Welcome to the eighty-second episode of Access On, the National Federation of the Blind's Technology podcast.
Episode
Listen to the eighty-second episode of the Access On podcast (Browser).
Or listen on your preferred podcast platform.
Timestamps
This week on Access On:
- Getting a Waymo in Austin 0:00:00
- Summarize and transcribe podcasts with Snipcast 0:02:46
- Apple won't refund my subscription to an inaccessible app 0:27:56
- Microsoft said to be dropping SMS authentication 0:30:15
- More discussion on detecting deep fakes 0:35:57
- WhatsApp for Windows accessibility regression 0:37:34
- Giving consumers the best choice based on need, not price 0:38:53
- Thoughts on WWDC 0:47:47
- Comment on recent episodes and webinars 0:51:44
- Number Pad Commander in MacOS 0:53:36
- Accessibility developments at Microsoft 0:55:19
- Tech Tip, use Gemini to recognize handwriting 0:57:45
- Closing and contact info 1:00:04
Transcript
Speaker 1:
Live like you want.
Speaker 2:
Access On.
Jonathan Mosen:
Welcome to Access On, the technology podcast of the National Federation of the Blind. This week, too many podcasts and too little time? Use Snipcast to summarize your podcasts. Apple declines to refund a subscription for an app that's not voiceover accessible. Are people being given access to the right screen reader to meet their specific requirements? And in this week's tech tip, send Gemini those difficult handwritten documents. It's Jonathan Mosen at the Jernigan Institute in Baltimore, Maryland, welcoming you to episode 82 of the podcast. And there is a chance that you are listening to this episode as you wing your way to Austin, Texas, or travel by some other means to attend the 2026 National Convention. And one of the questions that we are getting a lot here is, how do I get a Waymo in Austin? Because there's much awareness that Waymo is operating in Austin, and this could be the first time for many people to have a chance of being in a self-driving vehicle.
Understandably, plenty of excitement about that. In Austin, Waymo does not use its own app, it uses the Uber app instead. And there are few things that you can do to increase your chances of getting a Waymo ride. It's important to note that you have to do these things when you're actually in Austin. If you try and do any of these things in a market where Waymo and Uber don't have this partnership, the settings simply won't appear. So, wait until your wheels are on the ground in Austin at least before you fire up the Uber app and give this a chance. When you are in Austin, open the Uber app and at the very bottom of the screen, the final tab, you will find the Account tab. So, double tap that to get into Account, and then choose Settings. On the settings screen, you will eventually find a Ride Preferences section, and from there you want to choose Autonomous Vehicles, which will appear when you're in Austin.
You'll want to toggle on the setting that says More Autonomous Rides, then you also want to toggle on Descriptive Audio. So, that's got you opted into receiving Waymo rides. Then, when you are calling an Uber, you will find that if you select Electric Vehicle as the type of Uber that you'd like to hail, that's another great way to increase the chances of being paired with a Waymo. Good luck, and if you do find yourself in a Waymo, share your experiences with us. [email protected], let us know how it went for you. [email protected].
In our role here at the National Federation of the Blind, we are, of course, advocating to provide more access to technology, that's important. But to get some sense of perspective, I do reflect sometimes on how far we've come on the accessibility journey. There was once a time when blind people were deprived of a lot of information, and we still are left out of some of it, but now if we have the technology to access it, we can suffer from information overload like everybody else. There is so much interesting content out there and finite time to consume it all. If you're someone who struggles with podcast backlog, or you want to retain more from what you listen to, Snipcast is worth checking out. And this section is all about Snipcast. It is a service that analyzes podcast episodes, extracts the main ideas and key takeaways and delivers a concise written summary directly to your email box.
At the most basic level, you find a podcast episode you want summarized, you copy the link to that episode, you paste it into Snipcast along with your email address, and within a few minutes, typically two minutes per hour of audio, a summary lands in your inbox. No account registration is required for the free tier, there's no switching between apps, and there's no disruption to whatever you're already doing. Snipcast works with any publicly available English language podcast from the major platforms. That includes Spotify, Apple Podcasts, Pocket Casts, Overcast, and Castro. If you have a link to a podcast episode on any of those platforms, Snipcast can summarize it. There are two limitations worth noting. First, Snipcast doesn't currently support Spotify-exclusive podcasts because there's no way for the service to access the audio files. This is a licensing restriction, not a technical one. Second, while Snipcast supports dozens of languages, some languages are newer to the platform than others.
If you're thinking inventively, you can actually take this platform beyond podcasting. If you have content that isn't a published podcast, the premium plan lets you upload your own audio or video files directly to Snipcast for summarization. This is incredibly useful if you're attending presentations, meetings, conferences, or webinars that you want quick takeaways from. Let me give you some concrete scenarios where Snipcast really does work well. Let's say that you follow an industry podcast or an education-focused show, but episodes run two or three hours. I do actually have podcasts like this. You find the topic interesting, but you don't have the time commitment available. You paste the link into Snipcast, get a summary in your inbox, and within five minutes, you know whether you want to return and listen to the whole thing. If you're on the Premium Plan, Snipcast will even provide a timestamped summary so you can jump to the exact moment a topic you care about is discussed.
Or what if you are a chronic podcast hoarder? You hear about interesting shows, you subscribe, but then within weeks you are drowning in a backlog of podcasts. You have episodes piling up faster than you can listen to them. Instead of feeling guilty about unsubscribing, you could use Snipcast to quickly triage what's actually worth your time, get summaries of recent episodes, and decide which ones deserve your full attention. Well, what if you're a professional who listens to industry podcasts but struggles to remember the specific takeaways for use in meetings or projects? Snipcast summaries give you something concrete to reference, search through, and save. If you want to, you can build a knowledge base of what you learned. One way I do this is by using NotebookLM. We have covered NotebookLM on Access On before, but if you're not familiar with it, is an AI service from Google, which only searches information that you put into it.
