From the Lighthouse for the Blind, Inc.: Picture the Possibilities: Enhancing Image Description with AI using Today’s Leading Screen Readers: Wednesday, March 27, 7:00 PM Eastern Time groups.io/g/tech-vi/message/64…
Join technology scholar LAURA DENARDIS for a conversation with authors ARAM SINNREICH & JESSE GILBERT about data surveillance, digital forensics, and generative AI. REGISTER NOW\n
Accessible Social, a beginner’s guide to creating inclusive social media content, by Alexa HeinrichAccessible Social, a beginner’s guide to creating inclusive social media content, by Alexa Heinrich
I am currently working on #accessibility in #forgejo. And I discovered a problem with focus where it is not possible to navigate the page with "tab", a dropdown makes the focus skip to the end of the page and you end up in a loop.
Can someone recommend ways or tools to debug focus in webbrowsers? I have a hard time to see how the focus skips there. Any hints are welcome.
@modulux Thank you for the pointer. Yeah, I tried to use it. The problem is that it does not live-update and is not helpful when the page is dynamically modified with Javascript, for example. Anyway, we found a hotfix in the meantime.
Are you coming to the CSUN Assistive Technology Conference in Anaheim, California next week?
We're looking forward to catching up with everyone there! Don't forget to swing by Orange County 1-2 at 2:20 PM on Thursday 21st March for our session. Learn what is new and coming up in NVDA, meet our new CTO, Gerald Hartig, and have the chance to ask questions of the team!
It's SCaLE21x week! We're dragging out our suitcases and making our packing lists - can't forget the stickers! We'll be at Booth 215 Friday through Sunday, and we hope you come visit us. It's our first year at SCaLE and we can hardly wait. 🐧 🌴
And in a completely, totally, honestly unrelated question, we've been thinking about the kinds of Easter Eggs you can't eat. What are YOUR favorite hidden surprises? Any media counts! 🐣 😉
Update on Newton, the Wayland-native accessibility stack I'm developing for GNOME and (eventually) other desktops: I have an end-to-end prototype, using a Wayland protocol extension for the connection between applications/toolkits and the compositor, and D-Bus for the AT-to-compositor interface. I have an experimental branch of Orca with basic focus announcement and mouse review working. 1/?
Instructions for setting up and trying the prototype are available in the README in that Orca branch: gitlab.gnome.org/mwcampbell/or… Note: To get this working, you have to install both Mutter and the wayland-protocols package from experimental Git branches. But if anyone wants to try out what I've been working on, the pieces are all there. Next step: implementation in GTK; the current test setup uses a sample app from the Rust-based egui toolkit. 2/2
I should also explain why getting mouse review working in this prototype is significant. In upstream Orca, using AT-SPI, mouse review doesn't work on Wayland, because there isn't currently a way to find the window and the associated AT-SPI accessibility tree under the mouse pointer, or even to be notified about mouse movement across the whole desktop. Orca currently depends on X11-specific solutions for these problems.
One other note about the Newton prototype: I've validated that it works with the application running in a tightly restricted sandbox that allows access to the Wayland socket but not any D-Bus instance. Note: there's no support for running assistive technologies (e.g. Orca) in a sandbox; in my opinion, it's debatable whether it even makes sense to do that.
And I have to admit that while I was reluctant to switch to D-Bus for the AT-to-compositor interface, switching away from a Wayland protocol extension for that purpose does make the sandboxing more foolproof and easier to verify.
sweet seeing some progress here!! dumb question, is it tied to any toolkit or some internal mutter stuff? can it be eventually used on a wlroots compositor or kwin without pulling half of gnome? /genq
@halva While the only compositor implementation so far is in Mutter, I certainly want it to be as straightforward as possible to implement in other compositors. While the interface between assistive technologies and the compositor is now D-Bus-based, as preferred by the Mutter/GNOME Shell team, I anticipate that somebody could design a Wayland protocol extension for that instead, and then write a separate daemon to expose the D-Bus interface.
@halva I think you should probably get agreement with the KWin and Plasma folks before settling on that D-Bus interface. Once the interface is locked in, it's going to be very hard to fix later.
