Welcome to the second episode of Access On, the National Federation of the Blind's Technology podcast.
Episode
Listen to the second episode of the Access On podcast (Browser).
Or listen on your preferred podcast platform.
Timestamps
The show is segmented by chapter, making it easy to move between segments of the podcast if you have an app or player that supports chapters. Below is what's on the show this week, and when you can hear it.
- Introduction, follow us on Mastodon 0:00
- Samsung working on Meta Smart Glasses rival 1:52
- Blindshell releases the Blindshell Classic 3 2:33
- Firmware update for the BT Speak 4:25
- Ads coming to Threads 5:16
- Interview with Google Gemini on its dedicated iOS app 6:06
- Whatsapp has added drafts 10:49
- Upcoming CENA boutique on teaching someone to use a screen reader 11:05
- What will Apple Intelligence mean for screen reading? 12:27
- How is Apple's Visual Intelligence going? 15:42
- Getting descriptions of videos using Seeing AI and PiccyBot 17:27
- Mike Buckley, CEO of Be My Eyes, discusses the Ray-Ban Meta collaboration and other partnerships 28:18
- Latest episode of the Nation's Blind Podcast 1:04:01
- Tech tip, reading complex emails in Microsoft Outlook 1:04:46
- Closing and contact info 1:06:55
Transcript
Speaker 1: Live the life you want.
Jonathan Mosen: I'm Jonathan Mosen and welcome to Access On, the Technology Podcast of the National Federation of the Blind. Be My Eyes comes to the Ray-Ban Meta Smart Glasses. We discuss this and much more with Mike Buckley. If you prefer a smartphone with buttons, BlindShell has just released the BlindShell Classic 3.
We give you a quick preview of Google Gemini on the iPhone, by interviewing Gemini itself, and we briefly explore a couple of options for getting description of your own videos right from your smartphone. Good to have you here for episode two and thank you so much to everybody who has provided such positive and warm feedback on our first episode. We're glad so many of you have found us.
I mentioned last week that you can be in touch with us via email by attaching an audio clip or writing the email down and sending it into Accesson, all joined together, @nfb.org and many people have done that this week, which is wonderful. We will go through a selection of some of the emails that we have received a little later and
I'm also pleased to let you know that Access On is on Mastodon, you can follow us there, [email protected]. And I'd encourage you to check out the nfb.social instance, for a range of National Federation of the Blind related Mastodon accounts, including our primary Mastodon account at [email protected]. But do follow us, [email protected], we do respond to messages there.
We're getting a bit of conversation going about various accessibility initiatives that we are pursuing. So by all means, check us out there and join in the conversation.
Now it's time for this week's news bits. Credible sources from supply chains say Samsung is working on new smart glasses, slated for release of this stage in Q3 of 2025, known as the XR glasses. They'll feature the same kind of chip, battery and camera, currently available in the Ray-Ban Meta Smart Glasses, which have become a huge hit in the blind community. Unlike the Meta Smart Glasses, which understandably use Meta AI, Samsung's XR glasses will use Google's AI model, Gemini.
They'll support paying for products right from your glasses, QR code recognition, gesture recognition, reminder and translator functions. If you want smartphone features, but don't want to use a touch screen, BlindShell has released the third generation of its popular line of smartphones. The company says BlindShell Classic 3 comes in two new colors and starts much faster. There's more RAM, 3 gigabytes on this device versus 2 gigabytes in the previous generation and the phone apparently feels more responsive. Internal storage is up from 16 gigabytes to 32 gigabytes and in something of a philosophical departure, users can now install third-party apps from the Aurora app store.
When using those apps, you have the option of using the phone's physical buttons as you always have, but now you can explore by touch when the screen becomes a touchscreen, running third-party apps. The Classic 3 has a new AI assistant, it's called Luna. You can ask Luna for facts, conduct searches on the web, open apps and ask it to complete various tasks for you.
BlindShell has created its own proprietary messenger program called Tandem. The Tandem application allows you to communicate seamlessly with friends and family members, even if they don't have a BlindShell phone. When the other person installs the Tandem app on their iPhone or Android phone, you'll be able to call, message and communicate with them directly. Tandem supports voice calls, video calls and text and voice messages. All your data is encrypted while using Tandem according to the company and using Tandem, you can remotely control a BlindShell Classic 3.
If you need technical assistance, a BlindShell Classic 3 iPhone or Android user can take control of your mobile phone with your permission to assist you. There are plenty of other new features. Access On has reached out to BlindShell. We hope to have them on the show to explore the product in the near future.
Blazie Technologies has released the November 2024 update for the BT Speak. It's free for all BT Speak owners and it includes many new features, such as putting the Getting Started audio tutorial into smaller sections, determining the level of punctuation you'll hear when reading in a file. Improved navigation when reading by character, support for streaming YouTube videos in traditional mode and accessing files on Google Drive.
The revamped audio tutorial produced by Stephen Clower is divided into 21 sections and in addition to being available on the BT Speak itself, is available for free download from the guides and media page of the Blazie Technologies website, that's at blazietech.com. So this is a good way to explore the product in depth before you purchase and decide whether it's for you.
Threads, Meta's answer to X, will soon introduce ads to the timeline, potentially creating clutter for screen reader users. Sources say a team within Instagram's advertising division is leading the effort and a small number of ads could start appearing as early as January. It's unfortunate timing for this news to have leaked, because Bluesky, another X alternative is having another growth spurt. The company says it has surpassed 15 million users and for a time the app was the single most downloaded app in any category on the iOS App Store. The service has accumulated 2 million users in the space of just one month. By comparison, X still has 500 million active users and Threads has 275 million monthly active users.
Google has finally released a dedicated iPhone app for its Gemini AI service and you can have a conversation with it, using a feature called Gemini Live. And since we can talk to it, I thought we'd interview Gemini right here on Access On and find out what all the fuss is about. Gemini, it's a pleasure to talk with you. Welcome to Access On, the National Federation of the Blind's Technology Podcast. I appreciate you doing this.
Gemini: It's my pleasure to be here. I'm excited to talk about technology with you and the listeners. Let's get started. What would you like to talk about first?
Jonathan Mosen: I want to talk about you actually. Tell me why the release of the Gemini iOS app is such a big deal and what can the iOS app do, that you couldn't do on the iPhone before with Gemini?
Gemini: So the Gemini iOS app is a pretty big deal, because it puts a dedicated Gemini experience directly into people's iPhones. Before you could only really access Gemini through the main Google app or a website. Now there's a whole app focused on getting the most out of Gemini's abilities.
