A federal judge issues a temporary restraining order barring DOGE from accessing personally identifiable information from the Social Security Administration. storage.courtlistener.com/reca…
The British and German governments have updated their advice for travellers seeking to enter the U.S. with fresh warnings about the risk of arrest or detention. A French scientist was denied entry to the US, because of text messages on their phone containing a 'personal opinion' about the #Trump administration. Veronica Cardenas, former assistant chief counsel at the US Department of Homeland Security says #deportations are eroding public trust. #US #travel #immigration
Officials in Paris are expressing their dismay after a French scientist was denied entry to the US, because of text messages on their phone containing a 'per...
Tonight's weird tech question: Is there a DOM API/JavaScript hack of some sort I can use to track, save, and restore the screen reader's position in a relatively static HTML document? Say, for instance, you were reading a book in an HTML document with your screen reader, then closed the window. Now imagine that window was an app, and I wanted to make certain your position was restored when that app opened again. Is there an API I could hook into for that?
It's not quite focus, because that'd require tabindex. It probably isn't one of the text properties, right? Because you're not exactly in charted territory when you're arrowing through a paragraph not in a writable element like an input or textarea. Can I track that at all?
No. The closest thing to an automatic solution to that is probably tracking scroll position, and you may or may not be able to make that more granular by making the text quite big (I haven't tried). But even then, the best you're gonna be able to achieve is to track the closest element and put focus back there to restore the position, without character-level accuracy.
I know that Mozilla and NV Access have done some work to allow selection of text within the NVDA browse mode buffer to be communicated to the browser for on-page actions that require a selection. But: 1. That doesn't work across browsers; and 2. your use case seems targeted at reading, not selecting.
Thanks, that's what I was thinking. And just to check an assumption, setting the scroll position won't update the screen reader's position in the doc--I'd have to use focus shenanigans for that?
For context, this is my attempt at a document reader that saves/restores position when the document is closed/reopened. I don't think I need character accuracy, or even paragraph accuracy, if I can open books or longer documents to roughly where the reader closed out.
FWIW I'm not just being lazy and asking, I'm trying right now and it isn't working, which I suspected it wouldn't. It's also possible I'm using my web framework wrong or that something is behaving silly under Linux.
This is, as far as I know, the primary reason why QRead, Christopher Toth's book reading app for Windows, uses a rich edit and not a web view; the rich edit has a caret, so the application can save and restore its own cursor position. The screen reader's virtual cursor is invisible to the application.
You can sort of try and hack this by giving all your divs a negative tabindex and then calling .focus on them, we tried this a few years ago when thinking about QRead web, but the biggest problem is the inaccuracies. I'm okay with being a paragraph back in a book, but not a paragraph forward, and because of the nature of screen readers and the DOM we'd likely get the latter in a lot of cases.
There are some instances in which a webpage can move the scroll position without explicitly setting focus to the target element, and have the screen reader's reading position follow. But that can be less reliable, particularly if the target element is visually obscured, and setting focus is a more explicit/guaranteed way to do it.
Note that if you're setting focus to things like headings and paragraphs that aren't focusable by default, you'll need to dynamically inject a `tabindex="-1"` for the best results.
Perfect, thanks, thinking I'll just try grabbing the first visible element after the last-saved scroll position and do the tabindex dance to position the screen reader. Not character-perfect but good enough for my purposes.
Alizée Delpierre publica 'Servir a los ricos', una investigación sobre lo que pasa dentro de las casas de milmillonarios franceses y las "desiguales" relaciones que se establecen entre el servicio y sus empleadores
I’ve been working on a #freebsd from scratch blog post series and was going to skip right over Xorg (preferring #wayland and #sway)…and then I heard about #NSCDE - a reimplementation of #Solaris #CDE look-and-feel via #fvwm
Thanks to work by Christian Moerz, it was a snap. Literally `pkg install Xorg nscde` a change to my .xinitrc and I was up in a pastel bliss again. Run #interlisp on it is a beautiful joy.
I was so impressed AND I got to avoid abandonware security holes and bit rot— so I sent some librepay love. It’s a great project!
The Department of Education being closed is one of the dumbest things I've ever seen.
Most #maga supporters want it as #trump and #doge convinced them schools were federally funded, and they were going to overhaul and fix it.
They're too uneducated to know that it's already funded by states, and The Department of Education exists to fill gaps in resources, especially for #disabled and disadvantaged children.
The Alzheimer's Association has uploaded the video of the "Narrative Showcase: Dementia, Life Course, and Storytelling." Here is the link. You just need to register and it is free to watch. I hope you find it valuable.
