Show HN: HipStat – Track Inches, Not Pounds | Hacker News Reader

Show HN: HipStat – Track Inches, Not Pounds

play.google.com

33 pointstomeglenn65 comments

Hey guys, I've been a lurker of HN for a while and thought I'd post up a Show HN thread for my new Android/Web app geared towards tracking body measurements rather than body weight.

This is a super early release so there are bound to be issues (see: http://genius.com/4126912/Genius-the-genius-isms/Take-the-ro...).

Really interested in getting some early feedback, so your comments are very much appreciated.

The plan is to focus on next implementing goal tracking and then to add some simple social elements (a la Instagram).

Cheers!

PS: If you don't have an Android device, you can access it via the web at http://app.hipstat.com

I have always wanted to make an app like this. Never got around to it.

1. Please add more photos of the UI 2. Show trend lines / moving average 3. Allow users to use it without signing up. If you really want to ask users to sign up offer facebook/google sign up 4. Remove username requirement. Just use email.

Because of (3) I didn't install. The friction is a bit too high

Thanks for the feedback. I initially had Facebook and Google logins, but since its a meteor app there is a less than desirable user flow (it opens a Cordova browser which looks awful and possibly even looks fake).

I'll take your feedback on board though thank you!

I think the large picture of the woman in the bikini might be a little too revealing. You might go with something tamer. I imagine it would make the target audience more comfortable.

I think it's fine. Although it should show either a fit man or woman, depending on what sex you choose. Something to inspire you with the will the continue on the iron path.

Yeah, the picture is super off-putting; I liked the app concept but any interest I had disappeared when I saw that login screen. It looks like something more suited for Tinder than a health tracking app, and frankly, I'm sick of seeing skimpily clad women as the metric for idealness. It's demeaning and demotivating as hell if you can't fit into that mold. Not to mention that some titillating picture of a woman's ass isn't in any way appropriate for a general interest app in the first place.

Thanks for the feedback. I will be looking to change this to a less intimidating image.

Thank you to everyone who has provided feedback. I have tried to reply to as many as I can. I greatly appreciate all of your comments, the good and the bad!

Please keep the feedback coming. This is exactly the reason I released this app early! :)

I may be missing it, but the cool thing about self-quantification usually is that you're not responsible for data acquisition. Things like Fitbit/Aria and Mint give you some automation and make it easier and more fun to track data (and ostensibly turn it into action).

Of course, there's no smart tape that I know of, but I think it would be CV-trivial to rescale & superimpose silhouettes of progressive pictures on top of each other to track progress. I know a lot of people would shy away from doing something like that, but it would be pretty cool. As it is, measurement with a tape is a pain and worse, subject to a good deal of human error.

Isn't it possible to do the measurements visually using Kinect?

Possibly? I don't know how accurate they'd be, but that's the general idea I'm getting at. Most people have a hard time committing to long-term work for self-quantification.

Measurement of progress autonomously means less work, more data and perhaps more chance at success.

It should be _possible_ with even a smartphone camera.

You'd have to do some clever image registration to calibrate/ re-scale the image.

I did a little (undergraduate) research in this area. Granted, I'm not very skilled in this regard, but my conclusion was that the kinect still wasn't quite accurate enough to make proper measurements like this.

I definitely agree with this - I think minimising the friction of the data acquisition has been a big part of their success.

As a personal example I've wanted to track my weight over time for a while but even the friction of standing on the scale and then manually inputting the data was enough to mean I only ever kept it up for a few days at a time before I'd forget or it became tedious.

I recently bought a Withings scale and the removal of that seemingly tiny amount of friction of manually recording the weight has made all the difference.

In high school I tracked my weight and what I ate everyday in a mini composition notebook I kept in my back pocket at all times. I was exercising (not to mention younger) at the time and lost 30lbs over 4-6 months. I now have a withings scale (along with access to a myriad of food tracking apps) and I can't help but feel that my pen and paper approach was easier and yielded better results... I always feel like entering what I eat into an app is more trouble than it's worth especially when it starts listing calories because I feel compelled to get it as close as possible so I'm busy searching for calorie counts of what I'm eating. In HS I just wrote "Ham and cheese sandwich, sandwich bag of chips, fruit cup" and that was it. MUCH simpler IMHO. I know calorie counting "works" but I just always feel like the counts aren't accurate b/c I cook a lot of things myself and have no desire to calculate the calories of the dishes I make. If they aren't going to be accurate then why do it at all?

I did what you did at one point in my life. In fact, I got perversely granular.

It's not sustainable unless you have a very real mental condition. I think Withings is compatible with a number of apps such that you should't have to enter anything into it to have it tracked.

Food is notoriously difficult to track, and even harder to accurately quantify. A burger and fries meal at one place could be +/- 500 kilocalories despite looking the same. After a while I determined food tracking to only be worthwhile at high level - ie, "ate burger" - rather than to do the legwork necessary to get accurate kilocalorie counts.

>"not sustainable unless you have a very real mental condition"

How does a mental condition correlate with sustainability?

Food is easy to track, if you cook yourself. And fairly easy to estimate if you're eating out.

I've tracked ~96% of my meals for the last 3ish years. As well as my weight every morning. It takes an extra 10 minutes max per day, the benefits of the data is well worth my time.

That was somewhat tongue-in-cheek, but I mentioned it because it requires tireless devotion to get right, for ultimately not that much payoff.

I did it for five years and I don't think there was much benefit over writing down what I ate.

The Weight Watchers points system (the older one) was essentially trying to solve this. Points are essentially calories divided by 50 (with adjustments to emphasize low fat and high fiber) and are easily memorizable. It doesn't matter exactly which bread you ate, because it's probably 1 point per slice anyway.