And thanks to Snipcast, I have set up two additional notebooks in NotebookLM. In one, I put Snipcast summaries of technology podcasts that have taught me something useful. I can then ask a technology related question and the AI will search those Snipcast summaries for me and give me the answer right away. I've also set up a similar notebook for self-care and health-related podcasts, and put the Snipcast summaries in there. Another scenario worth considering for Snipcast is if you are not a native English speaker and passing complex topics through audio alone is difficult. Written summaries are far easier to process in your own time, and you can look up unfamiliar terms as you go. You can also put summaries through translators if you wish. So, Snipcast has quickly become an important part of my podcast consumption workflow.
Snipcast operates on a freemium model and that means you can try the service without any commitment. The free plan gives you two summaries per month. There's no credit card required to use the free plan, and you can test whether the service fits your workflow. Then there's the Premium Plan. It is $5.99 every month, and it includes 50 summaries per month. I should add here too that Snipcast doesn't just summarize, you also get access to the full transcript of each episode, which Premium and Patron subscribers can copy and save. The timestamped summaries let you jump to specific moments in the audio, you can upload your own audio and video files, and get API access if you're a developer who wants to build this functionality into their own tools. The Patron plan is $9.99 per month. For that, you get 150 summaries per month, plus early access to new features as Snipcast rolls them out.
I know we have a good number of students listening to Access On and you will be pleased to know that students get a 50% discount on the Premium plan. If you're a student, all you have to do is email Snipcast from your educational email address and they'll send you a discount code. The simplest way to use Snipcast is to go to Snipcast.io. That's S-N-I-P-C-A-S-T.io. I'm on the Snipcast IO website now, and I'm logged in already because I have a Premium account with Snipcast. If you are not logged in, this page will look a little bit different, but you will have the opportunity to enter a URL of a podcast, to enter your email address, with no obligation, and to have that summary sent to you so you can try it out. What I'm going to do to demonstrate what Snipcast can do is bring up my list of links.
Speaker 3:
Links list dialogue. Links list view.
Jonathan Mosen:
And I'll press M.
Speaker 3:
My summaries. Seven of 26.
Jonathan Mosen:
And there's my summaries. So, I'll press enter to activate that.
Speaker 3:
Snipcast.io AI generated podcast summaries. Banner region. Navigation region. My summaries link.
Jonathan Mosen:
And I have quite a few podcasts that I've summarized. I'm going to search on the word Jaws.
Speaker 3:
Jaws finds dialogue. Heading level, three links. Features for mastering complex documents and webpages.
Jonathan Mosen:
This is a recent episode of the Freedom Scientific Training Podcast, and I decided after listening to it to summarize it so that I could put it in NotebookLM. I will press enter to activate the link that opens the summary.
Speaker 3:
My summaries. Snipcast.io. Freedom Scientific Training Podcast. May 29th, 2026. Freedom Scientific.
Jonathan Mosen:
Let's navigate by hitting, I've gone to the top of the page just to be sure I'm there and I will press H.
Speaker 3:
Six JAWS features for mastering complex documents and webpages. Heading level one link.
Jonathan Mosen:
I'm going to perform a say all now so that you can hear the kind of summary that Snipcast generates.
Speaker 3:
Heading level one link, six JAWS features for mastering complex documents and webpages, Freedom Scientific Training podcast. Copy summary button. Copy transcript button. Link graphic. Freedom Scientific Training Podcast. Heading level three about the episode, studio microphone in this episode of the Freedom Scientific Training Podcast. Host Liz and Rachel discuss six JAWS features designed to help users navigate and master complex documents and webpages. They cover tools like Picture Smart AI, the AI labeler, Page Explorer, and Smart Navigation, offering practical tips for extracting information and improving efficiency.
Heading level three, key takeaways, electric light, bulb bullet one, 48 Picture Smart AI for visuals. Picture Smart AI uses artificial intelligence to describe images, slides, and other visual content. It allows users to ask questions about the image and queries two different AI platforms, ChatGPT and Claude, to provide descriptions, helping to verify information and reduce hallucinations. Specific commands like Insert + Space + P + S can describe the entire screen. While Insert + Space + P + E can describe a specific control. Bullet 708 AI labeler for unlabeled elements. The JAWS AI labeler assists in labeling unlabeled webpage and PDF elements. It uses AI to scan images and text on a page to suggest names for unaccessible controls, which JAWS then remembers for future use. This feature is crucial for improving the usability of forms and documents where elements might not be properly defined by the creator.
Bullet 937, table layer for navigation. The table layer, activated by Insert + Space + T, provides a dedicated mode for navigating tables. Within this layer, users can use arrow keys to move between cells and control, plus arrow keys to move to the beginning or end of rows and columns. To see all available commands within the table layer, users can press Shift + /. Bullet 14, 25 Page Explorer for summaries. Page Explorer summarizes webpages and documents, providing information on content type, layout, and navigation tips. It can also identify important links and allows users to chat with the AI to ask specific questions about the content. This feature is particularly useful for untagged PDFs and complex webpages, helping to extract information or understand the page structure.
Bullet 43 hours and 43 minutes form, field navigation Insert + F5. The command Insert + F5 or Caps Lock + F5 opens a select form field dialogue, listing all form fields on a page. This allows users to see the total number of fields, the title of each field, and its current state “e.g”. Comma for a checkbox. It's a helpful tool for reviewing required fields before submitting a form and identifying potential issues. Bullet 45 hours and 47 minutes, Smart Navigation for layout understanding. Smart Navigation, Insert + X or Caps Lock + X, helps users understand the layout of controls and tables as a cited person would see them. It can announce group controls and allows for navigation using arrow keys, providing a more intuitive way to interact with complex or cluttered interfaces.
This feature can be configured in the settings center for controls, tables, or both. Bullet 48 hours and 45 minutes, place markers for frequent navigation. Place markers allow users to set specific points on a webpage for quick navigation. Permanent place markers can be set using Control + Shift + K, while temporary ones can be set with Control + Windows + K. These are especially useful for frequently visited pages or for navigating to specific sections that lack proper headings or quick navigation keys. Bullet 52 hours and seven minutes, Smart Glance Highlight, letter Y. The Smart Glance Highlight feature accessed by pressing the letter Y identifies text that stands out visually on a page but may not be navigable via standard quick navigation keys.
