I now listen to audiobooks, podcasts for ~2 hours daily, while walking my dog. If there's a way to control apps, navigate options, perform useful dictation, I haven't found it. I can barely get Siri to reliably start and stop audio playback.
I'm now curious how blind people manage these things. Also, I dimly recall Marvin Minsky or Donald Norman or Alan Kay... writing about early experiments with conversational user interfaces. It really seems to me that we have to go back to the basics, to start over with new assumptions. (I still have the books, I might be able to relocate those notions.)
To wrap up, much as it's clear my Apple Watch (w/ AirPods) has the potential to replace my iPhone, and my clear desire for that to happen, today it's not on the horizon.
Oh, here's my gratuitous geek cred: We were trying to use voice control in early 90s. For stuff like AutoCAD. While some of my users liked it, one even excelled with it, most didn't. So I've been waiting for this for a long time.
The key ingredients that make iOS an excellent computing platform for those hard of sight are
* enforcement of UI guidelines that enable the screen-reading to work well
* a collection of accessibility gestures meant for no-screen operation.
NOT Siri.
google assistant is getting close
Continuing my rant...
I know that Siri is impaired by choices Apple's made wrt to privacy or some such.
But why does local voice recognition fail so often? Like while hiking or I'm in the basement. Surely some hybrid is possible. Does Siri really need network access to process a stop playback command? (I do fuss with the Voice Control features every few months, to see if things have improved.)
Warning, idea incoming:
I want the ¡Tchkung! voice interface. Pops, clicks, buzzes, Looney Tunes sounds effects. Why are we limited to just speech recognition? Why can't I beatbox to control my phone?
In the 80s, there was a comic who'd straight read serious writing, adding sound effects for the punctuation. Hysterical (at the time). Sorry, can't remember name. But I want to make sounds for punctuation, backspace, back one word, newline, etc.
But I see younger people making presentations at work that are littered with animations. To me it looks extremely unprofessional (both from the distraction and the usual childish nature of the images themselves), but its common enough that it's clearly within mainstream norms.
I also use my watch in ways the author does (though not to as extreme degree!) and do find it handy. But for me it's a convenience feature when my hands are full or I'm literally on the run. I was impressed by the level of effort described.
That being said: the idea of an audio-first future horrifies me. Audio and video are quite linear while a text article (like this post) are much easier to skim, read closely, and hop back to reread something. It can also be read in much less time than voice would take.
No.
It comes with a lot of drawbacks:
- Speed. A lot of people can read a lot faster than your usual conversation speed (and in their own pace), while audio is at a rather constant speed, which you might not even be able to control. Additionally, you can easily re-read something when you missed it or did not understand it - doing so with audio is not that easy.
- Privacy. When you're using a voice assistant, you're broadcasting your inputs to your surroundings and maybe even the output. Yes, people can already look on your screen, but the field is limited and the behaviour is discouraged. You're basically becoming that annoying person on the phone overheard by everyone, all the time.
- Scalability. Somewhat related, but a full bus with people on their phone is not a problem (in non-pandemic circumstances). Now, imagine 40 people talking to their phone in that bus.
- Clarity: Human-to-human understanding is quite imperfect already, especially in noisy environments or with distance. Text does not have any problems with this [0].
- Information density. "A picture can say more than a thousand words". Try reading a scientific paper with graphics via audio only, it'll be fun.
That's not to say that voice is useless. When you're driving or your hands are full while you're in your home, it is ideal. It also opens up accessibility to people which might not handle screens that well (for example very old persons).
But given these limitations, I can not imagine voice replacing screens as the main input method any time soon, or at all. For the edge cases above, sure. But in the general case? No.
[0] There are still understanding problems with text, but voice has those, too.
I'm planning on eventually getting an Apple Watch so I can leave my phone at home, and want to try using a Tap keyboard with it. The Tap has mixed reviews, but I'd like to try out being able to input text using a mobile keyboard into the watch. I don't know if it would result in a satisfactory experience, but that's why I want to try it.
I've noticed I'm wasting a lot more time checking my phone during the pandemic (trivial things, and, shock horror... even too much HN at times).
Anything with a "feed" can definitely be addictive. It's too easy for me to browse my RSS feeds, HN, and podcast feeds on my phone. And I'm doing it a lot.
Not sure I could go quite to the extreme of Watch-only but it has made me consider whether I could make more use of my Apple Watch instead of my phone to reduce distractions.
Contrast this with my favorite phone, the MOTO FONE[2] from 2006 which also had an e-ink display and had dimensions of 47 x 114 x 9.1 mm.
This has been true of many of the "modern dumb phones" - dimensions, particularly thickness, have been surprisingly large.
