Why I'll never start making videos

Sat, 04 Nov 2023 12:57:04 +0100

Jupiter Rowland

jupiter_rowland@hub.netzgemeinde.eu

Thinking about making videos, dropping it because of accessibility requirements; CW: long (2,940 characters)

Artikel ansehen

Fediverse

show all 4 comments

Sun, 05 Nov 2023 14:52:17 +0100

Scott M. Stolz

scott@biztechtonics.net

Instead of doing the pausing stuff, you could instead write a transcript that appears in the description of the video.

Another option is to provide the same information in another form, such as a blog post.

You post the video on PeerTube, provide a basic description in the video and the text description, and then link to a blog post that describes it in detail.

And on the blog post, you embed the PeerTube video and have the detailed description underneath.

Personally, I think that a 40,000 word description is overkill, and I wonder how many people with impaired vision would actually want to hear all of that. They may wind up doing their equivalent of TL;DR (Too Long; Didn't Listen?).

Think about it this way. Before there was TV, there were radio shows, where everything was described using voices and sound. Or if you want a more modern example, think of audio books. Since the audience cannot see anything, the entire narrative needs to be described to them using words.

But, audio books and radio programs don't tell you every single detail; they only tell you the important ones.

This is where the two tier approach comes in. The video or picture that you post has a brief description with the most important details, and then you link to a blog post that goes into greater detail, if they want greater detail. Some will, and some won't. But you gave them the option. This way you satisfy those who want a short description and satisfy those who want a long description.

Mon, 06 Nov 2023 11:15:48 +0100

Jupiter Rowland

jupiter_rowland@hub.netzgemeinde.eu

@Scott M. Stolz

Instead of doing the pausing stuff, you could instead write a transcript that appears in the description of the video.

Another option is to provide the same information in another form, such as a blog post.

You post the video on PeerTube, provide a basic description in the video and the text description, and then link to a blog post that describes it in detail.

And on the blog post, you embed the PeerTube video and have the detailed description underneath.

Neither of which is actually accessible.

Both the WCAG and the Mastodon community demand that content and description always have to be delivered together and as close to one another as possible. The description must not be separate from the content. This is also why both giving an image description in a comment and giving an image description in a separate article doesn't count as accessible.

For videos specifically, the WAI does not mention a fully separate written description as an option, only a real-time audio description in some shape or form that automatically plays at the same time as the video. Worse yet: A blind Mastodon user has once told me that she could not access an image description of mine in a Hubzilla article with her screen reader. This means I can't rely on Hubzilla articles for accessibility.

The Mastodon community is trying hard to push all Fediverse content to WCAG level AA (see here for a quick reference). And as over 300 completely off-topic Mastodon users from all over the instances make up the vast majority of my connections, not counting those Mastodon users who see my posts on their federated timelines, I can't simply ignore what the Mastodon community says just because I'm on Hubzilla and not on Mastodon.

Now, "community" doesn't mean instance admins and moderators. It means regular users. Users of the kind that see posts with pictures without alt-text on their federated timeline and reply to them by asking the posters to add alt-text. And that alt-text better be actually descriptive and informative.

The only alternative to a separate audio description edited in in "post-prod", actually what's recommended for new videos (see here), would be to describe the video during the making. My videos wouldn't be scripted, so this would mean that I'd constantly be rambling and describing and explaining what appears within the camera's field of vision as I walk through the place where I'm doing the video. This, however, would make a shorter version with less description impossible unless I do the same video in the same setting twice, making one of them appear much less spontaneous that way.

Personally, I think that a 40,000 word description is overkill, and I wonder how many people with impaired vision would actually want to hear all of that. They may wind up doing their equivalent of TL;DR (Too Long; Didn't Listen?).

Think about it this way. Before there was TV, there were radio shows, where everything was described using voices and sound. Or if you want a more modern example, think of audio books. Since the audience cannot see anything, the entire narrative needs to be described to them using words.

But, audio books and radio programs don't tell you every single detail; they only tell you the important ones.

These can hardly be compared.

Audio books are basically spoken novels. Novels are concise in describing the setting, but only if they take place in our real world. The more fictional the setting becomes, the more verbose its description becomes. Once you go full fantasy, once you completely leave the real world, descriptions can become very extensive. And they have to. Fantasy literature can't get away with mere name-dropping. People have to imagine what that fantasy world looks like, but they can't if they don't know how because nobody tells them how.