This helps users discover important information such as calls to action or key messages that might otherwise be missed on poorly structured web pages. Bullet 53 hours and six minutes, navigating text blocks, letter N. The letter N, November, can be used to navigate through blocks of text on a webpage. This is helpful when reading content that is interrupted by links or images, allowing users to move sequentially through the main text. Page Explorer can also be used to extract just the text from an article. Heading level three, chat with this episode, ask a question about this episode... Edit.
Jonathan Mosen:
That's the comprehensive summary Snipcast offered and a couple of things to note. The misreading of the time is related to the text to speech engine I'm using. That's not a Snipcast problem, it is a TTS issue. Second, most of the time handy commands to activate the features that Rachel and this were talking about were in the summary but not always. Page Explorer I think is a case in point there, where I don't believe the summary picked up the keystroke. So, you can chat with your transcript and I'm going to navigate to the edit field.
Speaker 3:
Ask a question about this episode... Edit.
Jonathan Mosen:
I'll turn forms mode on. And I will type, "What is the keyboard command for Page Explorer," and press enter.
Speaker 3:
Six JAWS features for mastering complex documents and webpages, Freedom Scientific Training podcast document.
Jonathan Mosen:
Now, I'm not getting any feedback about that so I'll have to navigate, I'll go back to the edit field.
Speaker 3:
Ask a question about this episode... Edit.
Jonathan Mosen:
And work my way up with the arrow queues.
Speaker 3:
Ask a quest... The keyboard command for Page Explorer is Insert + Shift + Caps Lock + Shift + E, if you're using laptop layout mode.
Jonathan Mosen:
Fantastic. That gave me what I was looking for right away. I'm going to go to the top of the page.
Speaker 3:
Six JAWS features.
Jonathan Mosen:
And now I'm going to press the letter B.
Speaker 3:
Copy summary button. Copy transcript button.
Jonathan Mosen:
I'll press the space bar wrong copy transcript.
Speaker 3:
Six JAWS features for mastering complex-
Jonathan Mosen:
And now that's copied to the clipboard. So I'm going to open Microsoft Word.
Speaker 3:
Search bar. Opening Word. Opening Word. Document one, Word. Document one. Document one. Document one, Word. Edit.
Jonathan Mosen:
Yeah, I think we might be in document one. I'm going to paste this now.
Speaker 3:
Pasted.
Jonathan Mosen:
And let's go to the top of the file.
Speaker 3:
Top.
Jonathan Mosen:
And just read a little of it.
Speaker 3:
Hi everyone and welcome to our web, and here are Six JAWS features for mastering complex documents and webpages. I'm Liz, and I'm Rachel, thank you for joining us today. We are talking about how to navigate and interact with documents and web pages, because as we all know, not all documents and pages are created equally.
Jonathan Mosen:
So, if you want to, you can copy the full transcript to the clipboard and paste it into something like NotebookLM so that you don't miss a single pearl of wisdom that Liz and Rachel dispensed in that podcast episode. Now, I'm one of those people who never listens to podcasts on my PC, I'm always listening on my phone because I can listen to podcasts when I'm doing other things. I might be on the treadmill, I might be doing something around the house and just want to hear podcasts while I work. So, the iPhone is what I always use to consume podcasts. So, let's talk about that scenario. Snipcast provides a prebuilt iOS shortcut that integrates into some iOS players. You add the Snipcast shortcut to your shortcuts library and you will find that on the Snipcast website, and then it will get integrated with the share sheet.
You might want to move it around a little bit and promote it, and the iOS share sheet is sort of interesting, but you will eventually locate it, I'm sure, and move it somewhere prominent. One thing to be mindful of is that the first time you execute the shortcut, it's going to prompt you for your API key for Snipcast. The reason why that's important is that once it has your unique API key, it knows who you are, it knows about the email address to which you want summaries sent, and all those important things. That is a one-off process, you do have to go find the API key, copy it to the clipboard, paste it into the right place in shortcuts, but it's a simple enough process, and it's also set it and forget it. So, once you've done it once, you won't need to do it again.
I use the Castro Podcasts app on my iPhone. Without it, I wouldn't be able to subscribe to as many podcasts as I do. I triage a lot of episodes every day and Castro's kind of like email, everything appears in your inbox. If you want to listen to an episode, you add it to the queue, and that leaves you with all the episodes in your inbox that you don't want to listen to once that process is finished, and then you just triple tap the inbox tab and clear them all. So, it's incredibly efficient to go through a very large number of podcasts. So, I'm in Castro right now and here's an interesting episode.
Speaker 4:
Are you getting the most out of RSS? School of Podcasting. Expert Tips for Launching and Growing Your Podcast. 35 minutes. Episode 1041. Published today. Are you getting the most out of your RSS feed? To figure out if you are, you first have to understand how a podcast workflow works, which means you need to understand what an RSS feed is, how you make one, where you get the best feed, and how you make sure you're getting the most out of it. Don't make these mistakes in this episode. I'm going to explain some of the mistakes I see people making with RSS feeds, especially when it comes to getting the most out of transcripts.
Jonathan Mosen:
Well, that sounds like an interesting episode and something that I might want to add to my NotebookLM. So, I'm going to double tap this. How you do this will vary a bit depending on the podcast app that you have.
Speaker 4:
Back button. Share button.
Jonathan Mosen:
There's the share button, that's what I'm interested in, so I'll double tap that.
Speaker 4:
Share. Image.
Jonathan Mosen:
Now, the share sheet has come up and I need to flick through quite a lot of items before I get to the one I want. So, I'll do that quickly.
Speaker 4:
Zoom, Claude, more. Summarize with Snipcast. Button.