#Blind users, users that rely heavily on #keyboardNavigation or anyone else with a preference on that matter (please indicate what applies in the comments):
Do you have a preference for or comments on the format of URLs? During user research, we have learned that URLs that are easy to handle are a good thing.
Hey, André here. I used to test accessibility functionality with Otto once. First of all, thank you again for caring about accessibility, it's not very frequent in the open-source community, unfortunately. Yes, URLs are extremely important at least for some of us (like me), and here GitHub keeps doing great job. However, I didn't really understand your concern, what was before and what are you planning to introduce? Thank you. Also, you always know where to find me, please write here or to ap (at) oire.me. I'm willing to help.
@menelion codeberg.org/USERNAME/REPOSITO… is the URL scheme for repositories, so everything codeberg themselves use directly after their domain can't be used as a username anymore.
Yeah, the idea is to avoid adding more and more reserved usernames for every route and simply using patterns that are easy to differentiate from user accounts. For example, user accounts could be prevented to start with an underscore and you could do URLs that start with an underscore for "special use".
We wondered if there was a recommendation or best practice that is still acceptable accessibility-wise.
@Ember I've never heard about any best practices pertaining to URL design, so I guess the only guidance here is common sense. I.e., URLs must be: A) Meaningful (ID in a URL is okay, like /menelion/my-fancy-repo/issues/123 is okay, of course); B) Back-traceable (following the same example, if I cut off /123 I go to issues, if I cut off /issues I go to my repo, if I cut off /my-fancy-repo I go to my profile, etc., etc.); C) Consistent (if everything "new" is /new, it's a bad idea to sprinkle some /create in several places, and vice versa); D) Secure (check that I as a user cannot do /someone/repositories/new, for example, just typing in the URL in the address bar, — it might seem stupid, but believe me, it happens sometimes because "everyone click links" — spoiler alert, no!); E) You either have reserved words, like "repositories" in my previous example (to show user's repos), or you alter the URL altogether: if my user ID is 12345, I have /menelion, /menelion/my-fancy-repo, but to show all of my repos, I have /users/12345/repositories, for example (in this case, no one can create a username "users" — also kind of a reserved word) or, as you suggested, /_users/12345/repositories, or /-/users/12345/repositories, — also makes sense. In this case, I would do a redirect from /-/users/12345 to /menelion for better readability. If needed, I can elaborate this as you wish, in a Forgejo issue, oral or video conference, email or Slack (but not Matrix, I couldn't unfortunately find anything accessible for it).
Panic is selling refurbished playdates now, I wonder if I could pay them to just fix the screen on mine. 🥺 I need to look into it later when I have time. #Playdate
The thing about blogs: If it looks modern with a short URL and has tons of cookie popups and self-promo, stay away. If it looks like a plain HTML page with some basic CSS, found at a long URL at the bottom of your search result, it's probably trustworthy and written by a passionate self-hosting human being.
Basically, the more it looks like motherfuckingwebsite.com, the higher the quality of the material you're about to read.
@menelion I think the point was about a disproportionate amount of time being spent on presentation and embellishments rather than the message. How it's a probable indicator of the author's priorities in their blogging and maybe craft.
@dside @menelion Short wasn't the correct word. I should have said "professional" or "memorable". I'm talking about those worthless companies that have their own "blogs", which are just advertisements in disguise. The real blogs will use URLs like blog.arandomcomputerthought.org or ijustwanttoselfhost.org or something fun like that. They're just people doing it because it's fun or because they want to share interesting things they learned.
youtube.com/watch?v=lt_quvb5bW… CHARLIE ANGUS ISSUES TRAVEL WARNING AGAINST UNITED STATES Canadian officials must stop tiptoeing around the rising fascist threat posed by the Trump regime. The arbitrary detention of travellers and kidnapping of citizens by ICE is unacceptable. These are not the actions of a country that respects the rule of law. I am urging Canadians to avoid travel to the US.
Canadian officials must stop tiptoeing around the rising fascist threat posed by the Trump regime.The arbitrary detention of travellers and kidnapping of cit...
Still not fixing the menu not showing in XFCE. You know, the menu when you right-clock on the icon in the taskbar. Nothing shows. Nothing 😟 Started after v2 released.
@edgren the upcoming Gajim release contains a fix for the task bar icon menu in connection with XFCE. Please note: Gajim 2.0 switched to GTK4, and GTK4 removed its task bar icon feature. We had to implement a custom solution, which is why there are changes compared to Gajim 1.9.
