Comment by alialkhatib | Hacker News Reader

Comment by alialkhatib | Hacker News Reader

Parent Full thread

The more I look at these measures, the more I wonder whether Apple has a plan for all of this or if they're just throwing spaghetti on the wall and seeing what sticks. Some of these seem like there's tie-in (body measurements can be assessed through a Withings scale, sleep can be assessed through a Jawbone UP or similar device, and the "Me" section is mostly static), but does Apple have any plan to get people to track their Molybdenum intake on a regular basis? Or their BAC? Or their body temperature?

I'm not saying they need to, but in the past it has seemed like when Apple provided an affordance, they also had a clear use case. If nothing else, you could fully use the device/service in the way they describe, and that would sufficiently justify its adoption for many. With Health, I'm not seeing that. I wonder if this will threaten the overall adoption of Health, which in general seems like not a terrible idea - anything that gets people thinking more about their activity and diet seems like a good thing.

One could argue that their solution is to leave it to third party devices and integration. Apple doesn't need to make a scale to help you track your weight (Withings has already done it, as I'm sure others have), nor does it want to touch the quagmire that is nutrition by claiming to measure the nutritional value of any food. Let some food-related startup do that dirty work, as long as Apple provides some interface for the measurement to end up in this data silo.

This argument seems weaker to me than the argument seemed when Apple first started accepting apps for the iPhone (later iOS). Health currently leaves huge swaths of its UI barren for lack of data, and I'm not sure that hardware developers can swarm to fill in those gaps the way software developers could with the platform.

Maybe I missed the part of their talk where they made it obvious how they envisioned Health (and HealthKit) to be used in a fuller way; if that's the case, I'd appreciate a pointer.

What drove this home for me was the inclusion of a separate measure for BMI.

I've already given it my height (hopefully static, but apparently deserving of a line graph), and my weight. Yet apparently calculating my BMI from these inputs (admittedly a crude measure of health) is beyond the capability Health app?

BMI is only intended to measure health at a population level, it's too crude to have any useful input on personal health.

And height isn't static if you're an adolescent.

Exactly - and yet BMI is one of the metrics included. Meanwhile a manual input is required, despite having been provided with the relevant data.

Perhaps this is a concession so that people who do not want to work with BMI do not have to by default. As minikites says, on a personal health basis BMI isn't very useful so folks may want to skip it altogether.

There are devices which measure BMI using conductivity. Given that a device may report a value other than the simple function of height and weight, it seems prudent to have a spot to put it.

I think you are thinking of body fat percentage. BMI is literally a function of height and weight.

Implant-based blood glucose measurement devices[1] are becoming available, for example. These devices are well outside Apple's area of expertise, but interfacing with the data from these devices is well and truly within Apple's grasp.

Imagine that Apple was going to develop wearable technology of some sort (like, say, a wrist-watch) with inductance charging. Would the induction loop also be usable to interface with implanted inductance-powered sensors located in the wrist?

[1] http://www.diabeticlive.com/diabetes-101/diabetes-news/impla...

> body measurements can be assessed through a Withings scale

Nope, Withings still hasn't actually released an update that does this. Despite it being one of the "flagship apps" that were advertised with iOS8 to have the feature from the start.

None of the mainstream health/diet/fitness apps currently give anything to the Health app that I can find.

Definitely a very lacklustre launch. I was/am quite excited for this app but when it launched and nothing actually worked except your steps, it was a bit of a letdown.

The need to pull HealthKit apps from 8.0 probably meant Withings delayed their update (so they could stay on the store). With 8.0.2 out now they'll presumably have an update soon for it.

I think Apple is setting the stage for someone else to come out with an excellent virtual personal trainer app. Personal trainers are one of the reasons that the wealthy are more fit than everyone else. This could be a big deal for first world health.

> anything that gets people thinking more about their activity and diet seems like a good thing.

This makes intuitive sense; however, in practice, the truth is a bit more complicated.

We don't really know how to get people to change their behaviour and exercise more, eat less, etc.[0] - this is the holy grail of "lifestyle" diseases such as diabetes or heart attacks. (Disclaimer: my startup is a health-tech company aimed at improving medical outcomes through a number of factors, including getting patients to adapt their behaviour).

However, there's a lot that we know doesn't work. It turns out that simply making people more aware of what they're eating not only doesn't work, but in fact oftentimes causes people to eat more. There are a few hypotheses explaining this, both of which are supported by evidence (so it's likely to be a combination of factors).

First, people falsely equate awareness with action, psychologically[1], so making them over-aware of that they are doing makes them likely to think that they are taking more action than they actually are[2].

The second is that people don't have a good sense of how to process this data, so by (e.g.) placing caloric counts[3] next to food items, if they don't also know how many calories they should be consuming in a typical meal[4], they can end up overeating ('This dessert is "only" 800 calories!).

[0] by this, I mean a systematic process that is cost-effective on the population at large, not something that is (a) effective only for a small subset of people, or (b) effective but cost-prohibitive. There are a lot of strategies that are very effective for people who are already low-risk, and a few strategies that are effective but insanely expensive for people who are high-risk.

[1] This is analogous to what happens in politics with "slactivism".

[2] This is probably the origin of the advice that you shouldn't tell your friends about new resolutions before you have constructed a plan to enact them - telling others about it gives you a sense of satisfaction even before you've done anything.

[3] Not that calories are really the best way to measure diet, but let's set that aside

[4] Most often, people are familiar with the number '2,000 daily', but that doesn't help you determine how that should be allocated throughout the day (dividing up into three equally-sized meals is not ideal), and many people require significantly less than or more than 2,000 calories.

If people have no reference for how much they should eat, then it would be trivial for any app to notify them about it, e.g based on just height, gender and weight[0].

If the problem is inaction then you could have e.g the step counter show the user not just the steps they have done but whether they have reached a safe minimum. Fitbit does that - and it works.

[0]: it would be reasonable to infer their activity level based on their weight too.

Reply on news.ycombinator.com