Jonathan Mosen:
For me, I have moved the summarize with Snipcast to be right to the right of the more button in the share sheet. So, it's relatively easy for me to get to. I'm going to double tap this option, summarize with Snipcast and we'll hear what happens. Obviously use the shortcut regularly. So, for me, it is already set up.
Speaker 4:
Back button. Castro, success.
Jonathan Mosen:
And that confirms that it's worked.
Speaker 4:
Okay. Button. Your episode of School of Podcasting. Expert Tips for Launching and Growing Your Podcast is being analyzed, and the summary will be delivered to your inbox in a few minutes. Okay button.
Jonathan Mosen:
I'll double tap that.
Speaker 4:
Okay.
Jonathan Mosen:
And we'll come back to that summary in a moment. But now we're really starting to see the value of Snipcast. If you can integrate it with your podcast workflow and just choose the ones that you want summarized, it just becomes second nature. But now, here's where things get pretty clever. People who listened to Living Blindfully will already be familiar with this next service, but if you weren't a Living Blindfully listener, let me introduce you to a service called Listen Later, which you can find at listenlater.net. Listen Later does something different, it converts articles, newsletters, and text into podcast episodes. You email it a link, or you forward a newsletter, and Listen Later uses AI to convert that text into an audio episode in human-like narration. The episode lands in your personal podcast feed that you can subscribe to in any podcast app. You're essentially turning your reading list into a podcast playlist.
I find it very convenient to have long bits of text in my podcast app where I can listen to them, they just come up alongside other things that I'm listening to. Connecting these two services together produces a very convenient experience. When you get a Snipcast summary, it lands in your email as a written summary. But what if instead of reading that summary in your email client, you could listen to it as a podcast? Essentially, you take a long podcast you don't have time for and you convert it to a short podcast you can listen to in the same place. So, here's the workflow. The first step is to set up a Listen Later account if you don't have one already, at listenlater.net. When you do that, you'll be given a custom email address, it's unique to you for your Listen Later feed. This is the email that you send articles and newsletters to.
Next, you want to subscribe to your Listen Later podcast in your podcast app of choice. And because all of the episodes in my Listen Later feed are things that I have expressly added, I've got my Listen Later feed set up so that all the episodes immediately go into my Castro queue. They don't even show up in the inbox because I know that I want to hear them because I put them there. So, without any intervention, they're in the queue waiting for me to hear in the order that they came in. After that, instead of using your personal email address on Snipcast, use the Listen Later email address instead as the destination for your summaries. Now, when Snipcast generates a summary, it sends that summary to your Listen Later email address, Listen Later receives the summary as an email, sees a text attachment, and it converts that summary into an audio podcast episode using human-like narration that you get in your podcast app.
And that's what I have set up. So, we're going to go and check out the summary of the School of Podcasting episode that we just looked at. But now that I have all this set up, I found a podcast episode that I wanted to know about, I submitted it to Snipcast, by now, it will have well and truly finished generating a written summary. That summary went to Listen Later, it converted it to audio, and now I have a podcast episode that is the distilled essence of a longer show, delivered in audio form in my regular podcast queue. Now, we haven't lost any of the benefits of having text, even though Listen Later has done its magic. I can still go back to the Snipcast website and find the summary on the website if I want to take some other kind of action with it. So, let's go into the Castro queue on my iPhone.
Speaker 4:
Q Tab button.
Jonathan Mosen:
And let's see what's here.
Speaker 4:
Now playing. Are you getting the most out of RSS? School of Podcasting. Expert Tips for Launching and Growing Your Podcast. Listen Later podcast, three minutes, published-
Jonathan Mosen:
All right, so this one was originally about 35 minutes, I think, it's distilled it to three. Let's hear what the summary sounds like. I will go down to play.
Speaker 4:
Play.
Jonathan Mosen:
And let's have a listen.
Speaker 5:
Are you getting the most out of RSS? School of Podcasting. Expert Tips for Launching and Growing Your Podcast. In this episode of the School of Podcasting, host Dave Jackson explains the importance of RSS feeds for podcasters, using analogies to radio and the body's circulatory system. He discusses common mistakes like relying on non-RSS platforms for transcripts, and warns against moving away from RSS. The episode also includes a mailback segment, answering listener questions about dynamic content and podcast measurement. Key takeaways, RSS as the source of truth. RSS, real simple syndication is the core of podcasting because it syndicates your content to all podcast apps like Apple, Spotify, and Pocketcasts.
Everything in your feed is the source of truth. If you change your media host or feed, it's not outpatient surgery. Don't multitask. Dave compares RSS to a radio frequency. Your media host is the tower and the RSS feed is the signal that apps translate into playable episodes. 10, 11. Transcripts must be in RSS feed. Transcripts should be added to your media host's RSS feed.
Jonathan Mosen:
I will stop it there, but there you hear the transcript, it's comprehensive, and that's a pretty natural sounding voice, if you like that sort of thing, from listenlater.net. So, there is a lot that you can do with Snipcast, it's a great service, and when I contacted the developer initially, he didn't support Castro, but he'd added Castro within hours. So, he's responsive, and he's interested in accessibility. You can find out more by going to Snipcast.io.
Make a difference with the National Federation of the Blind's Keep NFB Advancing, Give $25 campaign. Each year, thousands of federation members and friends contribute to support blind people, but we still need your help to fund our programs in 2026 and beyond. When you give $25 between May 15th and July 7th, you're entered into the Keep NFB Advancing drawing. Each $25 increment is a chance to win. Your support helps us continue to build a network that advances the lives of all blind people across America. You could win prizes like round trip transportation for two to the 2027 NFB National Convention, hotel accommodations, registration, banquet ticket, or $2,000 cash.
Speaker 6:
And you can double your dollars up to $25,000, thanks to a gift from IRA, the visual interpreting service.
