Yes/No.

Accuracy: Please use this as a moment to empathize with old people that can’t distinguish between short press and long press or can’t double tap at the right speed. In my experience, the gesture is incredibly accurate, but you have to learn the pace of tapping. Not too fast not too slow. This is something we do all the time everywhere else, but have internalized and lost consciousness of. (The double tap also needs your watch to be awake which has leads to separate class of failures — will be amazing if they ever get a system that can skip that need.)

Utility: Hardly anyone thinks its current form is very useful. (It’s a little useful. I think of it as “hugging support”. If I wake up in the morning and have one arm around someone and want to turn off an alarm it’s great. Or if I have an arm around a friend and want to take a photo: also great. …Music would be useful even without hugs if I weren’t casting from watch too.)

The hope, I think/for me, is that this is gesture conversion 1. The watch needs multiple gestures to navigate meaningfully. Like with the accessibility version. Hopefully they bit by it add reliable, snappy versions of the accessibility gestures. Having multiple will be more than the sum of parts!

Hmm. That is strange. I don’t think of Apple as doing much artificial gating. This may be that (and I’m not opposed to charging for software), but it is notable that the Apple Watch keyboard does a lot of prediction. It works surprisingly well and clearly does a lot of guessing and prompting to deal with the fact that a finger is like 4 keys large. I don’t know much about the SE, but I wonder if they found the experience or be worse there. (If they’re going to give dictation it wouldn’t make a ton of sense not to give a keyboard if it were an option I’d think.)

Question: what language do you speak that dictation works poorly in? Just curious. I only recently tried it (dictation) and I was blown away by how well it works in English.

I don’t use it much, but it actually, very surprisingly, works quite well. It does more prediction and gives you more options as you type. I think it does more inference regarding your fat fingers and what they could possible mean. I was actually impressed.

Still not something I do much, but impressive.

Pretty reasonable to expect a keyboard to be there.

Tried both.
Love the batter life. Hate the lag in what should be a micro-moment to read. (Especially if at gym and timing between sets or using maps for navigation.)

I keep AOD on by default. But if travelling or out for long periods or likely to wake up somewhere unexpected: it’s a great way to extend battery.

Someone else here mentioned making a shortcut and adding to Smart Stack. I like that idea. 👍

We have very different sleep. For me: I get about 1/8 deep and almost always exclusive to the first half of the night. (Apple tracked sleep pattern that seems common anecdotally; I haven’t checked if a general sleep pattern in studies.)

Only thing I’ve noticed that increases deep sleep reliably so far has been, and I apologize if this feels rude, but it is relevant, is having had sex directly before. (Which is a great reason for it! :)

That said, as you can see, you’ve got much more atypical sleep.

Not an expert here. But I’d try recording yourself and seeing what’s going on at night. (I know there are audio apps for this, maybe video as well.)

If you can rule out snoring or not breathing (blood oxygen readings may help there) you can start looking at things like what you are, ambient noise, etc.

There are also drugs that help with sleep maintenance vs offset. And I know of at least one that purportedly doesn’t disrupt deep sleep. (A common problem with sleep drugs is they also disrupt sleep itself; but there do seem to be a few good ones out there.)

I’ve never gotten it to work.
If I can even find a contact via the manual rolodexing system they force you to use. It’s straight bizarre.

Disagree.

Battery life is huge QoL improvement. You’ll never have to worry about making it through a day.

Got a watch for a parent for health tracking. Ultra 2. (After she tried both regular and ultra and liked the ultra; not too heavy — she’s teeny, mind.)

Harder for them to scratch or damage. Comes with cellular option if they ever want it. Twice the battery life. (Supports sleep tracking more easily if desired because no need to charge overnight.)

Ultra is just a good choice for lots of things.

Regular watch is fine, but just how you live your life. (And aesthetics if you’re into that.)

Thanks.
I’m definitely not trying to balance out AW numbers. (And I realize that would be a fool’s game. They’re going to be very approximate at best.). I’m not even concerned with the AW expended numbers.

I just want to track food being eaten.

I definitely don’t have this issue.
You might try just getting a replacement UW2 first (assuming you’ve already done a hard reset to check software)

Also possible that this is a human variation thing. 🤷 (Is there a consensus yet? I’ve seen a few people mention this, but most UW2 users seem to have no problem.)

I assume this is a joke post. Or I’m misunderstanding.

But in case it’s a legitimate unfamiliarity with the watch’s function:

The back of the watch is where many key sensors are. If you obstruct it you’ll lose a lot of functionality.

I have no clue (though from my electronics meddling past I’d assume it would be salvageable if it’s currently working with some viscous non-conducting superglue like epoxy in the cracks. [viscous so it doesn’t drip too far in])

BUT: what’s the cost for repair? (Just curious as much as anything.)

You can always manually set it to sleep.

I’ve heard, but cannot verify, that even though that data is not displayed natively it is tracked.

Useful to know if so as it would allow different apps you switch between to uses that data as they see fit.

Would be great. Strongly support.

Difficulties:

• Different biology down there. Apple is doing a lot of unstructured learning on large datasets to make its inference algorithms from what we can see. It’s possible that they’d be set back a lot if they had to support an area they weren’t collecting data for. (Hard to say — I don’t know what the major research bottlenecks are for them. I agree that we’re probably a lot of great signals down there, but a great language you don’t understand isn’t communicative, I’d you get me.)

• It’s a touch screen. You’d need to very reliably get it to turn off when it faces away from you or you could get lots of accidental taps and movement from a leg or wet gym clothes etc. (And watches don’t have a camera; they don’t know when you’re looking at them)

• It’s a scratch able surface that you now have facing down. It’s going to be constantly run back and forth over metal laptops etc. This significantly increases the abrasion that a lot of people would be looking at.

No Uber (or equivalent) then no phone replacement.

If I could call an Uber on my watch I’d often leave phone at home for short outs and dates, I’m pretty sure.