The new app has a couple of cool things going for it. First it's really easy to use, you can just tap to start talking to Gemini Live, which is kind of like having a conversation with a super helpful friend. Second, it can create images, which is something that wasn't really possible before on iPhones. And finally, it works seamlessly with other Google apps. So if you're looking at something in Gmail or Maps, you can ask Gemini about it directly without having to switch apps or anything like that.
Jonathan Mosen: Listeners can choose a range of voices, I understand, according to their preference.
Gemini: That's right. We've got a whole bunch of different voices to choose from, so you can pick the one that you like the sound of best.
Jonathan Mosen: Can you tell me about accessibility for blind people who will be listening to this podcast and using voiceover? Is the app fully accessible with voiceover technology?
Gemini: Yes, accessibility is really important to us. The Gemini iOS app is designed to work well with voiceover, so blind users should be able to navigate it easily. For example, you should be able to understand what's happening on the screen, move around the app and use the different features with voiceover turned on.
Jonathan Mosen: At least for iOS users, ChatGPT has a bit of a head start, because they've had these premium features available for a while and I talk to ChatGPT a lot. What would you say the major strengths were of Gemini over ChatGPT?
Gemini: This is a tough one, since both Gemini and ChatGPT are really strong language models, but I'd say Gemini has a few key advantages. First off, Gemini is better at understanding context. It's able to keep track of what you've been talking about, even if you switch topics for a bit, this makes conversations feel more natural and less robotic. Gemini's also got a knack for generating different kinds of creative content, you can ask it to write poems, scripts, emails, or even code, and it does a pretty impressive job. Finally, Gemini's really good at connecting with other Google services, so if you're looking at something in Gmail or Maps, you can ask Gemini about it directly. This can be a real-time saver.
Jonathan Mosen: Are you able to tell me what's on my Google Calendar for Monday, for example?
Gemini: So I can't directly see your Google Calendar yet, when we are live, but this is something that will be possible in the future.
Jonathan Mosen: Well, we'll look forward to future developments. Thank you for coming on Access On. Got anything to say about the National Federation of the Blind? What do you know about us?
Gemini: The National Federation of the Blind, it's a great organization doing really important work. I know you're dedicated to empowering blind people and improving accessibility. You're also a strong advocate for things like braille literacy and pushing for better technology that works for everyone.
Jonathan Mosen: All right. I appreciate it, Gemini, thanks so much for coming on the show.
Gemini: Thanks for having me. It's been a pleasure chatting with you.
Jonathan Mosen: And finally, on news bits for this week, WhatsApp has unveiled the ability to save messages as drafts, so if you get interrupted while composing a message, you can return to it later. Drafts appear at the top of your list of chats and that's news bits for this week. This is the last opportunity I'll have before it happens, to tell you about the webinar for November from our Center of Excellence in Non-Visual Accessibility, CENA.
And the webinar this month has a slightly different emphasis from many of our webinars and it's a really important subject. It's focusing on how to train friends and family on a screen reader. Learning access technology can appear daunting to a newly blind person. Finding quality access technology training can also be difficult for several factors and this is where technology savvy friends and family can help someone start learning how to use a screen reader.
Join the National Federation of the Blind on Tuesday, November the 26th from 2:00 PM till 3:30 PM Eastern, to learn how you can teach a friend or family member how to get started with a screen reader. We'll cover how a screen reader works, how screen readers present content, common keyboard shortcuts to start with further training resources to learn how to use multiple screen readers and more besides. To find out more about this particular boutique webinar or others that we are holding in future, go to nfb.org/cena, that's nfb.org/ C-E-N-A, and we look forward to seeing you there on November the 26th.
Let's step into the Access On mailbox. If you'd like to write to us or send us an audio attachment, you can do that, [email protected]. This email says, "My name is Angel, though I often go by Farron. I'm writing to express my excitement about your new podcast, Access On, which I've been looking forward to. I used to listen in on your former podcasts, Mosen At Large and Living Blindfully, and I've always appreciated the insights you share. I currently use an iPhone 12, but I'll soon be upgrading to the iPhone 16 Pro in desert titanium, and I'm particularly excited about the Apple Intelligence feature.
As a voiceover user, I'm curious about what I can expect when using Apple Intelligence for the first time. Do you have any insights into how it might improve the accessibility experience? I'm especially hopeful that Apple Intelligence could enhance voiceover, making it even more responsive and capable of understanding on-screen content. Could there be a possibility that Apple Intelligence will help make voiceover more intuitive or adaptive to individual user needs over time?"
There you go, Angel, you can say yours was the first ever email from a listener, read on Access On. Isn't that cool? I think that screen reading in general is due for a bit of a shake up, thanks to artificial intelligence and Apple Intelligence and we'll see where that goes. We certainly know that there are users who have difficulty remembering the myriad of commands that are necessary to operate a screen reader and it might be possible for something to be developed that's very different from the current screen reading paradigm that makes screen reading easier for that group of people. That has to be done of course in a way that doesn't compromise the efficiency of those who do work well with screen readers and want to get the job done with the keyboard commands they know and understand.
One thing about Apple intelligence that is interesting and potentially exciting for blind people, is that it can be made aware of what's going on on screen. And that means that you will be able eventually, not with 18.2, but soon thereafter, to ask Siri questions about what's on the screen. Siri’s slated to become more powerful in forthcoming releases, and already in the 18.2 Beta, we're seeing Siri answering questions powered by ChatGPT. It is going to get even more on-device smarts over the next little while as well.
Maybe they'll build on the screen recognition feature, which essentially performs optical character recognition on the screen and that can help with inaccessible apps. Sometimes it can be just one splash screen that's causing a problem and if you can get past that initial splash screen at the beginning of an app, you're okay and certainly that screen recognition feature has helped in that regard. Maybe when iOS 19 comes out and Apple Intelligence is well baked into the devices by then, we will find some interesting use cases for voiceover that creep in.
That's what makes this field so interesting. There's so much change, so much innovation going on. We're never quite sure what's coming next, but we do know that Access On will be here to cover it and try and make sense of it all. And this email has a similar theme. It comes from Marissa and she writes, "Hello, Jonathan. First and foremost, congratulations on your new role within the NFB." Thank you Marissa. She says, "I am very happy that you are also back and doing podcasts. I wanted to ask, if I remember correctly, I believe you had gotten an iPhone 16 Pro or Pro Max?" That's correct, I'm running a 16 Pro Max at the moment. "Are you beta testing iOS 18.2 at this time?" Yes, I am, Marissa. "Have you tried the visual intelligence feature? How does it work with voiceover or does it still need fine-tuning?"
