Rebecca · @WordyAnchorite
70 followers · 906 posts · Server deacon.social
Gregory J. Rosmaita · @oedipusnj
376 followers · 272 posts · Server kolektiva.social

@AeroDerHasen This clip quite succinctly captures my feeling towards Mastodon as a fellow blind refugee from Twitter.

Video: The final 14 seconds of the film "Casablanca"
youtu.be/JMOBxjIvLBE

Description: On a heavily fog-covered airport runway, two figures are seen walking away from the camera, into the obscuring fog. The figure on the left wears the uniform of a Vichy prefect of the police, cap set jauntily at an angle. The figure on the right wears a light colored trench-coat & a dark fedora. As the music swells, the scene fades to black.

#describedvideo #BlindMasto

Last updated 2 years ago

Gregory J. Rosmaita · @oedipusnj
376 followers · 272 posts · Server kolektiva.social

@Burstingfoam

2. Best practices for creating accessible hashtags

2a) CamelCase Hashtags

The first & most important rule for creating accessible hashtags is always use CamelCase. That means, capitalize the first letter of each individual word that comprises the hashtag. This will make the hashtag much much more comprehensible to anyone using speech or refreshable braille. It also makes it easier to visually distinguish the components of a hashtag, an important consideration not only for those using high levels of magnification to interact with their device, but one which improves overall readability & reduces eye- & cognitive strain. If you do not capitalize the first letter of each word, the hashtag will be delivered to the AT user as a single—usually incomprehensible—undifferentiated string.

Here are a few examples of a hashtag using CamelCase :



2b) is it OK to use hashtags for individual words or terms in the body of a post?

Yes, as I have explained, most blind/VI users who use AT will have no problem with sporadic hashtags sprinkled into the body of a post. they also help us find pertinent information that we might otherwise miss.

2c) To group or not to group?

Grouping of hashtags is also fine. However, as a matter of personal preference, I prefer to encounter them at the very end of a post, rather than a grouping being the first thing I encounter in a post. This is the model I myself use, when associating multiple hashtags with a post.

A lot of accessibility is just variations on—& very specific use cases of—basic usability. If you can't access an object you can't use it; likewise, if you can't use an object, it cannot be called accessible.

I hope this answers your questions.

#TheArtOfALT #BlindMasto #BlindMastodon

Last updated 2 years ago

Gregory J. Rosmaita · @oedipusnj
271 followers · 107 posts · Server kolektiva.social

@emptywheel

2. Describing images

As a person who chose which images to include in their post, you are the best person to describe that image.

First of all, automatic image recognition is still at a very primitive stage. Most automatic image recognition of purely visual images, such as a group photograph, yield only very generic information. You, on the other hand, chose those pictures because of something intrinsic in those images that eloquently communicates a message to the viewer. It is that implicit message you were sending by using an image that you must capture to fully describing image.

Of course, image description begins with the basic “what” of the image. Identifying information, contextualizing information, etc. The same information you would need related to you over the phone if someone was describing the image to you. But, in order for me to build up a detailed and useful conception of the image—whether it is a meme, or a piece of recently captured photojournalism—I not only need to know the basic building blocks of the image—that is, the "what"—but also the “why” of the image.

Image description is as much of an art as it is a reasonable accommodation. You will grow more comfortable describing images the more you describe them. You will also find that describing images is a new outlet for your creativity and a new exercise in concise and succinct writing.

As with everything, if this is not sufficient, or if it merely causes more confusion, please ask for clarification in public, for I know for certain I am not the only person who benefits immeasurably from the inclusion of ALT on this, and every, platform.

#TheArtOfALT #imagedescription #BlindMasto

Last updated 2 years ago

Gregory J. Rosmaita · @oedipusnj
271 followers · 107 posts · Server kolektiva.social

@emptywheel

1. Images of text - Imperative

Images of text are useless to a screen reader or refreshable braille display. Screenshots, highlighted excerpts from a long article or legal document, etc: ALL images of text must be accompanied by a transcription in text. You can easily obtain a scanned text version of an image of text using Google Lens, no matter what device, app or operating system you’re using.

Note, OCR is not 100% accurate. Once you extract the text & paste it into the ALT field, please take a few seconds to review it to make sure it's not only a one to one reproduction, but that it makes sense when read linearly, from top to bottom, for that is how a screen reader or refreshable braille display renders the contents of the description to the blind/VI individual: as a single string text. It can only be listen to as a string of spoken words. With speech, you can’t go back & review the description word by word or sentence by sentence as you normally would do with text content. If you interrupt speech, you must listen again from the beginning through to the end to obtain the entire description. This is an essential concept to keep in mind

A few additional notes on images of text & equivalents. If you visually highlighted part of the image of text, that highlighting will be lost in OCR. One strategy is to reproduce in the ALT only the text you highlight, since that is what you were trying to direct the viewer to; everything else that happens to be in that screenshot is immaterial to understanding the intent of the author in adding that screenshot.

For highly formatted text, or in the case of a graph for chart, the rules change, because you're not trying to convey the chart or the graph textually—what you need to do is present the findings rendered graphically in the graph or chart in a manner that will make sense when read linearly, from beginning to end. Don't describe the graph or chart, explain the data.

#TheArtOfALT #imagedescription #BlindMasto

Last updated 2 years ago