Jonathan Mosen:
Want a chance to win a BrailleNote evolve from HumanWare? Become a federation connector. Ask friends and family to contribute before National Convention and indicate that you prompted their giving. The Give $25 drive supports the Kenneth Jernigan Fund, SUN Fund, TenBroek Memorial Fund, and the White Cane Fund. You can choose a fund when you donate. To enter, visit nfb.org/give25donate. That's nfb.org/give25donate. Call 410-659-9314, extension 2430, or send a check to National Federation of the Blind and mention #Give25 and the fund in the memo. The winner will be announced July 8th, 2026. Thank you for your generosity.
Let's go to our busy email box, [email protected] if you would like to write in. [email protected], you can attach an audio clip if you want, or you can write down the email. Scott Rutkowski in Australia is writing in. "A few days ago, I purchased a subscription to the Udio app because the free version is very restrictive. Unfortunately, after paying for the subscription and exploring the app, I discovered that it contains many unlabeled buttons and is completely inaccessible. Following this, I submitted a refund request to Apple, explicitly detailing that I am a voiceover user, and that the app is entirely unusable for me. Shockingly, my request was denied. I immediately appealed the decision explaining the situation again, but it was denied a second time. In an attempt to resolve this, I spoke with two Apple senior advisors, one in accessibility support and another in media services, but unfortunately they were unable to help me.
What really frustrates me is that Apple's automated and manual refund processes completely fail to take into account those of us who rely on voiceover. Purchasing an Apple subscription in good faith only to be denied a refund when it turns out to be entirely inaccessible is incredibly disappointing and upsetting. This has been an ongoing issue for some time now, and Apple seems to refuse to make any sort of meaningful change to support users with disabilities in these scenarios. I'm reaching out in the hope that the NFB might be able to liaise with Apple on this matter. Apple needs to implement a dedicated refund process for users who buy software in good faith, including a specific refund request for voiceover users to claim a refund when an app is fundamentally inaccessible."
Seems very reasonable to me, Scott, I would be interested in hearing from others who've had similar experiences. I'm not sure if I've ever tried to get a subscription refunded because typically what happens is I try the app first, and if I find that it's not accessible, then I don't get a subscription. But I can certainly envisage a situation where certain features are just locked out until you subscribe, and in that scenario you have no way of knowing that the app isn't accessible until you've subscribed. If you have anything you'd like to share on this, drop us an email, [email protected], and Scott, I appreciate you raising this with us.
Stephen Hudson says, "Way back in episode 15 of Access On..." Wow, that's going back away now. "You talked about Google possibly dropping support for using SMS as a form of two-factor authentication, and the accessibility problems that might cause. Ultimately, we were assured by Google that this wasn't going to be the case in the foreseeable future. I've recently been made aware that Microsoft is now in the process of quietly phasing out SMS as a form of two-factor authentication for personal Microsoft accounts. So, I believe we now face a similar situation. To briefly summarize what Microsoft is saying to justify this, SMS is vulnerable to phishing and SIM swap attacks, and is no longer considered to be secure. Microsoft has also been aggressively pushing users to start using pass keys instead of passwords. People can still use a verified backup email address or authenticator app as forms of two-factor authentication, or sign in with pass key. It's also still possible to generate backup codes that should be printed out and stored somewhere safe.
As many of us are surely aware by now, passkeys and most forms of two-factor authentication assume that a user has access to more than one device. That device could be a smartphone, tablet, or another PC. It's also possible to store pass keys on security keys, such as UbiKey. The problem comes in when a user relies on one device. Perhaps they don't really need a second PC and they use a flip phone rather than a smartphone. If their device breaks, is stolen, et cetera, then they're probably out of luck in most cases. Even if a blind person has a smartphone, authenticator apps require the user to scan a QR code. Some of us use computers without screens connected, and even if we do use a computer with a screen, scanning a QR code isn't always easy. Here's what I believe is the worst part of all this though, Microsoft requires a Microsoft account just to set up a new PC.
There are ways around this, but the average person isn't going to know about or bother with them, and Microsoft sometimes closes those loopholes. Let's assume that someone doesn't use a smartphone or tablet and they're replacing their PC that is broken, they have two-factor authentication set up, but because they don't use an authenticator app, and SMS will no longer be supported, they have to use a backup email address. In order to access that backup email address, they have to set up the PC, but in order to set up the PC, they need to access the backup email address to get a code to verify that they're the owner of the Microsoft account.
The only option at that point would be to create a new Microsoft account so they can access the old one. It's not exactly a problem that affects blind people only, since anyone could end up falling into this trap. But considering that the price of devices is increasing at an alarming rate, and some of us might not be able to afford to buy more than one device even if we wanted to, it does possibly affect us more than others. I don't really expect Microsoft to reverse course on this, since they've been pushing pass keys pretty hard for a few years now, and they're correct about SMS being vulnerable to attack. But it may be worth starting a conversation about it anyway so they're aware. Security is important, but this feels like actively punishing people for not having access to more than one device and possibly encouraging them to turn off two factor authentication entirely."
Thank you, Stephen, I appreciate that. I will raise it with Microsoft. A couple of things to note, the first is that Microsoft is about to release a very easy way for you to bypass the Microsoft account requirement when you set up a PC. This is something that many of us have been requesting for a long time. As you say, there are workarounds, but they're a bit convoluted and sometimes they disappear. But Microsoft is actually providing a sanctioned way to get this done in the very near future. So, this is probably happening in parallel with some of these changes. The second thing I would say is that I use pass keys on one device all the time, I do use 1Password, and on Access On we have provided a tutorial on 1Password and passkeys. And if you use a password manager like that, you don't need a second device.
Third, while scanning a QR code is certainly the default behavior, it's not the only behavior in any scenario that I've seen. Normally there is a code that you can type in. And again, although there might be a bit of a security compromise in doing this, you could certainly use 1Password to do the authentication on the same device. So, quite often with two-factor authentication, I have a unique password and then I have to authenticate with what would normally be an authentication app, but 1Password completes that step for me too. So, it is possible to log in using 2FA with just one device. So I think there are workarounds. I don't think it completely negates what you're saying though in the sense that there is some degree of technical proficiency required for all this.