It's working pretty well, Marissa. We are going to do a deep dive into iOS 18.2, either just before or soon after its release. So we'll cover the visual intelligence feature then. But I will say by way of a teaser, that my experience at the moment with it and it is likely to keep evolving, is that it's essentially just giving visual descriptions of things either with ChatGPT or you can look something up, you can look an image up on Google and then get information back from a Google search for the image.
So it's early days, but it is accessible and it is reasonably promising. I don't think the prompts are of course fine-tuned for blindness specific use cases in the same way that Access AI from Aira, or Be My AI from Be My Eyes have done. So I don't know whether you'll get the deep level of detail and nuance that you would from those apps. But yes, it does work and having it so conveniently available on the camera control button, which is available on all iPhone 16 phones is certainly handy. So we'll watch it develop.
Jim: Hello Jonathan, this is Jim from Florida. It's great to have you back and also working with the folks at NFB, Access On is great. I wanted to first ask you about, we've got a new way of doing audio description. I had heard a demonstration of using Seeing AI to listen to what I would call real-time audio description of a video someone had created. And I use an iPhone and so of course I got all excited and started playing with my iPhone 13 team, and learn from talking to Apple Accessibility that the format that Apple uses when you or others create videos on their iPhones is not compatible, and so I've been trying to work with Apple to get the right format. Wanted to know if anyone knew how to do that, if there's an accessible app to convert files created on iPhones or any type of digital recording device to that format?
Jonathan Mosen: Glad you're enjoying Access On, Jim, and thank you for getting in touch. For those unaware, this is a new feature that Jim is talking about that arrived in the most recent build of Microsoft's very popular Seeing AI app. In the specs it says that it only works with MP4 videos and even though iPhone doesn't record in that format, I have had no difficulty at all sharing my videos with Seeing AI and getting an audio description of those videos. So I can only conclude that when you share it with Seeing AI, there's some sort of on-the-fly conversion going on that we don't have to worry about, because it's working perfectly well. Let's go through how we do this. I have the Seeing AI app up and running now and I'm going to go to the bottom of the screen.
Speaker 5: Channel, short text, adjustable.
Jonathan Mosen: I'll flick up.
Speaker 5: Document, product, scene preview.
Jonathan Mosen: Now I'm on the scene channel which is in preview. I'll flick left.
Speaker 5: Browse photos button.
Jonathan Mosen: And right there is the browse photos button. Perhaps it would be a little less ambiguous if they renamed that now that this does the audio description as well, but that's where you need to go. So I'll double tap.
Speaker 5: Seeing AI, back photo, October 18th.
Jonathan Mosen: What I also like about this is that you have a clear list of photos and videos in reverse chronological order. I know that I did some videoing on the 13th of October. I remember this very clearly, because that was when we had a ride-share refusal from Uber. So I'm going to flick right.
Speaker 5: Photo, video, photo, video, photo, photo, photo, video, October 13th, video, October 13th, video, October 13th.
Jonathan Mosen: I took several videos as a result of this and I should say that this video was actually taken on my Meta Smart Glasses, but the principle is the same and I have been able to confirm that videos taken on my iPhone are also working for me. I'll double tap.
Speaker 5: Expand video, date, Sunday, October 13th, 2024, 07:00 PM.
Jonathan Mosen: I'll flick right.
Speaker 5: Toolbar, describe button.
Jonathan Mosen: And double tap the describe button.
Speaker 5: Cancel button, processing. Please wait while descriptions for the video are generated. This may take several minutes, depending on the length of the video.
Jonathan Mosen: They are not wrong, it can take a long time for the description to actually occur. I guess it is uploading the video to the cloud, it's then analyzing the video and giving you a description. Now what happens with this particular implementation that Seeing AI has done, is that it pauses the video, gives you the description of what's going on at that particular moment and then it resumes the video.
That's a different way of doing it from PiccyBot, which we'll cover next. This is handy obviously for reliving moments that you want to, but also where I find it quite interesting is going through my Ring video doorbell footage and having it described video from that to me. I'm going to pause the recording, because this may take some time and then we'll resume the recording when we get a description.
Speaker 5: A street scene at night with a parked car and reflections of colorful lights on the wet pavement. The car door is open and someone is getting into the vehicle. Two people are standing by the car appearing to have a conversation. The camera pans to show an empty street with reflections of lights on the wet ground, a close-up of the wet pavement with reflections of colorful lights. The scene remains focused on the wet street with no visible people or vehicles. The camera continues to show the empty street, highlighting the reflections.
Speaker 6: Unfortunately, if you don't take her, you'll be debarred from-
Speaker 5: The scene remains unchanged, focusing on the wet pavement and reflections.
Speaker 6: ... the platform and you won't get any more work.
Speaker 5: The camera slightly shifts, still showing the wet street and reflections.
Speaker 6: It's the law.
Speaker 5: The scene continues to focus on the wet street with colorful reflections. The camera pans slightly, maintaining the view of the wet street. A view of the street with a large advertisement board visible in the background. The advertisement board is more prominent, with the wet street in the foreground. The camera focuses on the advertisement board and the reflections on the street. The scene shows the parked car again, with reflections on the wet pavement. The camera remains on the parked car and the colorful reflections on the street.
Jonathan Mosen: Now we don't get the rest of the conversation with that particular video, because I double-checked that I was videoing and started a new one, but it is really bringing videos to life for blind people. That's the Seeing AI way of describing videos. There is another app that's pretty popular in our community and it's called PiccyBot, I believe it is developed by the same person who did the Google Goggles app that was also very popular some years ago. It handles this situation a bit differently and I'll show you what happens there. I'm in the PiccyBot app now and if I go to the top of the screen.
Speaker 7: Settings button.
Jonathan Mosen: We have settings. This has a range of options in terms of the way that it describes video. Some of it is very, very chatty and informal, that's not, as my kids would say, my jam. So I've made the descriptions pretty straightforward, but if you prefer a little bit more flamboyance in your descriptions, you can do that as well. And if I flick right.
Speaker 7: PiccyBot heading, query PiccyBot with your question, text field. Welcome to PiccyBot. Take or choose a photo or video and let PiccyBot speak out the description for you. Video button. Video gallery button.
Jonathan Mosen: That's what I want. I'll double tap to go through the videos I've already taken.
Speaker 7: Toolbar, sort and filter button.
Jonathan Mosen: We'll go to the top of the screen.
Speaker 7: Cancel, selected videos, collections, search your library, microphone, video date, October 15th, video date, October 15th, video date, October 13th, video date, October 13th.
