#captions meta

Hello!

Tags I have seen for this include “screenreader unfriendly” (or “screen reader unfriendly”) and/or possibly “eyestrain” if applicable. And then if you copy the text in an accessible format (no decorations and default font), this is called plaintext and could be tagged as such.

If anyone else knows how these are tagged, please comment below.

while i can't find any sources on "industry standard" descriptions being ten words or fewer specifically (i did see fifteen from one source, though,) i do see that alt text should generally be around 200 characters at the most, as alt text should be concise and give just enough detail to understand the image clearly while not including anything that isn't vital to that understanding. some screenreaders aren't compatible with alt text that goes over that limit.

however, a lot of what you do on this blog is outright transcriptions, which isn't quite the same. aside from that, what most people (including you) who write descriptions on here are writing are long descriptions, which are intended for describing images in more depth than alt text and aren't meant to be concise in the same way. i think part of the problem is that people on this website tend to use "image description" synonymously with "alt text," implying that alt text and long descriptions should work the same way and aren't made with distinct (though related) goals in mind.

on websites outside of tumblr, long descriptions are usually used for images containing lots of vital information like charts and graphs, in conjunction with a short description in alt text—the alt says something like "a chart showing (general data)," and the long description provides the specific information, sometimes linked on another page or put in a collapsible menu so that the user can choose to access it. on tumblr however, especially when you're not the original poster of something, i don't know what a good equivalent to this practice would be. you could always warn at the beginning of a description that it is meant to be a detailed/long description or a transcription of lots of text.

image description guidelines are also not the same as guidelines for describing audio or video. this isn't something i do as much as image description, so i don't have any specific knowledge about it, but i'd keep that in mind too.

i don't mean to come off as contrarian to that anon (because i think their criticism is justified and i'm open to correction if there's something about their perspective i'm missing,) or condescending towards you (as i'm sure you probably know some of this already, being a caption blog,) i just thought i'd provide some context i don't see often that might help you figure out what you want to do with your captions ^_^b (happy thumbs-up emoticon).

Hi!

In short, captions benefit anyone who cannot understand or see images or coloured/formatted text on their own. Many visually impaired people, for example, use screen readers, and without extra information, screen readers will just say “image” when an image is present. There are also people with intellectual or learning disabilities who need help processing images or who can't read fancy fonts or colourful text, and these are just a few examples.

I also always add that I like adding captions because it normalises them and promotes accessibility online.

I've gone into some more depth on this in this ask, and I also encourage you to check out some of the links in my pinned post for more on why and how people write captions (also called image descriptions or IDs for short).

The stock screen reader on my Android reads a lone hashtag as "number" but will read it as "hashtag" when it is attached to a word. But that's just one example, so I'm not sure more broadly what is preferred.

Hey there!

My personal approach to videos depends on the type of video, but I will usually try to break it into “scenes” and give each scene a paragraph. If there’s dialogue, I’ll work that in with the description of the scene so that it hopefully flows like a narrative. If there’s a lot of text on the video itself (a lot of educational videos have this, for example), I work that in with what is being said and what makes sense. Some videos require more detail, and some you can just say things like, “And then they write the formula on the board.”

As for the actual process, I take videos in chunks of 4-5 seconds and then watch the whole thing a few times to make sure the description matches the video. If there’s a lot of narration, I might start by transcribing that first and then weave it in as I describe the actual video.

There are definitely other ways of doing it, though, so if anyone wants to drop tips in the replies/reblogs, please do!

Also going to promote the People’s Accessibility Discord server again if you are a Discord person - you can request descriptions and/or get help with your own there! Just for clarification, I don’t run this server; I’m just in it. Here is the link: link.

Hi there! That’s a great question and one that actually comes up frequently in the accessibility Discord server I’m in, so shoutout to them! The server is linked in this post if anyone’s interested.

From what I understand, the built-in captioning feature on Tumblr is buggy at best and downright unusable at worst. Sometimes screen readers don’t pick up on the captions written in those images, for example, and the feature seems to behave differently on mobile than it does on desktop. Adding alt-text (which is a different feature) sometimes works but has a pretty small character limit of 200 unless you edit the HTML directly, in which case you would have to do it on desktop since mobile doesn’t have that option anymore.

Overall, I generally encourage people to add ID’s under media for the above reasons and because there are people who benefit from captions but don’t use screen readers. This includes people with certain learning disabilities and people with sensory processing disabilities, among others (more on that in this ask and the beginning of this post). And as a last little note, adding ID’s where people can see them helps normalise them and makes it easier for people to find them if they’re scrolling the notes.