EDIT: On the other hand, the "Light Phone" linked from the article has dimensions of 95.85 x 55.85 x 8.75mm which is, at least, not silly ...
[1] https://www.kickstarter.com/projects/mudita/mudita-pure-your...
In fact I’ve been aggressively turning of most notifications across the board.
iOS’ new(ish) “deliver silently” option is helpful here.
Which is to say, if I wanted to carry an apple watch in my pocket, and didn't want the sort of lame, and unifinished, aesthetic of bare/empty band connectors ... is there some pendant or medallion or ... something ... that you can attach the watch to that isn't a wristband ?
OH, AND ALSO ...
The op speaks of the necessary iphone that he leaves in his drawer, at home, like a router ... now that we have apple silicon, etc., can I run my "iphone" on a mac mini, in an emulator ? I mean, if you were developing apps, etc., for Apple Watches, or doing Q&A, you wouldn't have 50 iphones somewhere, right ? RIGHT ?
I like the idea of only carrying around a watch + headset. I'm actually thinking of doing this next month.
In my country there is no covid so life was practically normal. My office has blanket wifi and I discovered the watch would easily do a day if connected.
The most annoying part was how slowly my airpods connected to the watch when taken out of their case, it made the first 15 seconds or so of incoming phone calls very awkward.
My phone I use for geo-recording points for ArcGIS, reading server status and data in realtime, testing network connections, thermal camera, and more. Not everyone lives in the land of sitting in an office all day and going to meetings, some people physically work for a living.
Good for the person in the article loving their tech, but it's not the future for everyone.
Sadly most tablets are not waterproof, and are quite awkward as cameras. Otherwise I’d probably still go without a phone (I have zero need for telephone calls or SMS)
LTE is not one thing. It's a bunch of different radio frequency bands that are selectively used by different networks in different countries.
Because of that, cross-network phone interoperability is a consumer nightmare of technical minutia. See https://en.wikipedia.org/wiki/List_of_LTE_networks and https://en.wikipedia.org/wiki/LTE_frequency_bands
But furthermore, an Apple Watch with cellular bought in the US, cannot use cellular data in e.g. the EU. Not sure why, but it must be localized on a software/hardware (?) level. At least that’s what the Apple genius told me on my last visit to the states..
Also, audio-first apps for watches will be huge, and already a great niche to experiment with (although ability to code apps is still limited)
First privacy, I don't want my audio going to any server unless its used in a Google search. This is about on device ml. Also this requires internet access just to use voice commands, which is silly.
Another issue is all the health apps logging my bodys stats. Again if this was in device it would be okay.
Their was also a ux issue using the device is very slow with all the http , you have to speak, probably repeat because it missed something then wait until it repeats you intent. This is a very long process just to check something simple that would take a few seconds in a phone.
Also you look stupid repeating outloud the same thing in public to your watch.
The battery life is way too low as well
My 2 cents.
This is especially annoying if you're listening to streaming radio (like NPR) and going for a walk. It will just drop. This has been the case since the very first version of the cellular watch, I suspect as a battery-saving feature.
The only way to really get around this is to force cellular mode before you leave. This is what I do, and it's embarrassingly un-Apple in experience:
(1) Put my iPhone in Airplane mode. I think I could accomplish the same thing just turning off BT, but I want to be really sure the phone and watch are cut off from each other.
(2) Turn WiFi off on my Apple Watch. Wait a few minutes(!) before I go for a walk -- make coffee, leash up my dog, etc
(3) Walk outside, and issue my Siri command to trigger music/radio
This usually works fine. But its annoying. If you're not especially dedicated to going phone-free (it drains the battery quickly) I really wouldn't bother.
No, you can use AirPods on a non-cellular watch just fine, it has bluetooth. Cellular only affects your ability to stream content (eg. Music or podcasts), but you are also able to download/cache those while on WiFi on the non-cellular watch.
...he posted in 172 characters.
In 2006 predicting that the future of the internet wasn’t going to be in longform text would’ve been pretty bold and pretty prescient.
Look at what’s happened to longer text content since then. The concept of “long reads” is all but dead now. Been replaced by short snarky comments, tweet threads, videos, and social feeds.
To me, this indicates that accretion, not replacement, is the better model. Your mileage may vary.
As Paul Graham said, "Live in the future and build what's missing!" - Going Apple watch-only might allow us to find a ton of interesting audio-first ideas like clubhouse.
going into audio mode is a completely different attention pattern, and definitely not casual.
Laptops are far better for visual since you can focus on longer-form content and are less distracted.
Is this a dystopian prediction? How can we ensure that the future doesn't turn out to be so bad?