Jonathan Mosen: In this case, I'm going to choose the middle video, which is where the action really starts. Now what happens with this is that you don't get to hear the dialogue with the description. It doesn't pause in the same way that Seeing AI does, but you will get a real feel for what actually took place with this ride-share refusal. So I'll double tap.
Speaker 7: Choose video heading, movie scrubber, current position, zero seconds, cancel, but play button, choose button.
Jonathan Mosen: And we can play the video, just to confirm that this is the video we think it is, I'm confident it is. So I'm going to double tap the choose button.
Speaker 7: Choose.
Jonathan Mosen: And now let's look at what's on the screen.
Speaker 7: Choose video heading, settings button, PiccyBot. Query PiccyBot with your question. PiccyBot heading, what is in this video? Dimmed video, uploading video button. Video gallery, but camera, but photo gallery, photo gallery.
Jonathan Mosen: The key thing that was uploading.
Speaker 7: Cap with video, analyzing video, analyzing video.
Jonathan Mosen: Now it's telling us what the status is and soon we will get a description. You can hear the little indicator there that it's processing. And again, I will pause the recording until we have some action.
Speaker 8: Here's a description of the video. The video opens with a nighttime-
Speaker 7: Video.
Speaker 8: ... rain soaked city street scene, two men are talking near a silver car. The first man appears to be in his fifties with short light brown hair and glasses, he's wearing a dark jacket and carries a smartphone. The second man is slightly younger with longer dark hair, wearing a dark-colored patterned shirt. The camera follows them as they walk toward a building, identified as Rydges Wellington. The scene then switches to a view of the silver car, which seems to be a taxi parked on the street. The two men are seen talking near the driver's side door, the younger man seems to be explaining something to the older man who listens intently. The conversation centers around the older man's refusal to take a guide dog into his taxi.
The younger man insists that by law the driver must transport the guide dog. The older man argues there's an Uber assistant option to bypass the necessity of bringing the dog. The younger man counters by firmly stating that the presence of a guide dog is non-negotiable, citing legal requirements and threatening to call the police if the older man refuses. The older man seems hesitant, frustrated, and increasingly annoyed by the younger man's insistence. The younger man appears firm and unwavering, calmly, but resolutely reiterating his point about the law. The video then shows the younger man using his smartphone to potentially make a call to Uber support. The camera angle shifts to provide views from inside the taxi. The older man seems reluctant, but eventually agrees to take the guide dog.
Jonathan Mosen: I'll stop it there, because I don't particularly want to relive any more of that, but that's an example of how PiccyBot does it. I am using the premium version of PiccyBot. Seeing AI has a limitation of five minutes per video and if you do go the full five minutes, it is going to take quite a while to process that video. So those are just a couple of options. If anyone has any tips and tricks to share on describing videos that are on your smartphone, by all means be in touch. [email protected] is my email address, you can attach an audio clip to the email or you can write it down.
The Ray-Ban Meta Smart Glasses are the talk of the blind community at the moment. Large numbers of blind people are purchasing these stylish glasses at an attractive price point and in a future episode I want to feature your feedback about your daily use of the Ray-Ban Meta Smart Glasses. But if I were to try and summarize the feedback that we are receiving, it's that people appreciate how good they look and how comparatively little they cost.
Being able to take pictures with the glasses is convenient, particularly when one of our hands is already taken up with a cane or a guide dog harness when we're traveling. It's easy to snap a picture for posterity or for posting to social media. And when my wife and I were recently refused a ride in an Uber because of her guide dog, it was quick and easy for me to video the incident, get a clear picture of the license plate in the process and send that on.
We can access Meta AI with the glasses and while people appreciate that, there has been some criticism that to obtain the detailed answers a blind person requires, you have to play 20 questions with it or be extremely precise in your questioning. Getting it to read an entire document can be done, but it's a bit hit-and-miss. It won't describe people at all, even though a sighted person has access to this information simply by looking. But the general consensus is that the pros far outweigh the cons and that it's likely that these glasses will become increasingly capable.
I'd be most interested, as I know with many other listeners, in your experience with the Ray-Ban Meta Glasses, if you have them and I'd be delighted if you'd record an audio clip or write something down and send it to [email protected]. That's Accesson, all joined together, @nfb.org. Then there's the ability to receive sighted assistance through the glasses. One way and the most seamless and integrated way to do that is rolling out now and that's with the official integration of Be My Eyes with the Ray-Ban Meta Smart Glasses. To talk with me about that and other happenings at Be My Eyes, I'm delighted to be joined by Mike Buckley, he's CEO at Be My Eyes. Mike, welcome to Access On.
Mike Buckley: Thank you so much, Jonathan. Congratulations on the new gig and the new podcast. Excited to chat with you today.
Jonathan Mosen: Thanks so much. I really look forward to working with you. We value the relationship that the National Federation of the Blind and Be My Eyes have. We will have listeners who aren't aware of all the practical details. So talk me through what you can do with Be My Eyes on the glasses today and where it's available and how you get it working.
Mike Buckley: Basically there's a seamless integration now, where you can connect to a Be My Eyes volunteer through a verbal prompt and it's rolling out initially in the US, Canada, UK and Australia. It's fast and it's easy and it's a simple download when you download the Metaview app. And so we do anticipate that this will be rolling out to other countries all around the world in the future, but literally after you've set up your glasses, if you say, "Hey Meta, Be My Eyes." You'll be connected with one of our 8.1 million global volunteers in your language of choice and in your time zone of choice.
For example, if you are in the UK and it's 3:00 in the morning, we're not going to connect you with a volunteer geographically in the UK because we'll wake them up, but we'll connect someone who speaks your language in a part of the world that's up. So it's 24/7, just like our volunteer services as well and again, it's live right now.
Jonathan Mosen: I understand that it is necessary to go into a new part of the Ray-Ban Meta app called Accessibility settings and link your Be My Eyes account. Is that correct?
Mike Buckley: Yes, that's right. We have community help pages up on how to do that, accessible of course, as does and Meta has some information on how to do that as well.
Jonathan Mosen: We are recording this just a few days before we publish this episode and what I'm seeing on Mastodon at the moment is that some people have this accessibility settings option, but others don't. Is it a staggered rollout?
Mike Buckley: Yes, I believe as of Thursday in the United States it was at 50% and so I think we'll be complete in the near future.
Jonathan Mosen: How does the transition between smart glasses and phone camera work? I imagine there'll be some situations where it's better to use the glasses, but then you may find, okay, I've got an even more powerful camera in my pocket here and I've found a use case where I need to switch.
