest signs. In the light red “Don’t bring” section, it says, “cell phone without first turning off Face/Touch ID, going on airplane mode, and disabling data, jewelry, anything you don’t want to be arrested with, and contact lenses.”
Now, if you're ADHD, you probably saw that big block of text and said, "I don't want to read that." Even with the extra bolding I added.
Another thing to remember is that IDs are for sighted people, too. Maybe someone can't read the font on the graphic and need it in plain text form. Maybe someone has bad wifi and the image won't load (that's me. If an image isn't loading and I see the ID, I can still get something out of it. And if I know what it is, like a detailed art piece, I might decide to wait for it to load so I can see it. But if there isn't an ID there, I wouldn't know.)
I see a lot of people writing IDs that are just a big block of text. Those aren't accessible. Screen readers take forever to read them. And non-screen readers likely wouldn't bother.
Here's my take on IDing that graphic:
[Image description: Graphic titled "protesting safely."
What to wear: Goggles; mask; nondescript, solid colored, layered clothing; emergency contacts written down; hair tied up
What to bring: Water for drinking and tear gas, cash/change, ID, washcloth, bandages, first aid supplies, earplugs, protesting signs
Don't bring: Contact lenses; jewelry; cell phone without first turning off face/touch ID, going on airplane mode, and disabling data; anything you don't want to be arrested with
Notice what I cut out. With graphics, no one cares about the text color or background color, so I cut all of those out. I separated the information into different lines as a way to mimic the organization of the graphic without going into tons of detail about how it's formatted. The example ID also used "it says" over and over, but I cut that out.
I can only think of one example where you would want to know the text color or background color, and that's if an artist used it for a certain purpose.
I normally wouldn't use my IDs from earlier in the year as an example because I wasn't very good at them, but here's an example. In this piece of art, the black background was used to elicit emotion - I can't read the artist's mind, but at the beginning it shows Ed's loneliness and at the end it hits you in the face.
(also side note: why doesn't that piece have more notes?? Y'all go reblog it with the ID right now)
Since this is a fandom blog, I want to go over how to do comics too. I know the one I used was technically a comic, but again I wasn't great at them at the time.
It may be tempting to describe every single facial expression and everything happening in the background when describing a comic. Especially when the artist has done a great job with expressions and details! But that leaves us with huge chunks of text like:
X says "this", smiling. Y crosses their arms. The background has a couch, a clock, and a cat. Z jumps up and down, eyes closed, laughing. X stands on a pink chair, on their tiptoes, saying "this" while waving their left arm and swatting Z with their right arm blah blah blah
I know that's a terrible example, I just don't want to pull a real one, lol.
Remember our favorite meme format? Incorrect quotes?
Format your comic like that.
Here's an example. Notice how the actions of the characters are built into the dialogue. Instead of:
Edward's phone rings. The caller ID says "Dad." Hohenheim has an incredulous expression on his face. Edward grins, and picks up the phone. He says, "Hello Colonel Roy Bastard!"
Actions are built in, either with *asterisks* or next next to the person's name ("Hohenheim, seeing the screen:")
Here's an example of putting comics into that format. It doesn't always work perfectly ("Roy opens the door. He can hear voices" is a break from that format) but it's much easier for everyone to read.
Why? Because the point of an ID is to describe what the creator of the image wants to get across. Even if their backgrounds and poses and expressions are phenomenal, the punchline is in the dialogue. So that's what you want to focus on in your ID.
The best way to get better at ids is to practice, and to look at other people's IDs to pick out what's good and what isn't. The main thing to keep in mind is you want to describe what people need to know, not what catches your eye.
Writing this guide has been overdue for a bit, and I'm glad I got it done! I'd appreciate it if people reblogged this so it could help more people write better IDs.