Folks you are absolutely sleeping on this series and you need to rectify that. Naomi Novik’s “Scholomance” series is like the platonic ideal of the “surprise, the villain is actually the structural effects of capitalism” story and you need to read it mastodon.social/@glyph/1141927…
Without looking it up, do you know what a "maw-mouth" is? Would that be a resonant reference for talking about AI training data?
(Don't look it up. If you don't know what it is, read A Deadly Education, The Last Graduate, and The Golden Enclaves.
I started the series last night on your original recommendation (when you posted about maw-mouths). And yes, I did get far enough into the first book to at least begin to learn what a maw-mouth is.
Now I understand what a maw-mouth really is. I'm not sure I could convince anyone else to stop using generative AI, though, simply because it now makes me think of this fictional monster, horrific though it is.
@matt the analogy I wanted to use was not so much that I want to argue others out of using it but to describe my own misgivings. Trying to find ethical ways to use a local LLM for a real project feels like finding an ethical spot to live inside an enclave
It used to be quite challenging to set the precedence of individual #CSS rules without a lot of headaches. (This encouraged the practice of authoring many small utility classes that do very little on their own.)
Twenty-six Social Security Administration offices across the country are expected to close this year, due to actions taken by Elon Musk’s Department of Government Efficiency.
Check out DrDonk/OC4VM: OpenCore for Virtual Machines (OC4VM) has been built to run macOS VMs primarily on Intel based Apple Macs. It may also be used on other PC hardware using VMware Workstation. It provides an OpenCore disk image that can be used to boot Intel based macOS using VMware Fusion and Workstation, and the open source QEMU program. Made by the same developer who makes the Unlocker: github.com/DrDonk/OC4VM/ (experimental support for some AMD systems) If you want OpenCore VMDK images for AMD, I suggest you check this forum post: forum.amd-osx.com/threads/mac-… It even has images for 15.4. All you need to get started, if you are resourceful, are found in those two links.
ah yeah, very good point for sure. I think it's mac-OS only unfortunately, I have a Macbook Pro here with M1 so might try this after work tonight, but in make.sh there's a `# Bail out if not macOS` line which checks if it's a Darwin platform or not. Looks like all the files and scripts are there to build the vmdk file and make.sh guides you through it interactively. The repo got updated yesterday so I think it's still a work in progress but hopeful there'll be a release there soon.
@bill yeah, if you run make.sh from a Mac-OS machine, it should build the disks folder for you with the various ones, hopefully. Will have to check it out since it replacing unlocker itself is also fairly huge.
Several big gaming companies have formed the Accessible Games Initiative to establish standardized tags that clearly label games’ accessibility features like narrated menus and subtitles.
The problem, aside from the product recalls, the fact you can die inside when it catches fire and the doors lock, the substandard programming and the fact they look like the entire genre of vaporwave shit itself in vehicle form, is the fact that at this point it is known the only people who want one are fascists and/or smug dicks with raging compensation issues. This is the truck you buy when you hate yourself and everyone else.
This is a repository of apps to be used with your F-Droid client. Applications in this repository are official binaries built by the original application developers, taken from their resp. repositories (mostly Github, GitLab, Codeberg).
Das F-Droid Repo von IzzyOnDroid kennzeichnet reproduzierbare Apps (Reproducible Builds) seit heute mit einem simplen, aber effektiven Schildsymbol, um sensible Nutzer der Reproduzierbarkeit und somit der Integrität ihrer Apps zu versichern.
@s3nnet BTW: Schade, das auf Eurer Seite nur "asoziale" Buttons proprietärer Netzwerke (Facebook, WhatsApp & Co) angezeigt werden – aber kein einziger für's Fediverse (Diaspora/Friendica, Mastodon, Pixelfed etc)…
Oops, stimmt ja. Das Geraffel muss ich noch rausnehmen, danke!. Allerdings gibt es Mastodon sehr wohl, hab nur noch nicht das "Icon Pack" für Publii gekauft (kostenpflichtiges Plugin). Werd ich aber nachholen, jetzt da ich mich zu Paypal hab hinreißen lassen
@s3nnet Nein, in-App gibt es das noch gar nicht. Aber in unserem Repo-Browser (Web) gibt es das auf App-Ebene schon seit August, siehe android.izzysoft.de/articles/n…
Und wenn Du auf App-Ebene jetzt schaust, gibt es da "neue Icons". Zum Beispiel hier, für Catima: 3 Builder haben bestätigt, einer davon war ein "unabhängiger" ("independent builder") – also einer, der nicht von unserem Team betrieben wird.