@Conan_Kudo I wonder if it would make sense to use a portal for that interface, even though assistive technologies like Orca are not currently sandboxed, and it arguably doesn't make sense to try to sandbox them.
Portals are bad because there's no pre-authorization support (such that the distro or admin can pre-grant it instead of the user approving it individually) in xdg-desktop-portal. As long as that continues to not be a thing, portals are wildly inappropriate.
@Conan_Kudo Would one use portals for ATs, a way to ensure it can be preconfigured by a distribution is indeed a must, and a perfectly reasonable thing to add to the portals ecosystem, not only for ATs. It's not a reason to not use portals though.
@jadahl For now, it absolutely is. So far, there's been zero appetite to fix it for all the other use-cases. At this point, I don't see it getting fixed for AT either.
@Conan_Kudo The context is allowing ATs to run in sandboxes, and to pre-authorize a theoretical portal or whatever other way for handling IPC does not exist anywhere at all. And implying there "is no appetite" to fix it in portals is also a mischaracterization of why there currently doesn't exist a solution for other portals.
@jadahl We've had a need for it ever since the Remote Desktop portal was introduced. We also have requests for it for the Input Capture portal. We even have requests for it for the Screenshot and Screencast portals! It's not theoretical, these requests go back *years*.
And frankly, every single portal should be capable of pre-authorization. It should be in the xdg-desktop-portal frontend interface itself. Instead, the backends have to hack together crappy workarounds to do it themselves.
@jadahl Don't get me wrong, the portal infrastructure is useful, but it has big gaping holes in the design and interaction model because people keep thinking of it as a Flatpak-oriented system.
We need to stop thinking of it that way because it *isn't* that anymore. It's used for system software all over the place. It's *required* for Wayland environments now.
It needs to be treated as the user interaction hot path it is.
@jadahl And specifically in the Newton AT case, mandating portals basically makes it impossible for wlroots-based Wayland environments to use it, because xdg-desktop-portal-wlr will *never* implement it.
@jadahl xdg-desktop-portal-wlr has a policy of only implementing portals to bridge to Wayland protocols. This is because the wlroots developers do not like having D-Bus as a mandatory dependency. Every portal must be connected to a Wayland protocol.
The only portals implemented are screenshot and screencast, and only partially at that (you cannot select individual windows because the Wayland protocol wlroots uses for this does not support it).
Currently there is a portal for registering global shortcuts in the works for xdg-desktop-portal v1.16. I've just created this issue here to: Gather opinions on how this might want to be implemente...
@Conan_Kudo @jadahl Well then, maybe I could resurrect the a11y-consumer Wayland protocol extension, get it approved by a plurality of Wayland compositors (presumably excluding Mutter), and then it would be supported by xdg-portal-wlr. I might have to do that work on the side though.
@Conan_Kudo Or, if the consensus is that it doesn't really make sense to sandbox ATs, then as an alternative to doing a portal, then my newton_consumer client library could directly support both D-Bus and Wayland protocols. As long as that project remains under my control; since the GNOME Foundation is paying for my work on it (so far), I don't know if I'll end up having to get approval from others for additions to that package.
@halva And it's certainly not tied to any toolkit. The current demo uses egui, only because that was a toolkit that already supported AccessKit, and Newton builds on the work I've done in AccessKit (without requiring Rust in the compositor!). For the integration in GTK, the plan is to use AccessKit via its C API. Maybe somebody will implement Newton in a toolkit without using Rust at all, once we stabilize the serialization format.
Opening this thread to discuss and schedule dropping 32-bit official Windows curl binaries, leaving x64 and ARM64. This would make development leaner, builds faster, scripts simpler. Non-Windows ta...
Today's #AndroidAppRain at apt.izzysoft.de/fdroid brings you 10 updated apps. One of them is HideDroid: if you had it already installed from here, you'll need to uninstall and reinstall, as the signing certificate was changed; its author replaced the debug key by a proper release key for signing 🥳
This is a repository of apps to be used with F-Droid. Applications in this repository are official binaries built by the original application developers, taken from their resp. repositories (mostly Github, GitLab, Codeberg).