Mike Buckley: Yes, you can switch back and forth and it's relatively seamless. The feedback we've gotten from the beta testers was, to your point, that that was a really important functionality, because there may be certain features and power that your phone camera has, or it may be better for certain things than your glasses' camera. So you are able to switch.
Jonathan Mosen: Can you elaborate on those certain things? So when might I get a better outcome from my phone rather than the glasses?
Mike Buckley: It's an interesting question and I will tell you, Jonathan, what I've heard directly from our users, but I think you'll probably hear a lot more specificity from the people that you talk to that use them. Sometimes you may want a wider angle view, you may want something that's slightly better on optical character recognition, or the ability to narrow or widen a field of view.
So it's really going to be a little bit of a trial and error. What we heard initially in the early days of beta testing was that for, say reading out a printed page of words, sometimes it required you to hold that quite close to the glasses, and so there's a little bit of trial and error. And so I really think it's going to be a lot about individual user preference on those things.
Jonathan Mosen: One thing I have noticed so far is that you have to get used to the fact that when you look at something with these Meta Smart Glasses on, the camera is not dead center, it's slightly, I think to the left of your field of view. So you've got to get used to the fact that if you want to look at something directly, you're going to have to look slightly right.
Mike Buckley: I think that's true and I think it's also one of these things where any relatively new piece of technology is going to have its nuances. The way we think about it is, as of today, this is the worst this experience is ever going to be, it should only get better from here. And I think Meta is already aware of some of those nuances, with respect to the camera being on the left and some of the optical character recognition issues as well. And I can assure you that they're working very hard on future development and iterations.
Jonathan Mosen: I recently sat in on a call where I heard several Be My Eyes users talking about their personal experience of having the service on the glasses, there was a lot of enthusiasm. What kind of feedback have you received about the difference that it makes, having Be My Eyes on a wearable like this?
Mike Buckley: It's significant, Jonathan, and I've gotten a remarkable amount of lovely feedback. Now I've also gotten some criticisms on things like how it could be better, but the overwhelming sentiment that has been expressed to me is that it fundamentally alters the power of the user. I don't know, just think about setting up your router at home, or being able to work on a keyboard as you're getting visual assistance.
I think it dramatically changes the game and the freedom and the power of the person using the technology and that's what really excites me. You and I have talked about this, I don't know, a dozen times about you ultimately want to design anything with the community and not for the community. And so the hundreds of beta testers that we had, putting this product through its paces before it was released really helped us make it better. But all I can say is the overwhelming sentiment that I'm getting is that it's a lovely experience and it should be more power.
Jonathan Mosen: In your initial press release you referenced people working with Be My Eyes at airports, and you and I have discussed this separately. I must admit it stopped me in my tracks when I read that, because I hadn't considered using Be My Eyes at an airport before. My rationale was that the quality of volunteers may vary, I don't know how much time they have available to help me out. And then when you ask someone for directions, the answer you get back is extremely variable. Some people will give you great directions and other people will say, "It's over there and they can't translate that any further." How common is it for people to use Be My Eyes in this sort of situation and what are users telling you about that experience?
Mike Buckley: It happens every day in an airport, it happens every day in a hotel, it happens every day in various settings. I guess the way I think about it, Jonathan, I'm never going to claim perfection. Sometimes there's going to be a back and forth between a member of the blind and low vision community and a volunteer, to get better, more specific, more relevant information, or even the way something is described.
But the way that I genuinely think about this is there are very few limitations on human kindness and these interactions that happen, of which by the way more than 90% are successful, based on the feedback that we get, is there's just a human back and forth that works. And the proof of that is we do well over a million calls a year, the calls are increasing, despite the fact that we've introduced AI to handle maybe some more basic needs, but it just works, not perfection, but it works well.
Jonathan Mosen: Do you connect any data on how often in an environment where people are being asked for directions it works and where perhaps somebody has to call back and try another volunteer?
Mike Buckley: Yeah, I don't have that specific data. It's not something that we collect where, we don't know if you're asking for help in an airport at this moment or somewhere else. What we have is generalized call data that says successful or not successful, and allows the community to rate the calls. And those calls are rated at better than 9 out of 10 effectiveness and so it's a more generalized stat as opposed to something specific. But I will tell you that 7 to 10%, quote-unquote, failure rate that we have, probably 75% of the time as a result of connectivity issues, where there's either weak internet, or the call breaks, or something like that.
Jonathan Mosen: We'd be most interested in feedback from listeners who are using Be My Eyes in that use case as well, just to tell people about good experiences or otherwise that they have had. One criticism I failed to mention at the beginning, that I should have I think, is the battery life of the Ray-Ban Meta's. People would love to see that go longer and that's a huge engineering challenge, because of the form factor. Have you got any tips in terms of maximizing battery life for Be My Eyes use? And indeed, what expectations should a user have about what battery life is available when you're using it a lot for Be My Eyes cause?
Mike Buckley: Great question. You're right, there is a limitation. Most people are saying somewhere around two hours, maybe a little bit more, a little bit less, depending on how of use. And so I don't have a lot of great tips on extending that battery life, because there are limitations of the technology and as you mentioned, that's a combination of both power and miniaturization. I do think that over time that will improve, but I'd be curious if folks have tips and tricks as we go through this that they use.
While an expensive proposition and I don't want to suggest it's easy to do, I've talked to a few users who have said that they've bought two sets of the glasses, so that they can rotate them back and forth and basically keep them on for longer periods of time. Now look, that makes your purchase 600 dollars rather than 300 dollars, but I will say relative to the cost of other assistive smart technology, it's better.
Jonathan Mosen: And for those not familiar with these glasses, they do come with a charging case and that doesn't have to be charged that often so you can pop them in when you're not using them. But the dilemma I have is that I like to wear these, because you never know when there's a moment where the glasses may be handy and you just want to take a picture, or use the AI. And even wearing them with the glasses switched off, I suppose is a middle way, but it does take a few seconds to power them back up again. So it is a bit of a dilemma.
Mike Buckley: But Jonathan, as a technology enthusiast, I would bet you share my optimism that it's going to get better.
Jonathan Mosen: Oh, sure.
Mike Buckley: It's like the iPhone battery life has improved over time and this will happen, but I share your enthusiasm for wanting to just keep them on.
Jonathan Mosen: Yes. I mean, for example, we didn't obviously know that we were going to have an Uber refusal, although I suspect that it was possible, because it happens so often. So in that case, I was ready with the glasses powered on, powered up, and as soon as the refusal started to happen, I just started videoing.