Spannende Neuigkeiten im IzzyOnDroid Repo: Jetzt haben wir reproduzierbare Builds, spezifische Unterstützung durch einige der beliebtesten F-Droid-Clients und mehr!
@RealMo Show me that newer version 😉 IoD ships the NoNet variant, which hasn't seen a new release for 2 years now (none since 4/2023, see e.g. github.com/PhilippC/keepass2an…). The new release you see there is a) a different variant and b) much too big for the IoD repo (50 MB, our limit is 30 MB per app).
Checks I have read the FAQ section, searched the open issues, and still think this is a new bug. Describe the bug you encountered: I recently improved your build system so that we could have also n...
"Retain users with convenience, honesty, and kindness." This is at the core of what we do at @thunderbird and it drives our features implementation. Sometimes we fail at that, we're aware of it, but we're human and we always try to do better.
In our inaugural State of the Thunder, Alessandro Castellani, Director of Desktop and Mobile apps, leads a discussion on the 2025 desktop roadmap with membe...
The four principles of WCAG plus two more related to real-life accessibility. Set in Neue Helvetica Heavy as black text for lighter colored shirts. There is a version in white text and also matching shirts using Atkinson Hyperlegible.
“Now more than ever, the U.S. needs a service like VOA to promote its values, counter disinformation and ensure that the true story of America is heard worldwide.” thehill.com/opinion/internatio…
I blogged about my first time back to #CSUNATC25 since 2018 in San Diego. #CSUNATC is my favorite #a11y conference and community! I hope to get back every year. I had an amazing time hanging out with everyone and meeting you all in person! linkedin.com/pulse/csun-assist…
@dennisl Thanks for the comprehensive article! One of these years I really have to go back to CSun! :) BTW, Linkedin tricked me again. There is something going on with their "sign in with your GOogle account" flow. It literally signs me in every time I go there. I then have to go back and cancel my account (I do not use Google Login). I really have to report it one day! This feels like a scam!
I wrote an article about why Linux distros should use compiler hardening flags in C/C++
If mainstream distros had opted into -ftrivial-auto-var-init=zero , they would have mitigated the rsync RCE attack chain. That flag is estimated to mitigate 10% of C++ vulns.
Think about all the router and video game software built in CI without sane compiler defaults. It lasts for a long time. Distros need flags in the compiler directly.
Chainguard's defense-in-depth security strategy protected against multiple rsync CVEs before they were even reported. See how we did it, using compiler flags.
When I was porting librsvg to Rust, sometimes the code would run into a non-invertible matrix. I thought that maybe it was a fact of life that SVG would sometimes produce non-invertible matrixes; the code would then ignore those objects and not render them.
In the end it turned out to be uninitialized data from the C code. When it all got ported to Rust, those bugs disappeared *and* the tests showed that there were no non-invertible matrixes anymore.
James Scholes
in reply to Nolan Darilek • • •No. The closest thing to an automatic solution to that is probably tracking scroll position, and you may or may not be able to make that more granular by making the text quite big (I haven't tried). But even then, the best you're gonna be able to achieve is to track the closest element and put focus back there to restore the position, without character-level accuracy.
I know that Mozilla and NV Access have done some work to allow selection of text within the NVDA browse mode buffer to be communicated to the browser for on-page actions that require a selection. But:
1. That doesn't work across browsers; and
2. your use case seems targeted at reading, not selecting.
Nolan Darilek
in reply to James Scholes • • •Thanks, that's what I was thinking. And just to check an assumption, setting the scroll position won't update the screen reader's position in the doc--I'd have to use focus shenanigans for that?
For context, this is my attempt at a document reader that saves/restores position when the document is closed/reopened. I don't think I need character accuracy, or even paragraph accuracy, if I can open books or longer documents to roughly where the reader closed out.
FWIW I'm not just being lazy and asking, I'm trying right now and it isn't working, which I suspected it wouldn't. It's also possible I'm using my web framework wrong or that something is behaving silly under Linux.
Thanks again!
Matt Campbell
in reply to Nolan Darilek • • •Nolan Darilek
in reply to Matt Campbell • • •Quin
in reply to Matt Campbell • • •James Scholes
in reply to Nolan Darilek • • •That's mostly correct.
There are some instances in which a webpage can move the scroll position without explicitly setting focus to the target element, and have the screen reader's reading position follow. But that can be less reliable, particularly if the target element is visually obscured, and setting focus is a more explicit/guaranteed way to do it.
Note that if you're setting focus to things like headings and paragraphs that aren't focusable by default, you'll need to dynamically inject a `tabindex="-1"` for the best results.
Nolan Darilek
in reply to James Scholes • • •