An app designed specifically for playing conventional audiobooks. You have to get the books yourself, but once placed in the proper folder, the app will play
Hope this helps with having a healthy ARM version of Arch! It was such a pity to have lost the Asahi dev team as a downstream partially due to Arch Linux ARM not being as well maintained as the x86_64 version.
I would use a pre-quantum (e.g. EC) and post-quantum (e.g. KYBER) at the same time. The math behind KYBER Is rather new and other algorithms in the NIST competition have been cracked (rather late in the process!) in a couple of hours on classical hardware.
It kind of boggles my mind how quantum safe encryption is even a thing that can be achieved, especially without the use of quantum computing itself.
Does anyone know a good basic primer for someone who understands standard encryption (just) but isn't any kind of computer scientist / theoretical mathematician etc? @Tutanota
The blog post says what projects are being used to encrypt, but the only mention of how is one sentence about that they use problems considered "too difficult" for a quantum computer to figure out. There's no mention, not even a sentence, of what kind of problem or why it would be difficult, which is exactly what I'm wondering.
Samsung has officially launched its latest mid-range smartphones, the Galaxy A35 and A55. The A55, the higher-end model, boasts significant upgrades over its
General Activities LibreOffice 7.6.5 was released on February 22 LibreOffice 24.2.1 was released on February 29 Olivier Hallot (TDF) added help content for Calc’s XLOOKUP and XMATCH functions, Navigate By in Find toolbar, Draw’s Shrink text on overfl…
«Tuta stellt Post-Quantum-Verschlüsselung TeraCrypt vor»
Ist @Tutanota momentan die "Ersten", die es in ihrem #Email-Service anbieten? Ich frage nur, weil ganz neu sind die Erkenntnisse von Post-Quantum #Verschlusselung mMn nicht. Nun gut und zugegeben, ich verfolge das Thema schon etwas länger, wenn auch nur am Rande.
In case you missed it, we're hiring for three positions: - Technologist - Program Manager - Bug Resilience Program Manager
Join the Sovereign Tech Fund's team to strengthen digital infrastructure in the public interest. Please share; we're looking forward to getting your applications!
India has officially outlawed nine types of #UX #DarkPatterns, including saying "Hurry, only X amount left;" adding "processing fees;" adding dire language to opt-out buttons ("No, I'd rather not protect my purchase"); forcing people to agree to a EULA; forcing people to call a phone number to unsubscribe; using confusing opt-out language ("No, don't unsubscribe me"); blending ads into editorial content; and forcing people to click "remind me later" every day. bootcamp.uxdesign.cc/dark-patt…
The Indian government recently established guidelines to ban dark patterns in India. These guidelines are made to address underhand practices in digital design. These rules aim to protect consumers…
So @gnome is removing the x11 session, leaving just the Wayland one.
If this goes out before Orca, the GNOME screen reader, is fixed to work on Wayland, it will mean that people who rely on screen readers will have no way to use one on GNOME. And thus on the major Linux distributions.
So I’m hoping the plan is that this change will not land until GNOME has a working screen reader.
GNOME folks are well aware of the problems with Orca on Wayland, and actively working to fix them. There's even funding for this work, thanks to the Sovereign Tech Fund. I'm personally working on a new Wayland-native accessibility stack that aims to eventually replace AT-SPI and support sandboxed apps, but there are also efforts to fix problems in the existing stack in the short term. cc @sonny
It's the default because it's already quite functional, except for the problems you point out. Additionally, if you increase the number of users on Wayland, it is easier to identify problems and fix them. If it weren't the default, who even knows how long it would have taken us to identify that there were problems with Orca
@nah @matt @sonny If the GNOME folks didn’t realise that Orca was broken before making Wayland the default that’s even worse. It would mean no one on the GNOME team is testing accessibility. If they knew and made it the default anyway, it just shows that accessibility is seen as not essential. So I’m actually not sure which is worse. And we can’t go back in time so hopefully lessons will be learned and this will remedied as soon as possible.