Imagine Tolkien only name-dropping Hobbiton without describing what it looks like, what hobbit houses look like etc. Exactly this is why Tolkien's works are so wordy.

As for radio documentations, they're always from the real world, so the same description requirements apply to them as to novels and radio plays that take place in the real world.

My pictures aren't from the real world. They're from obscure 3-D virtual worlds and from settings there that may or may not be inspired by the real world, but that aren't the real world. They tend to be chock-full of stuff that doesn't exist anywhere else. Thus, I can't get away with mentioning the stuff that's there.

I write my descriptions for two target audiences.

One, blind or visually impaired users. If I only mention something without describing it, a blind or visually impaired user will ask, "Yeah, but what does it look like?"

Two, all kinds of users who love to read detailed image descriptions because detailed image descriptions help them understand pictures or elements of pictures which they don't understand just from looking at them (or using technical aids such as alt-text plus screen reader if they can't actually see the pictures). Again, my pictures are from 3-D virtual worlds which hardly anyone knows. I can't expect everyone to be immediately familiar with the setting and everything in the picture just from looking at the picture. Instead, I have to expect them not even knowing what they're looking at unless I tell them.

To make my point clear: This is the post with the 40,000+-character image description. Had I only provided the short "descriptions" in the alt-texts, that would have raised more questions than it would have answered.

The latter group of users would have asked, "What is this place? Where is it? What are those boxes? What's that stuff that is in the boxes?"

The former group of users would have asked, "What does that place look like? What do the shelves look like? What do the boxes look like?" And if I had just mentioned that there's something written on the boxes while describing what they look like, they would have asked, "What IS written on the boxes? Please transcribe.".

There's only one way for me to reduce how much I have to describe. That's by reducing how much there is in the picture in the first place. This wasn't possible here.

This is where the two tier approach comes in. The video or picture that you post has a brief description with the most important details, and then you link to a blog post that goes into greater detail, if they want greater detail. Some will, and some won't. But you gave them the option. This way you satisfy those who want a short description and satisfy those who want a long description.

If anything, I'd have to offer the same content, the video in this case, in multiple versions. One for sighted users with no description. One for impatient non-sighted users with just as much description as necessary so that they don't all come complaining. One for curious non-sighted users who really want to understand the video with a full-blown detailed description edited in.

Mon, 06 Nov 2023 20:38:01 +0100

Scott M. Stolz

scott@biztechtonics.net

I think you misunderstand.

You write a suitable accessible description in the alt-text like you are supposed to. But you write what is IN the picture, not what each item is. Then you write the long description in a manner that ALL people can read.

To give you an example.

You have a picture of a spaceship that is sitting in the crater of the moon.

The alt-text should say is it is "a picture of a spaceship that is sitting in the crater of the moon." Maybe give the name of the ship if the ship has a name, and some interesting physical characteristics about it. It should be short and sweet. You should NOT describe every single minor detail that most people do not even care about.

Then, as part of the text that is visible to ALL people, you tell us about the ship, some history about it, what features its has, what is significant about the text on the side of the spaceship, etc. This text that is visible to all people is supplemental and can be displayed in the same post or in a separate article.

You put way too much detail in your descriptions. Most people with perfect eyesight won't even notice half of the stuff you describe in your alt-text. If people who can see won't notice the detail, a blind person probably wouldn't care either.

Besides, the recommendation is that text that is beneficial to both sighted and blind users be displayed to everyone.

Since Mastodon does not allow long descriptions, then the workaround is a short alt-text and a link to the longer description. Until Mastodon changes how they do things, then your only option is a workaround.

That being said, you can make PeerTube compliant via posting a longer description for the video. So at least PeerTube would be compliant, even if Mastodon is not.

Mon, 06 Nov 2023 20:43:27 +0100

Scott M. Stolz

scott@biztechtonics.net

And speaking of PeerTube, you can bypass the 500 character restriction by posting your description on PeerTube. Anyone on Mastodon should get the whole description, I would assume, if it works like Hubzilla. If that is the case, you just have Mastodon people follow your PeerTube account.