It was absolutely fantastic to know that the whole incident was being captured, that the driver couldn't deny what had gone on and I'm still in New Zealand at the moment and so I was able to send it to our regulator here and say, "Here you go, what else do you need?" And it was a slam dunk, so having them on and ready to video, you just never know when you might want to do that, or for that matter, just say that magic word and get a Be My Eyes volunteer on the call.
Mike Buckley: Yeah, exactly.
Jonathan Mosen: There is no Be My AI on the glasses and I wonder if there ever will be, because you're using ChatGPT and Meta has its own AI, so it would effectively be giving license for a competitor to be on the platform. The AI is often considered the weakest link at the moment. Will we get Be My AI on the glasses at all, or at least AI that provides more detailed information?
Mike Buckley: On the latter, will we get AI that provides more detailed information? Almost certainly, this is the worst version of the AI on the glasses that's ever going to be, it's going to get better. Ultimately, I would imagine that you will be able to set it up with settings that work for you as an individual user, if you want more language, or less language, or shorter descriptions and things like that. I think that Meta has initially optimized the AI model for speed and short bursts, whereas I think Be My AI, the reason most users really like that is because of the robustness of the description.
And so very openly we're talking with Meta about what it would look like to put Be My AI on the glasses. There's no resolution there, but the thing that gives me optimism is twofold. One is that we've agreed to a deep development partnership with Meta, this isn't just a one and done thing. We're iterating together, our engineers are sitting in rooms together and so that I think is exciting. And the second thing is that the Llama model will continue to improve a great deal. You and I laughed when we talked last about how when GPT-4 was initially launched in whatever that was, March or April of 2023. We all loved it, but if we had that experience now, we would be complaining about it because of how limited it is compared to what the capabilities are now. And I suspect that Llama is going to follow a very similar path.
Jonathan Mosen: I mean, I agree with all of that, but I also think there's something else going on. I use the subreddit for the Meta AI Glasses and I find it very good, because it's just a mainstream forum. It is not specifically for blind people and they have a lot of pretty clever engaged people playing with these glasses. Someone managed to reverse engineer the prompt that these glasses are receiving and it's being deliberately asked to be terse. I imagine that's to save tokens or save bandwidth or whatever. So the Meta AI that we have now is absolutely capable of more detailed descriptions, it's just that we don't have a way of getting at them without the 20 questions thing.
Mike Buckley: I think that that's plausible, Jonathan, and I don't have any specific information, but I do think it is optimized to be terse and quick. But ultimately, Meta is smart and if there is a user desire to have a particular experience with the AI, it would shock me if they didn't have in their roadmap, ways to satisfy multiple user preferences there, it just makes sense for them.
Jonathan Mosen: It would, and there may be a number of ways of doing it. For example, if you link a Be My Eyes account, that's clearly a signal that you are a blind person and that information could be used to trigger more verbose descriptions.
Mike Buckley: It's like you were in our engineering team meeting last week. We literally talked about that, because there's knowledge about when you link accounts and maybe you optimize the Meta experience for that individual based on more robustness, but literally that's something we're talking about right now.
Jonathan Mosen: That's very good to hear. Another approach, perhaps a bit more controversial is that screen readers can present themselves, so it's possible for a developer to know whether TalkBack or voice over is running.
Mike Buckley: Yeah. What's neat about almost all of these models is when you do interact with them, you can sort of prompt them and even overcome some of the other system prompts. So if you say, "Can you give me much more detailed information," or ask those follow-up questions, generally that works. But I think the point you're making is, why not optimize for user delight and user desire at the front end?
Jonathan Mosen: Right, because obviously efficient access to information is important for many of us. And then there's the question of drawing public attention to yourself. Bonnie, my wife has commented on this actually, that when she's talking to the Meta AI and she's in the street, or at the supermarket, or something like that and she's trying to get the information she wants, she attracts unwanted public attention. People come over and start interceding and trying to help her, because the Meta AI hasn't helped her quickly enough.
Mike Buckley: I hear that and it's a real issue, but it has not dampened my enthusiasm or excitement about where we are. I mean, just think about where we've come in 18 months, Jonathan, and I can't imagine the capability 18 months from now. But you're right, a lot of what we think about at Be My Eyes generally is, okay, these models are going to be everywhere. They probably will be linked with almost every device we have, whether it's your Amazon Echo, or the glasses, or even your computer or laptop, which I think in the not too distant future you will be able to talk to get it to do things.
The real question I think for the community that we work with is how to really make it optimal for that community and with that community. And so we are never going to beat OpenAI, or Microsoft, or Meta on model power, because they have billions of dollars and resources to put into training that we'll never have. I think where we can make sure that we fight, advocate and win for the community is on those optimizations that make the experience better.
Jonathan Mosen: Absolutely. And the National Federation of the Blind shares that view, good quality products are made when blind people are around the table directly, at the conception stage and we can make these products work. So we are also optimistic about the future, but we have to make that future, we have to actively advocate and intercede and ensure that that future happens.
Mike Buckley: Yeah, we got to be loud sometimes and whether that's the Uber protest that happened last month, or just keeping our foot to the pedal on trying to get all companies to do the right thing here. I don't think that battle's going to end anytime soon, but it's one I'm happy to continue fighting.
Jonathan Mosen: And it doesn't necessarily always have to be a battle, it's just making sure that the conversations continue to take place. And I must say that we are having some very positive conversations with Meta at the moment, so I'm certainly not suggesting that the conversations aren't taking place. So it is exciting, but I think one of the things that we are trying to do on this podcast as well, is point out where things are going well and where there's room for improvement, because it's not our job to sell this technology. So there are some improvements to be made, but I'm also optimistic that they will be.
Mike Buckley: I agree. I obviously have the lens of our own business here that colors my opinion of these things, but the number and depth of conversations that we are having on a daily basis right now with enterprises is tenfold what it was a year ago, which excites me.
Jonathan Mosen: For those who are interested in, I guess more of the technical side of this, and we do have some very tech-savvy listeners. My understanding is that while Meta is embracing the success of this product in the blind market, which is wonderful to see, they didn't initially perceive the glasses as being a platform that third parties could just hook into. What has been involved in making this partnership happen? I mean, we often don't give thought to this, we give our Be My Eyes command and are delighted when it works, but it sounds like it has been a considerable commitment from both sides to get to this point.
Mike Buckley: Huge. But let's go back to the premise of your question and then the second part we'll do second. The first one was this product probably wasn't initially conceived as a platform and certainly wasn't conceived as a source of huge utility for the blind and low vision community. I think that's absolutely true. Mark Zuckerberg is on record as saying that he was surprised at the success and trajectory and really good feeling that consumers have had about the glasses, and I think that that altered Meta's thinking.