Ok, but I remind you that GNOME is not a company, it is a project created by a lot of volunteers. Everyone works on what they want/can, the idea is precisely that there are people who test Wayland and come to help and fix the problems
I think we all agree that accessibility is important, and we all want to create software that is inclusive enough for everyone... but we must also come down to reality, where as long as the problem can be worked around, a lot of ppl think that nothing needs to be fixed
Removing the possibility of workarounding and shoving problems in your face is a strategy to attract volunteers to help improve the software
Finally: public appeals where bad faith is assumed and the project is criticized as if it were a company with a team of full-time developers is a strategy to reduce people's motivation to work on things. You don't need to shit on GNOME for things to be fixed, but on the contrary, the strategy is to show up with proposals and willingness to improve the situation
This is the first step towards deprecating the x11 session, the gnome-xorg.desktop file is removed along with the gnome-session-{x11, @} systemd targets. But the x11 functionality...
@nah @sonny @matt Accessibility is a human right. Drastic regressions to the accessibility of a system that disabled users rely on are a human right violation.
We can cut a lot of slack to volunteer contributors, but not “human right violations are permissible” amounts of slack.
Ultimately, it’s the responsibility of the project’s leadership (volunteer or not) to set norms like “we can only accept your contributions if they’re accessible”.
@nah @sonny @matt As a norm, that’s not particularly new in open-source. I’ve contributed to several projects who only accept contributions that 1) align with project goals, 2) pass peer review and 3) pass multiple test suites.
That can definitely be a barrier to contribution. But there are also smaller hobbyist-focused toy projects with lax norms that people can contribute to, if strict norms are a deal-breaker.
Privacy and security are equally important; however, people's solution here is for us to continue using X11 by default until the alternative is "perfect" and we can switch to it.
The change to Wayland is huge, and requires many improvements across multiple projects (many of which don't even belong to GNOME as such), so it's not exactly trivial. We need more people working on this, and therefore we need to raise the interest of potential contributors to iron out all the remaining rough edges.
As I said in a previous post: until Ubuntu put Wayland by default, many people were not even willing to work with Wayland. It's important to understand this: the world of free software moves by motivation. If you're a software user, and you have a problem with said software that you can work around... how much motivation do you have to improve said software? You'll probably continue working around the problem until you can't do it anymore, and only then you'll contribute to solving the problem once and for all.
Right now, the workaround is to continue in X11, even despite the security and privacy problems that this implies. Many people don't realize how insecure X11 is, and they minimize these problems. GNOME is aware of this, and made Wayland the default several versions ago, also because Wayland was already usable for many users. X11 was still available as a fallback in case Wayland wasn't ready for you yet.
The proposal now is to eliminate X11 once and for all, and to do so a goal has been set: GNOME 48 (which will not be released until March 2025). Blocking issues for this also began to be collected, the current state of Orca being one of them (see Sonny's post), and this has been 4 months now.
I must highlight that accessibility is a fundamental pillar within the development of GNOME, we seek to create a desktop that can be usable and accessible by everyone, without discrimination. However, I must also highlight that there are many more aspects to take into account in order to achieve this. It's not as simple as it seems to be, and at GNOME we always seek to achieve the best solution. Assuming bad faith and criticizing the project as openly as Aral does is not productive. Perhaps putting pressure will work in contexts such as governments or companies, but not in free software.
This is the first step towards deprecating the x11 session, the gnome-xorg.desktop file is removed along with the gnome-session-{x11, @} systemd targets. But the x11 functionality...
@nah @fvsch @sonny @matt No one is assuming bad faith on the part of the volunteers. The pressure is for the corporations who profit from GNOME. Take GNOME away tomorrow and what does Red Hat do? What does IBM do? What does Canonical do? They have millions. Can they afford to make GNOME accessible? Yes. Should they be ashamed it doesn’t have a working screen reader? Also yes.
@nah @fvsch @sonny @matt But here’s the thing: Wayland would never have been made the default if, say, fonts didn’t render correctly. Not having a functional screen reader is as big an issue for people who rely on screen readers. So at some point, someone at Canonical decided that it didn’t matter that people who use screen readers would be excluded. And so they should be ashamed.
Two GNOME cycles, when people are actively working on accessibility this year, is a long time.
Before panicking, consider: a reason it's been targetted to GNOME 48 (a year from now) and not 47 is, devs are collectively aware that not everything is ready for this to happen in 47, and that it is reasonably likely that a11y would be working by 48.
The cited link (!99) says pointblank it is a blocker.
