Why YSK: If you want to make Lemmy a more accessible place for all, you need to know how to add labels (aka: "alt text") to embedded images
As many of you are already aware, you can embed images in comments using the following markdown: ![](https://example.com/image.jpg).
That works, but the image will be left unlabeled which leaves screenreader users out of the conversation. This may seem like a theoretical problem, but it's not -- Lemmy already has an active blind community: https://rblind.com/
So, here's what you can do to add labels when embedding images:
![Earth viewed from the Moon](https://example.com/image.jpg)
That's it! Any text within the [] becomes "alt text", which is what screenreaders use for describing images. It's a small thing, but it makes a world of difference.
Remember: Reddit took something from many of us here. For some, it was a mobile app. For others, it was the ability to operate their own communities. Comment with solidarity; use alt-text.
![A soldier in the film Starship Troopers speaks directly to the viewer: "I'm doing my part!"](https://i.kym-cdn.com/photos/images/newsfeed/002/182/171/eb0.jpg)
That was me, and my bad. As a blind person myself, I’ve never tried to post an image. I knew including alt text was possible, but I didn’t realize the method was undocumented, and Lemmy doesn’t prompt for it. If I had, I would have offered help, not just snark.
No worries! I never really considered blind people browsing apps, I only became aware of the technical details since the API changes at reddit, so it's still far from a reflex to think about.
YES! But please don't spoil. I am on the 4th book, Rythm of War, for many months now.. I have read 50% but it just goes so slowly... The first and second book, The Way of Kings and Words of Radiance, are my favorite books ever. Wit is one of my favourite characters, just so quick and witty and mysterious...
Lemmy should do it similar to how I have seen this done on Mastodon. When you upload an image an overlay shows up to remind the user to describe the image before submitting. The typical user is not going to remember to do this.
You can click the little image button when commenting or posting, but it uploads it to the instance you're on, which costs them space, so it's probably limited in most instances
it should actually be a description of the image, for blind people.
also, web users who aren't using some accessibility tool can't even see it without right-clicking and inspecting the element or clicking lemmy's "view source" button (as i just did to quote your comment here). so, please don't use alt text for things other than descriptions of images!
Excellent question. For image-based posts, there is currently no mechanism to directly embed alt-text. This is likely to change in the future, however, so subscribe to lemmy#1086 for updates.
For now, the general advice is to use the post body to describe the linked image (on Lemmy, all posts are allowed to have "self-post"-like text). This would work by linking the image directly as usual and also embedding the same image in the post text with alt-text included there.
There are no descriptions for image posts. You can however make a post that is both an image and text post at the same time, and throw the description there, until a proper way to put alt text gets implemented.
I'm guessing it is in settings? Also a jerboa user. But I am seeing the comment images at a reasonable size. I'm using the dark theme, and I think I changed the font size, but it's been almost a month since I first installed, so not super sure.
Indeed, but just FYI: doing this is actually even worse than leaving a completely empty alt-text!
Why? Because some browsers have experimental support for AI-generated alt-text. If an image has no alt-text, it'll try to generate one, but only if the alt text is empty. This feature will probably become more common in the future, so please try to refrain from putting non-descriptive stuff in the alt-text area.
You shouldn't do that. That's not gonna help screen readers and in fact will just waste their time trying to read off a URL because it thinks it's describing the image. The alt text is intended to be human readable.