I don't know, I wasn't in the room with them when they were having these discussions two or three years ago, Jonathan. But then as you see that initial user adoption curve coming, then you think about, okay, how do I make this more of a platform? I don't know that it was conceived as a platform initially, maybe it was, but that's not my sense.
First of all, look, I want to give credit to Meta for bringing us in. Now we pushed and we called and we emailed and we did a whole bunch of things there, and so it was a two-sided process, but I think they ultimately realized what the power was here of working together. But as you say, it was hard, Jonathan, it was the majority of our engineering team with a fairly large group of engineers at Meta for six months, Jonathan. This is definitely more than a 1,000 human hours spent on this, it was an engineering slog to get through this and there are two quick stories I'll say there.
One is one of intimidation, where actually, President Riccobono, our vice chairman, Bryan Bashin, and I visited Meta to get a demo. We walked in and there were 18 technical people around the table, Jonathan, for context, there are 20 total people at Be My Eyes, not technical, just everybody. So it was, oh wow, this is this large group of really smart, quite talented human beings. The second story that really warms my heart, is apparently there's a process within Meta where engineers can raise their hands and try to get on projects that they're interested in, and apparently the demand to get in on this integration with Be My Eyes was off the charts. It was just a cool story, it made us feel great.
Jonathan Mosen: So is this a commercial transaction for Meta or is it purely a philanthropic one?
Mike Buckley: Oh, I think there's probably elements of both. There's money to be made in this space just generally, if you build any piece of great technology, which in this case can also be used as a piece of assistive technology. I hope there's commercial benefit for them, because I think they're doing the right thing with respect to building this with our community and with thoughtful beta testers. But I also think Meta is not stupid, and they know that if they can really get a win here and a win for the blind community, it's probably good marketing as well and that's okay. I think companies that do the right thing on accessibility, I hope they win both commercial, as well as from a marketing standpoint.
Jonathan Mosen: I want to talk about some other partnerships while I have you there, because you seem to be establishing them in good number. And we've talked about OpenAI already, but part of the consequence of that partnership was something you unveiled all the way back in May of this year, regarding ChatGPT-4o and that frankly stunning video that was released, which was looking at the ChatGPT video features. And then we had Bryan Bashin, who did an impressive live demo of it at our convention in July. What's the holdup with that? I guess many of us were hoping we would have that by now.
Mike Buckley: I think it's testing, safety, a whole bunch of things that they're working on, Jonathan. Look, I want to be transparent, I'm not happy that this product is not out yet and by the way, this is no fault of OpenAI. They've been a great partner, I have tremendous respect for them, I think their product is beautiful and lovely and they've been excellent to us. However, if I would've known that we would be six months later and still not have that product out, I wouldn't have released that video.
I just think it created desire, expectation that we haven't met, and I do not like to under-deliver. I hate over promising and I think when we put that video out, I think people rightly thought, oh, we'll probably have this in a few months and we don't, and that's not the way I want to operate. We were so excited by it, Jonathan, and the power, and I still have a version of it on my phone and it's stunning. But really just no BS, you and me, I don't think I would've done it again if I knew it was going to be six months and it's still not out to the public.
Jonathan Mosen: How disruptive do you expect this technology to be when it arrives? I mean, the idea that you can be wandering around, this thing is telling you in real-time what's going on. It sounds disruptive, but there are also potential things we need to think about, like what do you do when you are relying on this thing and the internet connection becomes patchy, or what about hallucinations? So there are some really big existential questions here.
Mike Buckley: A 100%, and I think that this is not something OpenAI has said to me, but I think that you just highlighted some of the reasons why OpenAI is being really deliberate about this, because I think it's massively disrupting on multiple levels. If you have the functional equivalent of a visual assistant that operates with you in real-time, in 70 languages, in a queryable format, in an interruptible format, do you think that's societally altering? I do, I absolutely do. Do you think also that it poses risks to the community on potential over-reliance? Absolutely. Do we think that this technology, like any technology, will be used in ways that maybe are not advisable?
Yes, in the same way that someone drives a car a 100 miles an hour, I think someone will try to drive a model a 100 miles an hour and it may not result in good stuff always. So yeah, I think there are massive challenges and connectivity is another one that you said. The other one is we're going to have to figure out how to deal with background noise, because you're conversing with this assistant in real-time, all the time. If you're in a loud environment, you may not be able to get through seamlessly. What happens in loud environments then? I am, without question, 90% more on the optimism and excitement side, but that doesn't mean it's without risk. By the way, I feel this the same way about when the initial models were released. Although I will say I have not been able to get GPT to do a real hallucination anytime in the recent past. Have you?
Jonathan Mosen: It gets things wrong, but I don't know whether I would call them full hallucinations anymore, it just sometimes gets facts wrong. I mean, it may misdescribe something from time to time.
Mike Buckley: Yes. By the way, this is not a guarantee, or I'm not saying it's perfect, but I haven't been able to get it to make something up out of thin air. I don't know, in the last five months, six months. And you're right, sometimes it's quote-unquote, see something incorrectly, or it may get a fact wrong, but I haven't seen just an invention.
Jonathan Mosen: You've had this technology on your device for a while and most of us haven't. What would you say about use cases? So when would I be better using the GPT option and when is a Be My Eyes volunteer still a better option than that?
Mike Buckley: I don't know for sure yet, because we haven't put this through beta testing, Jonathan, and I don't want to speak for the community. I think that there are two things at play here. One is factual trust based, and the other is emotional. On the factual and trust side, if you're dealing with something really important, or maybe there's a medical information, or whatever, are you going to trust a human connection versus an AI? That's a philosophical societal level question, I don't know what the answer is there.
On the human side and emotional side is that we already know that there are times when people just want a human connection, if for no other reason, sometimes than just to hear another voice and to have that interaction.
It has genuinely surprised us that as our AI sessions have grown and we're well over a 100,000 sessions a day now, which is up more than 4X in the last year. It has genuinely surprised me that calls to volunteers have gone up as well, because I had assumed that there would be sort of a replacement factor and I'm sure that there's some of that. But I think what we're learning is that there's still a desire on the part of some people who use our services to connect with a human.
And whether it's trust-based, emotion-based, or I don't really know, but I think it underscores the fact that our job is to make sure that we give people choices and that they're able to interact with the world in the way that they want, not the way that we think is best, but in the way that they want.
Jonathan Mosen: You've got to pay the bills and one way that you do that is these commercial partnerships, where we see various commercial entities popping up on the platform, and you recently announced a partnership with Hilton. Can you summarize that for those who aren't aware of it and what benefits that has?