So, while SMS is vulnerable technology, there's a simplicity about it that is attractive. So, it's a difficult one, isn't it? I'll certainly raise this with Microsoft. Microsoft is always willing to engage with us, and we will find out whether they have looked at these issues through an accessibility lens. And while we're talking about security things, it seems like a good time to return to the subject of deepfakes. Here's Hayer Simkin. "I thoroughly enjoyed listening to your segment on deepfakes. Several observations came to mind, including the unnatural flatness of some voices, as if real people had been replaced by a text-to-speech program. An exception to this is John Wayne, who always sounded that way. Pay attention to the speed of their speech, it is unusually slow or fast.
More importantly, does the video have an attention grabbing emotional clickbait title, such as, You Won't Believe what Jonathan Mosen said, or This is What Happened When Jonathan Mosen... I rarely click on anything with a title like that because I know it is attempting to manipulate my emotions. I am unsure if deepfake videos begin with such titles, but it is likely that they do, as they're attempting to prevent viewers from scrutinizing the content or seeking the source. This is true of all clickbait material. Similarly, ask yourself, is this what I want to believe? Does this align with my biases? What does it exclude that contradicts my biases? Even the stopped clock is correct twice a day, but it is more likely to align with your biases and evoke a strong emotional response."
This is not necessarily true. These are simply safeguards we should implement consistently regardless of our vision, political beliefs, or any other factors. And here's Francisco Crespo, who says, "Hi, Jonathan. Here I am with my favorite issue, WhatsApp on Windows. You know, not my fault that WhatsApp breaks things for blind users all the time, and that you were super efficient at bringing things to their attention. So, thank you very much for helping out with the latest issue, and here's the new one. When I make a call on the Windows version of the WhatsApp app, a call window opens that displays all the call controls, such as audio settings, the end call button, the checkbox to mute and unmute the microphone, camera control.
If it's a video call, et cetera. Well, these controls now can't be navigated with JAWS at all. I tried with tab arrow keys, I can't turn the virtual PC cursor on there because it's not a web interface. Even the JAWS cursor isn't working. So, I would say it's not a JAWS issue, but it could easily be a keyboard navigation issue that affects more users than just blind people. I really hope that you can do your magic this time."
Thank you, Francisco, it's been a while since this message came through and I'd like to hope that it's resolved, but let's see. We'll try and check it out and see if we can escalate this with Meta. And now a thought-provoking email once again from Michael Evers, who says, "I have been following the recent discussions in the blindness community regarding JAWS, NVDA, vocational rehabilitation funding, and whether blind consumers truly need access to both screen readers. I thought I would share my perspective because I believe this conversation is about something much bigger than choosing between two pieces of software. At its core, I believe this is a discussion about consumer choice and autonomy.
For context, I'm a blind professional and former assistant director of disability services in higher education. Throughout my career, I have used both JAWS and NVDA. Like many blind technology users, I have found that each screen reader has strengths and weaknesses. There are situations where NVDA performs exceptionally well, there are also situations where JAWS performs better. In my own experience working with enterprise software, Banner 9, university administrative systems, and various web-based applications, JAWS often provided a more efficient and productive experience for the work I was expected to perform every day. Could I have completed some of those same tasks using NVDA? Probably, but possible and optimal are not necessarily the same thing. That distinction becomes important when employment productivity and career advancement are on the line. What concerns me is the growing trend of some agencies and organizations viewing NVDA as the default solution, simply because it is free.
I want to be clear, I am not criticizing NVDA, in fact, I think NVDA is one of the greatest success stories in assistive technology, and has opened doors for countless blind people around the world. My concern arises when cost becomes the primary factor driving decisions about what blind consumers should or should not receive. One analogy that comes to mind is healthcare. Imagine a patient suffering a heart attack. The cardiologist recommends a specific treatment because it offers the best chance of recovery and long-term success. Now, imagine the insurance company responds by saying, "We understand the doctor's recommendation, but we believe aspirin is sufficient for most people. Therefore, we will only approve the more advanced treatment if additional justification can be provided." Most of us would find that troubling. The reason is simple. The people controlling the budget are not the people living with the consequences. I sometimes wonder if we risk falling into a similar trap within the blindness field.
If a blind consumer says, "This is the technology that allows me to work most efficiently, compete effectively, and succeed in my career," should an agency's first response be, "Can you justify why the cheaper option is not good enough?" Or should the consumer's lived experience carry significant weight in that conversation? To me, the most important principle is that blind people should have meaningful input into the decisions that affect their education, employment, and independence. We are often referred to as consumers, but we are blind customers. We are the individuals who must live with the consequences of these decisions every day. If a particular screen reader, magnification program, notes taking device or accessibility solution allows someone to perform their job more effectively, that person's experience should matter.
I am not advocating for a world where every blind person uses JAWS, nor am I suggesting that NVDA is somehow inadequate, rather, I believe blind consumers should be exposed to multiple tools, understand the strengths and weaknesses of each, and then be empowered to make informed choices based on their own goals, careers, and circumstances. Technology should serve the consumer, the consumer should not be expected to serve the technology. I would be interested to hear your thoughts as well as the perspective of other Access On listeners. Are we moving toward a future where cost considerations are overshadowing consumer choice, or are we simply seeing a natural evolution in assist of technology? More importantly, how do we preserve informed choice while still being responsible stewards of limited resources?" Well, thank you once again, Michael, that's another great conversation starter.
This is an old issue in some respects packaged in a new way. I mean, I remember that back in the day where there was competition in the Windows screen reader industry, that was commercial in nature, sometimes you would find states that were JAWS states, and sometimes they were Window Wise states, and it didn't really matter very much what was best for the customer, because just like JAWS has strengths and NVDA has strengths, so did JAWS and Window Wise back in the day. But it seemed to come down to what the state preferred to fund or in some cases what particular technology trainers were willing to train and then recommend because it was what they knew.
