Justin Ekis

6 Followers
247 Following
2.2K Posts
Digital Accessibility consultant specializing in website audits. Proud Memphian. Christian. Sports fan. Apple enthusiast. #NeverTrump. Blind, autistic, and ADHD. Prolific booster. he/him
Alrighty folks. Hopefully all of my followers migrate properly. See y'all on the other side. In a few minutes, I'll be @JEkis. Mindly is indeed pretty cool, but I heard about this place and it just feels right.

Disney is updating its Disability Access program because they say it is being abused. Their solution? You must get prior approval by using Disney's virtual video chat system to speak with a member of the accessibility team.

Many autistics can’t use video chat because it causes serious anxiety. You would think their “accessibility team” would know that.

Nope.

#ActuallyAutistic @actuallyautistic

https://www.disabilityscoop.com/2024/04/10/disney-revamping-disability-access-policies-at-theme-parks/30825/

Disney Revamping Disability Access Policies At Theme Parks

Disney is overhauling its disability access policies after seeing an explosion in abuse of a program meant to accommodate visitors with autism and other developmental disabilities.

Disability Scoop

I think there's some commentary on the relationship between #infosec and the whole "why do my eyes hurt" google search memes

It's easy to laugh at people who clearly looked at the sun without eye protection

HOWEVER, there were a lot of fake eclipse glasses for sale before the eclipse. My wife bought two boxes of glasses (we gave a lot away), upon testing, one box was fake so they went in the trash. Had we used these, our eyes would also hurt

This is how we tend to treat people in #infosec. Users are so dumb, they click links, and install malware, and buy gift cards ... SO STUPID

But it's not that simple. A person was mislead and got hurt. Then we laugh at them. For fake internet points

ACB is pleased to see the final rule on web and mobile accessibility from the U.S. Department of Justice. Read the joint statement from ACB, AFB, NDRN, and NFB: https://www.globenewswire.com/news-release/2024/04/10/2860753/0/en/Joint-Statement-Applauding-Final-Rule-for-Web-Accessibility-Under-Title-II-of-the-ADA.html
Joint Statement Applauding Final Rule for Web Accessibility Under Title II of the ADA

Statement from ACB, AFB, NDRN, and NFB applauding the DOJ on release of the final rule for web accessibility under Title II of the ADA....

American Foundation for the Blind
[Forum Topic] Introducing Anytime podcast app and seeking VoiceOver feedback https://www.applevis.com/forum/ios-ipados/introducing-anytime-podcast-app-seeking-voiceover-feedback
Introducing Anytime podcast app and seeking VoiceOver feedback | AppleVis

Hi everyone, I have developed a free and open source podcast app, called Anytime Podcast Player, which is available for iOS and Android. I've been working on improving accessibility within the app and particularly the screen reader support, but I know there is more improvements to be made and I was recommended this forum - thank you Robert. I would be very grateful for any feedback, tips or suggestions that can help me improve the accessibility in Anytime and make it a better app if anyone would like to try it.

I decided I am moving to a different server. My top candidates are neurodifferent.me and mindly.social. Unfortunately, neither of those make their lists of moderated servers public. Can someone on those instances tell me whether they federate with threads.net?
@squish Hello there. I'm looking for a new instance because mine is blocking threads. I'm going somewhere that will allow me to follow Joe Biden and Barack Obama. Does Does neurodifferent.me federate with threads?

IRC: probably works fine with clients and servers from 20+ years ago

Discord: needs to install 15 updates because I haven't launched it since yesterday

Oh the idiocy!

Excerpt from a #Mashable article linked below:

Let's say someone owns the domain name "NetfliTwitter.com." Why would they own that domain name? Because if X is automatically changing anything that includes "Twitter.com" to "X.com," then that means posting "NetfliTwitter.com" on X would make it appear in posts as "Netflix.com," the popular movie streaming service. And if a user clicked the linked "Netflix.com" text that appears in that post, it would really take them to "NetfliTwitter.com." Because while X is changing the text that the user wrote, the URL it links and directs to remains the same as the user posted.

Source: https://mashable.com/article/twitter-dot-com-posts-change-to-x-dot-com-ios

X automatically changed 'Twitter' to 'X' in users' posts, breaking legit URLs

A scammer's dream and a disaster waiting to happen.

Mashable

Lots of things happening in the AI/LLM space that could have implications for #accessibility

Ferret-UI from Apple:
https://arxiv.org/abs/2404.05719

ScreenAI from Google
https://research.google/blog/screenai-a-visual-language-model-for-ui-and-visually-situated-language-understanding/

#a11y #ai #LLM

Ferret-UI: Grounded Mobile UI Understanding with Multimodal LLMs

Recent advancements in multimodal large language models (MLLMs) have been noteworthy, yet, these general-domain MLLMs often fall short in their ability to comprehend and interact effectively with user interface (UI) screens. In this paper, we present Ferret-UI, a new MLLM tailored for enhanced understanding of mobile UI screens, equipped with referring, grounding, and reasoning capabilities. Given that UI screens typically exhibit a more elongated aspect ratio and contain smaller objects of interest (e.g., icons, texts) than natural images, we incorporate "any resolution" on top of Ferret to magnify details and leverage enhanced visual features. Specifically, each screen is divided into 2 sub-images based on the original aspect ratio (i.e., horizontal division for portrait screens and vertical division for landscape screens). Both sub-images are encoded separately before being sent to LLMs. We meticulously gather training samples from an extensive range of elementary UI tasks, such as icon recognition, find text, and widget listing. These samples are formatted for instruction-following with region annotations to facilitate precise referring and grounding. To augment the model's reasoning ability, we further compile a dataset for advanced tasks, including detailed description, perception/interaction conversations, and function inference. After training on the curated datasets, Ferret-UI exhibits outstanding comprehension of UI screens and the capability to execute open-ended instructions. For model evaluation, we establish a comprehensive benchmark encompassing all the aforementioned tasks. Ferret-UI excels not only beyond most open-source UI MLLMs, but also surpasses GPT-4V on all the elementary UI tasks.

arXiv.org