@nekohayo @nah @fvsch @sonny @matt No one is panicking. By default, the most popular Linux distributions already ship with a broken screen reader. There’s nothing to panic about because this isn’t some hypothetical about a future harm. The harm (shipping an operating system with a broken screen reader) occurred when the first Linux distribution shipped Wayland as default.
@nekohayo @nah @fvsch @sonny @matt What I find disappointing is that instead of seeing this as a problem and taking steps to ensure it never happens again, I’m just seeing defensiveness that tells me that some folks still don’t understand that accessibility should be a showstopper, not a nice to have. That culture needs to change. And a policy adopted so it never happens again.
@nekohayo @nah @fvsch @sonny @matt You literally were just told - multiple times - that people know the problem and care about it and that this change is not even in the next version precisely because people are aware of this and other problems. You were told there are active and funded efforts to imprve the linux accessibility stack. Isn't that exactly what you want? Your last message in this thread feels very dishonest. Maybe there is a miscommunication here?
@lhp @nekohayo @nah @fvsch @sonny @matt Being told is one thing. Actions speak louder than words. Clearly the people who decided to make Wayland default before fixing the screen reader did not care about accessibility. It’s good that there are folks working on fixing the problem that do.
If you're going to keep repeating your "righteous" indignation, misplaced IMO, please at least remove Sonny from the thread, as I tried to do (after adding him). He's already doing everything he can. The past can't be changed, and I think there's no use belaboring it.
No one is assuming bad faith on the part of the volunteers. The pressure is for the corporations who profit from GNOME.
When you talk about GNOME, you're talking about the entire project and all its contributors, not just the companies that benefit from it. Keep that in mind: public attacks can reduce the motivation of contributors (as has happened many times before)
@nah @fvsch @sonny @matt I’m going to leave it at this: If I ran a large corporation that benefited from GNOME but didn’t want to support it financially or pay for it to be accessible, etc., I’d want any criticism of it to be seen as criticism of the poor, unpaid volunteers. It’s a great way for corporations to avoid criticism.
And I'll put it this way: if I work on a free software project for the benefit of the entire community, without earning anything from it, purely in my free time... and then someone decides to post criticizing the project from above to down about problems that I know and that both I and colleagues on the same project try to solve, without even helping... I really don't think it feels good to me.
I understand your point, but you also understand mine. If you want to attack companies, mention them directly, or criticize their products (for example, Ubuntu or RHEL). Attacking GNOME is useless in that case.
@nah @sonny @matt I am personally extremely critical of this framing of the GNOME Project as having no responsibility and no accountability, on the basis of most of the work being done by volunteers. This does not fly for other NGOs, and should not be used as an argument here.
Maybe time for the GNOME Foundation board and team to step up and set some clear norms and expectations for digital accessibility, if they have not done so until now?
- The MR's metadata does not show it as blocked by another issue or MR, as far as I can tell? - The comment you linked to says “potential blocker”, and the only response says it’s acceptable as a blocker *provided someone else commits to doing the work*. That is *not* a “hard requirement”, and probably requires an update.
This is the first step towards deprecating the x11 session, the gnome-xorg.desktop file is removed along with the gnome-session-{x11, @} systemd targets. But the x11 functionality...
@Myles124 @nah @sonny @matt I’m glad that “strict requirement” is the word from the highest ranking person here. And to close this loop, given that there is a variety of expressed opinions from contributors, I’d love to see written documentation on:
- Who has the authority to make this call (especially since the GNOME website says that technical decisions are not managed by the Foundation)? - What accessibility norms apply to GNOME releases. - How those norms are applied.
@Myles124 @nah @sonny @matt Personally, I’m wondering about the GNOME Foundation and Project's norms for accessibility. I’m reading developer.gnome.org/hig/guidel… but that’s mostly guidance about improving an app’s accessibility, not stating norms that the GNOME Project must follow or strict requirements.
@Myles124 @nah @sonny @matt Messaging from people involved in GNOME here and in the linked merge request has been mixed:
1. The MR author proposed blocking on this accessibility issue. 2. A reviewer replied that blocking on this issue is acceptable *provided someone else works on it*. 3. One contributor argued that accessibility is important but should not be treated as a strict requirement. 4. A GNOME Foundation director said it is a strict requirement.