So this is always a problem. I suppose there are a couple of imperatives, one is that yeah, these organizations are cash strapped, so sure, the price of NVDA does play a role, it is a factor that some are considering. The other thing too is that with the internet being so abundant now, it is possible for those who do have some technology knowledge already to do their own research, and perhaps come up with information about what might suit them in the best situation. If somebody is new to technology, perhaps they've become blind later in life or they're adopting technology later in life, then they may not know what they don't know. But absolutely we would certainly support the idea that somebody should be assigned the technology that facilitates their optimum participation in the workforce or in education, the cost should not be a consideration, it's facilitating the best outcome for that individual.
Now, how you determine that is of course the question. And sometimes it's really clear cut if there are particular scripts that are available for JAWS or a particular add-on that's available for NVDA. But at other times it is quite subjective. One of the things that I've always been passionate about as an individual is efficiency, because all those little micro-efficiency gains that you can gain over time, even to the extent of taking away superfluous things that your screen readers say, they all add up over time, and they just lessen fatigue, they make you more productive. But a lot of people don't seem to care as much about efficiency as perhaps we may have 20 or 30 years ago, and some of that, I think, is due to the drift that has occurred where mainstream software companies are playing a bigger part in the technology experience for blind people.
An example of this is Microsoft Edge, which is very verbose. If you load a page in Edge, by default, it's going to say, "Loading page," and, "Still loading," if it's taking long enough, and then, "Load complete," and that kind of thing. And what actually happened was that JAWS had to introduce technology to minimize that verbosity. So, Microsoft was sending all of the stuff to screen readers, and Vispero had to spend time and effort and limited development resources actually winding it back, shielding the user from the verbiage that Microsoft was inflicting. But then of course you will hear people who say, "Well, actually I don't consider that excess verbiage, I consider it helpful information." So, it is a really interesting question, and certainly we try to play our part here at the National Federation of the Blind in creating an informed marketplace.
So, we can point out where there are strengths and weaknesses in particular products, and again, some of this is subjective, not everybody will agree with every finding that we make, but we're putting things out there for people to consider and perhaps assess for themselves. So, two points for me. First, absolutely cost should not be a factor, we should be equipping people with the best tool for the job. And second, it is vital that we encourage informed consumer choice and that we encourage people to advocate for what they need having made that choice. It is not acceptable to be fobbed off with technology that is not as effective for a given use case simply based on cost. Hopefully others will chime in on this one as well. [email protected] is how you get in touch, that's all one word, [email protected]. You can attach an audio clip to the email, and that will give the audience a break from me, and they'll say, "Hooray for this."
Or you can just write the email down and I will read it as I read the email from Michael. Well, we're working through the backlog of email, and we've got to this one that was sent right after the Worldwide Developers Conference keynote, from John Reel, he says, "Hi, Jonathan. I just finished listening to your rundown of today's WWDC event. As usual, you, Mike, Judy, and Heidi did a thorough job. However, you all missed something in the presentation if an article from Mac World is correct. Namely, that the advanced AI features such as advanced dictation and expressive voices will only be available for iPhone 17 Air or 17 Pro users. I am very underwhelmed with all these supposedly nifty AI features, I admit I'm really old school, and I'm just now getting into using Claude. I do use Be My Eyes and Seeing AI, and think those apps are an excellent use of AI.
Second, as Mike pointed out, you've got to remember that these new nifty features exist, particularly if they're buried in menus or aren't obvious. I'm 73+ and I admit my memory isn't as good as it used to be, and remembering all these new features and new screen layouts is truly overwhelming sometimes. I would rate today's WWDC a three, with the caveat, don't talk about it, show me. And as an aside, I'm not sure I want an AI program scrounging around in my emails, chats, et cetera, to 'Help me do things.'" Good to hear from you, John. Yes, we recorded that WWDC recap immediately after the conclusion of the keynotes, so we were still getting the details coming in about the specifications at that point, and there was a lot of information to sift through.
So, the older the phone you have, the more reliant on the cloud some of these AI features will be, until you get to the point where you don't have the Apple intelligence features at all, of course. I'm running an iPhone 17 Pro Max at the moment, and I am running the beta of iOS 17. There's only a small number of enhancements this year, but they are quite consequential. For example, you have image descriptions on the rotor now, and that can make a really big difference if you just want to peruse your photo album and go through descriptions quickly, or maybe you're on Instagram or somewhere where there are a lot of visuals, and you can just so quickly now get a description of the photo, and it's quite a good description.
It's interesting the approach that Apple has taken with this actually, it's a little bit less opinionated than some of the other services that we have access to. So, you are going to get quite a detailed description, but you're not going to get something that says, "The room depicts a cozy setting," or, "The overall impression is that this is a cluttered space," or something like that. It's to the point, it's factual, but it's also pretty quick, and just having access to that on the rotor is quite impressive. The new Siri is absolutely streets ahead of the old one. There are some delays sometimes, but I don't know whether that is representative of the fine experience because we're in beta right now, and pretty early beta at that. So, overall, I would say the quality of what we have is very good, but if you're looking for a whole bunch of new features, you're not going to get a large number of them.
This is what people have asked for though, right? People have said we want Apple to just take some time out and fix some of the things that they have broken. So, I think another way that we can assess progress is how many longstanding defects that have been around for so long that we almost don't notice them anymore, get fixed by the time iOS 27 comes out in September. You could even argue, and I'm sure some Mac users would argue, that macOS might be in need of even a little more love than iOS in terms of voiceover support and stability. So, we will see where we get to by the time iOS 27 comes out.
Eric Duffy:
Greetings, Jonathan, Access On podcast listeners and fellow federationists. This is Eric Duffy from Columbus, Ohio. Jonathan, I want to congratulate you on the excellent recap of WWDC, very much enjoyed that, as I always enjoy your recaps of the Apple events. Then, I want to comment on the webinar on outdoor navigation, that was excellent. I use Strider with Mike and the team, walking from the Anaheim Hotel, which I don't remember the name of, to the gate of Disneyland. That was a very hot day and things kept coming apart and we'd have to stop and the team would glue things back together and continue on. I was sort of racing against a cane user without GPS, and would have won the race had things not kept coming apart, and making us stop, and glue them all together. It's been an interesting journey through GPS systems throughout the years and glad to see that Mike is still very much involved. Great work, and look forward to see what's coming next.