Mike Buckley: Starting in US, Canada, UK, any interaction that you have at Hilton will be supported, not only by our AI capability, but also if you want to talk to a human at Hilton for sighted assistance of any kind, you can do so. This is really a groundbreaking partnership, where every part of the hotel and hospitality experience, Jonathan, is more accessible. So if you're booking, you can press one button through the Be My Eyes application and talk to an agent. If you're onsite at the hotel, you can talk to an agent, or you can use Be My AI, whatever works.
We've already seen in the first 28 days of this partnership, several hundred interactions that have happened and the initial feedback is good. But you travel a lot and you know that there are, I don't know, a dozen pain points when you interact with a hotel, starting from the booking process to the hotel process. This aims to ease those pain points and just make the experience better and I give Hilton a tremendous amount of credit. They really put us through our paces on this, in terms of not just working for them, but also making sure that it would be a great experience for the blind and low vision consumer. So I give them a lot of credit.
Jonathan Mosen: I was interested in this, because yes, I do travel a lot and I also travel with a hearing impairment and some of those noisy, cavernous hotel lobbies are challenging, because it's just raw sound and it's often difficult to tell the direction that the sound is coming from. So I was really interested in this partnership. Who do I get at the other end, if I talk to a human at Hilton? Are they trained in any way to provide directions, or give instructions to blind people that are meaningful?
Mike Buckley: Yes, it's a trained customer service agent. There are a group of people at Hilton who are dedicated to the support of the Be My Eyes community. All have been trained by our very own Bryan Bashin, by the way, who's Vice Chairman of Be My Eyes. But yes, you'll get a person who is knowledgeable on Hilton and its systems and things like that, so the support should be of high quality.
Jonathan Mosen: One concern that we had expressed to us was that it might be a way of letting Hilton off the hook in terms of some inaccessible things that people are seeing creeping into Hilton's, like a panel in a hotel room that is entirely a touch screen and it's basically the nerve system of the hotel. It controls heating, it controls lighting and it's not possible for a blind person to use that touch screen panel on their own. Does that concern you that Be My Eyes may be used in that way?
Mike Buckley: I'm going to say no for two reasons. One is that in the entirety of the year long conversation and testing that we had with Hilton, not once did anybody at Hilton ever hint, intimate, or otherwise suggest that this absolve them of accessibility responsibilities. Never once did that happen and so I think that their interest is sincere and just trying to do a step above. But the second thing is practically, if you take that example that you just used of the interactive screen, or phone, or whatever the thing is in the room.
I would hope that through connecting to a dedicated Hilton customer service representative, where you can show them the problem you have, or what you're trying to do, that we will be able to navigate that piece of technology or that environment. I'm not suggesting it's a panacea, Jonathan, and I think you're right to say, "Hey, this doesn't absolve any company, not the least of which, Hilton, from hopefully designing accessible experiences from the ground up." But I will say I think we'll be able to solve a lot of the potential issues of concern.
Jonathan Mosen: Always a pleasure to catch up during these exciting technological times with Mike Buckley, Chief Executive Officer of Be My Eyes. And ironically, given that we talked about how dependent we might be on the internet and what happens when it goes down, it did exactly that. The internet on my side of the call went down as we were concluding and Mike has asked me to conclude with a couple of points that he made, but we never heard.
First, the data that Be My Eyes receives from understanding when a consumer calls or uses Be My AI in a Hilton context, can and will be used to inform future Hilton design decisions. That's one of the most powerful aspects of the partnerships, says Mike. And finally, Mike wants to hear from folks about their experiences with Hilton. Mike also encourages everybody to book with Hilton, because they have made a very serious and groundbreaking investment in an accessible hospitality experience.
I remind that we are very interested in hearing from you about a couple of things. First, how is Be My Eyes working out for you on the Ray-Ban Meta Glasses? Do share your experiences with us and also, how are you getting on with the Ray-Ban Meta Glasses generally? If you've used them for a while now, are they a tool you continue to come back to and that you find adds value? You can be in touch with an audio attachment or just a written email, send it in to [email protected]. If you haven't caught the latest edition of the Nation's Blind Podcast, from the National Federation of the Blind, I highly recommend checking it out and of course, following it in your podcast player of choice.
You can find the Nation's Blind Podcast where you found this one and in the latest edition of the podcast, our hosts, Anil and Melissa, are joined by two of the National Federation of the Blind's newest board members. And they discuss how to protect our dignity as blind people and how to properly respond to individuals who may challenge our dignity. They also reflect and share their thoughts on President Riccobono's 2024 banquet speech that focuses on dignity. It's a good listen, make sure you check it out, The Nation's Blind Podcast from the National Federation of the Blind.
Finally, on this week's Access On our tech tip and this one comes from the legendary, Curtis Chong. He says, "Greetings from Aurora, Colorado. When using Microsoft Outlook, I use Outlook Classic myself, I have often encountered several email messages sent to me that behave very much as if the message was a web page. I find that there are some times when the contents of a message must necessarily be read in the browser and not Outlook. And even though there is an info bar which gives you the option to view the message in a browser, I find that the result is not as good as what you will get if you do the following.
First, open a complex email message. Now press the F12 function key to bring up the save as dialog. Type a name that you will remember, I often use a name like Junk, to indicate that I can delete the file at any time. Press the tab key once to focus on the save as type combo box. Now down arrow to HTML, star dot HTM, star dot HTML and press the enter key. You will need to press the enter key a second time to perform the actual save. On my system, this places the file into my documents folder.
You may want to check the address bar in this dialog to be sure that you know where the file is going to be saved before you actually perform the save. Finally, with File Explorer, navigate to your documents folder or wherever your message is saved and find the file you have just created. Pressing enter on the file should bring the file directly into your default browser. This procedure works in Windows, regardless of what screen reader you happen to be using."
That's a very handy one. Thank you very much for sending that in, Curtis. If you'd like to send a tech tip in for any operating system or any hardware that's commonly used in the blind community, be our guest. We've got a few banking up now, but that's just great and you can send it in to [email protected]. That's [email protected]. That concludes this edition of Access On, the Technology Podcast of the National Federation of the Blind. If you'd like to know more about the NFB, visit our website at nfb.org.
You can phone us on 410-659-9314. And be sure to check out the Nation's Blind Podcast right from where you heard this podcast. To send in a contribution for possible inclusion for a future episode of Access On, email us, you can attach an audio clip to the email, or you can just write the email down. The address is [email protected]. That's Accesson, all joined together, @nfb.org. And always remember, it's the organized blind movement that puts Access On the agenda.