@Myles124 @nah @sonny @matt Well, you can read up in this conversation to find arguments stating that it’s regrettable if there are accessibility regressions but so is the life of this volunteer-led project.
I’ve argued against those arguments, and am happy that GNOME leadership also disagrees with those arguments. I’m hoping this also translates to written requirements and norms, both in this specific merge request and in the GNOME Foundation and Project's guidelines.
“Email is a minuscule text file that you transmit across the internet. To claim that you don’t have the competence to arrange that yourself is a sign of administrative incompetence”, says Michiel Leenaars in an interview on Radio 1. He’s is responding to a recent report of Clingendael Institute warning the Dutch government has squandered cloud sovereignty by outsourcing storing vital & sensitive information to American providers. Listen to the interview here [In Dutch] nporadio1.nl/fragmenten/spraak…
We zijn slaven van bedrijven als Microsoft en Google, vindt onze gast Michiel Leenaars. En dat is nergens voor nodig... Er bestaan veel veiligere en betere alternatieven voor Hotmail en Zoom.
It's not a big leap to say how much work went into Thunderbird for Android/K-9 this February! 🗓️
In the latest update, we talk about the newest stable release and discuss what's next: the journey to Material 3 (and how beta testers can make it as smooth as possible) and Android 14 compatibility. ✨
And as always, you can find the latest community contributions and find out how to get involved! 💪
Y mientras, en Austria, no solo la mejor alcalde del año en Europa es de Graz (y comunista), es que en Salzburg, la derecha ha quedado fuera de las elecciones y ahora es cosa entre socialistas y comunistas. #MarxismIntensifiestt.com/artikel/30877899/salzbu…
Die Gemeinde- und Bürgermeisterwahlen im Bundesland Salzburg haben die Landeshauptstadt wieder tiefrot gefärbt und in den 118 Landgemeinden die ÖVP-Dominanz trotz Verlusten um weitere fünf Jahre verlängert.
stbák je a stbák bude. Jakože jedna věc je někoho sledovat z nějakého koníčku, ale to co bylo vidět ze zveřejněné komunikace, tak to bylo nechutné. Jakože to, že nechá svého poradce, aby si případně vymyslel lež o tom jak "se pirát nesnaží a nechá čechy ve štychu"...ten stbák nemá žádné zábrany ani řádné svědomí.
From the bottom of my heart I congratulate you on your anniversary! I wish you all the best! You have done an amazing job. Thank you very much. One of the huge number of satisfied TutaMail customers !!!
Tuta is an email service with a strong focus on security and privacy that lets you encrypt emails, contacts and calendar entries on all your devices. - tutao/tutanota
@joolstorrentecalo It's available for Legend already. We need to monitor server load a while before we can also release it to Revolutionary users. Thanks for your patience!
@klh_io We're very sorry about that, but we discovered a bug after the release and had to disable import for now. We hope to release the fix next week. Thanks for your patience!
that's understandable, but some comms (given how requested that feature has been apparently) would be nice as I couldn't find anything anywhere about the issue. Happy to continue my migration once it's working again though!
Your email/webmail is now so secure that even your users are not able to access it. Also I cannot create a new free account using another machine and 'device', so I could hopefully transfer the old mail there (I could not even find any information if this is possible). I have an old tutanota account which I can only access now with one single 'device' - as you call it - because I used the recovery line. And that will not work with any other 'device'. But the machine that has that 'device' needs to change the OS, and there goes the 'approved device'. I tried to explain this by addressing you an e-mail but... I have no access to that, only to predefined questions. What you call 'device' is a browser, and nowadays almost everybody has several machines and browser that should be able to access a webmail account - F.C.S. I have done a lot of positive advertising since I was allowed to create that mail account. That will change of course if you don't allow me to keep it. Happy New Year for you too.
As a paid customer I am with Tuta(nota) since 28.07.2016. That time a yearly subscription was only 12 EUR. But without calendar and quantum-safe encryption protocol.
modulux
in reply to fnetX • • •Accessibility Inspector — Firefox Source Docs documentation
firefox-source-docs.mozilla.orgfnetX
in reply to modulux • • •modulux
in reply to fnetX • • •