Jonathan Mosen:
Thank you very much, Eric. There was something exciting about those times in those early days when we were experimenting with the technology, and I think we were a lot more tolerant of those sorts of bugs and we realized that we were on the cutting edge. I think now we're at the point where we say these things are a tool that we just want to get the job done with, we've passed that experimental stage for the most part. But they were very interesting times, and we were amazed at how far we could go. And yeah, it's great, Mike is a legend. He's contributed so much to that space, and it's great to hear him still thinking about the challenges before us. This email is from Paul Hopewell.
"Hello, Jonathan. I would be interested in your views on the below. Why do macOS commanders behave like this, and could Apple be persuaded to change? I use NumPad Commander to navigate voiceover on macOS, as it's much more convenient than using multiple standard keys. This all works fine until NumPad Commander is near to a secure edit field such as a password, then the NumPad Commander settings are ignored and pressing NumPad keys results in digits being entered into the secure edit field. When using NumPad Commander to navigate voiceover in 1Password, I often unintentionally change password fields. I believe that all commander settings are ignored for secure edit fields and it has been like this for several releases. Do you know why macOS commanders behave like this? Why are secure edit fields different from any other field? Is it somehow related to Braille? And if so, could the behavior be changed so that it only happens while a Braille device is being used?"
Thank you for writing in, Paul. I haven't had the opportunity with convention looming to go into the International Braille and Technology Center and sit down with a Mac and just verify this, but that does seem like a bug to me, I may be missing something so perhaps another Mac user who can explain this or tell us if they can duplicate it would like to contribute. It may just be something that needs to be drawn to Apple's attention. I cannot immediately think of a Braille related reason for this behavior, and you would think that if you've assigned the NumPad keys to actors navigation for your screen reader, that should apply no matter where you are, one would think.
And before we get to the tech tip, here's Aaron Espinoza who says, "I listened to episode 76 of Access On. All of the updates coming to narrator sound great, hopefully the days when the only reason anyone used Narrator was to download NVDA or JAWS will come to an end. I have mixed feelings that the Access Technology Affordability Act, which we in the Federation advocate for each year will drive innovation in the screen reader field. However, if it does, then bring on the screen reader wars, I like the current competition. Narrator is the built-in operating system screen reader, NVDA is an open source screen reader, and JAWS is a commercial screen reader. Another commercial screen reader in the mix would be a good thing as well."
Well, just pausing briefly on your email, Aaron, to grant your wish just like that, because there is another one and that is SuperNova from Dolphin, which is a combination of their magnifier and Dolphin screen reader. It doesn't have a large following at all in the US market, but it does exist, and it's a lot more popular in European markets. Aaron continues, "It's great that the NFB has the influence to talk with Microsoft about the Copilot key, and work with them to address the issue. I was reading some older NFB national convention materials and someone mentioned that they remember attending their first National Convention in 2001 when Microsoft was present.
They said the conversations with Microsoft at the time were completely different. Then they reflected that the fact that Microsoft now has a chief accessibility officer is a testament to the power and influence of this organization, and the value of working with it. I have a short story about the Copilot key on the new Windows computers. I got a new computer and was trying to copy and paste something, if I remember correctly, and Copilot kept launching. I had completely forgotten that the new computers came with that key. I spent far too much time trying to figure out what was wrong. I thought Windows had remapped the key or that it had something to do with NVDA. It wasn't until I intentionally pressed only the Copilot key, did some more research, and had checked with Ira that I realized it was the Copilot key causing the issue, and remembered that my new computer came with it."
Tech tip time. Tech tip time, you got to love that alliteration, and got to love this one from Catherine Samuel, who says, "Hi, Jonathan. I just made a wonderful discovery and wanted to share it with other listeners. If you have a handwritten document that you cannot get any of the usual OCR tools, like JAWS Convenient OCR, Envision, Seeing AI, or Be My Eyes to read, try uploading it to Google Gemini and entering the following prompt. "Perform OCR on the document and provide the full text without altering it." Including the part about providing the text without altering it in the prompt is important, so that Gemini doesn't editorialize or summarize the document. I tried this with two forms filled out by hand by doctors, so you know their handwriting wasn't the greatest. While the results from Gemini weren't perfect, the forms were at least readable and understandable." Thanks so much, Catherine.
I am using Gemini for a lot of this sort of thing, audio transcription, reading handwritten documents or difficult documents, and one of the things that we've done recently, it's actually just been Bonnie's and my wedding anniversary. Isn't it great? We got married on the same day. And we had a restaurant that we were going to, there were a couple of menu documents. They were separate documents, and they were untagged, not nice PDF documents that were just images, essentially. And what I wanted to do was just have a nice well formatted Word document with heading levels, and I gave the whole thing to Gemini. I uploaded both documents, I said, "I want you to create a full menu starting with the starters, and going on to the salads, and the entrees, and then add the second file, which is the dessert menu. I want everything marked up at heading level. I want everything described linearly and don't summarize the document. Do not exclude anything."
And it came back with a beautifully formatted Word document, perfect semantic structure. So, all the major sections, starters, salads, entrees were heading level one, each dish was a heading level two. Absolutely fantastic. So, yes, So you can use AI for these things and a lot of the time it will do an outstanding job. That concludes this episode of Access On, the technology podcast of the National Federation of the Blind. To send in a contribution for a future episode, email us, attach an audio clip, or just write it down and send it to [email protected]. That's [email protected]. To keep up to date with Access On, follow us on Mastodon, [email protected]. That's [email protected] on Mastodon. To subscribe to an announcement-only email list about upcoming episodes, send a blank message to [email protected]. That's [email protected].
To learn more about the National Federation of the Blind, visit our website, NFB.org, or phone us, 410-659-9314. That's 410-659-9314, and be sure to check out the Nation's Blind Podcast right from where you heard this podcast.