(I don't know much about the "new" points system. I think there are a lot of things that are just 0 points.)

I wonder if your point about pen and paper feeling like it worked better is because the app approach puts even more focus on the component data itself and not on what is probably more important - what we actually need do to reach our end goals which is not always so obvious.

A big problem I see with a lot of the current generation of health apps (at least those I have used) is that they've become pretty good at capturing and presenting data (making it pretty) but there's often still a huge gap to fill on actually interpreting this data and telling the user what they now need to do to improve and reach their goals. That's where I personally feel the real value is. (And to be honest I think this is a theme outside of health apps as well...)

Again another personal example - I've been getting into running more and more over the past few years and like a lot of people have been using GPS running apps to track the runs (Runkeeper, Endomodo, Strava etc.). The majority of these apps gamify and give encouragement on things like fastest mile/5k/10k, average pace for the month, frequency, longest distance etc. So as an unknowledgable and fairly ignorant runner I've ended up just trying to beat these metrics - by trying to push harder and run faster on every run.

The problem is it's only recently, now that I've become more keen and have started doing research, that I found out that this is completely the wrong way to do things. Almost any decent training program or coach would recommend running at different paces on different runs and spending up to 70-80% of your training at a relatively comfortable, easy pace [1]. This is much more effective at helping you to run longer, faster and avoid injury.

And it's this knowledge that was completely missing from these apps.

PS @tomeglenn, I love the idea of the app "measuring what really matters" - and in a way addressing the problem with BMI. If you can make the measurement as easy as possible that would certainly be a bonus. But for me combining it with coaching/education on how to actually reach goals (rather than just setting the goal and tracking/rewarding the deltas along the way) takes it to the next level, although is probably much harder to do. Best of luck, and congrats on shipping!

[1] http://www.goodrunguide.co.uk/TrainingGuide.asp

Thank you for your comments. I will definitely try to provide guidance and coaching in app to help people take accurate measurements.

Thanks for the feedback. I agree that autonomous data collection would be preferably. But with body measurements it would be difficult to achieve.

I understand that people may not be comfortable measuring their body accurately. I will look to provide help and guides in app to solve this.

Body measurements are the purpose of this app, so image analysis and estimation based on silhouettes would not achieve the desired outcome.

How about track cm, not inches?

Agreed.

This will be an option soon.

Before you sign up to try it - there's no way to remove your account or indicate you don't want email from HipStat during the sign up process or after you've joined.

The SPAM button in your e-mail client can take good care of that, should you find yourself in that position.

It's fine for me to throw eggs at your house because you own a mop.

Or its like me sending you a letter in the mail and you throwing it out...?

Pretty sure cleaning up after thrown eggs is significantly more work, and more public than hitting the spam button

Relax, I'm not defending their actions, I'm explaining that you shouldn't feel bad about pressing "SPAM" in that case.

There has been a subtle pr campaign to convince people you should press "unsubscribe" on any unwanted mail instead of hitting SPAM, lest you ruin it for everybody. I'm saying: don't believe the hype. In this case, just press SPAM.

I would use Mailinator: http://mailinator.com/

Note that emails sent to Mailinator are public.

The option to delete your account will be included soon. We currently do not send any emails at all. But when/if we do, there will be the option to opt out.

You can now remove your account. Since we don't send any emails yet, there is not option to disable email updates. But I will be pro-actively adding this option soon so that people can opt out before hand.

Please add SI units. I would like to use this app but I'm accustomed to m/cm and inches are meaningless to me.

Rest assured I'll be adding this soon. Thank you for your feedback.

There's the very obvious UX problem though, that for weight, all you need to do is step on a block of solid material, whereas this has a lot more friction in the process.

Not sure that's solvable, but that's been my main reason for not tracking these measurements for now.

Thanks for your feedback. I agree that it's more involved and effort to track measurements, but without developing a companion piece of hardware (smart tape?) I see no solution for this.

You get a much more precise indication of the way your body is changing for your help extra effort however. This app for example will be of far more use to a body builder than a weight tracker would be.

Just a quick suggestion as the husband of a woman who does not like tracking her fitness by pounds. She is always more concerned with which size dress she can fit into. Maybe it would be good to have a conversion mechanism from inches to dress size.

That would be really difficult to do as no two dresses are the same size. Same for male clothing to be honest!

Yes, it's a disaster area. I'll just plug a small web app written by a friend of mine that addresses this problem: http://sizes.darkgreener.com/

It uses least squares regression to work out, given your actual measurements, what "size" label from which shops is likely to fit.

This is great. Would your friend be interested in discussing a collaboration? I'm looking to build similar functionality into HipStat.

As a woman who also don't like to track her fitness by pounds, dress size is not a particularly useful metric, because it's so highly variable among manufacturers. I have pants that range from size 8 to size 16. If I tried to measure my fitness by what pant size I wore today, I'd go crazy in record time.

We will be providing clothing sizes based on leading UK manufacturers in the near future.

"So you have to wonder, what's really the point in measuring weight? Why not measure what really matters, the size of your body!"

I do have to wonder, why is it that the size of your body matters?

The relative sizes of various parts of your body correlate closely to health for the majority of people.

Well, it matters if you care about measuring your fat loss. Measuring fat loss by tracking weight is less accurate, because muscle is heavier than fat, if you are exercising to lose fat you may get heavier due to increased muscle mass, even though you are thinner, i.e. carrying less fat. There are other factors that affect your weight, such as water retention. As for why to measure at the hip, well, there isn't much muscle around that area compared to say your thigh, so the size of the muscle is not affecting the measurement output so much, meaning you are measuring fat loss, rather than fat+muscle loss/gain. (Muscle loss is generally not a good thing, it's very important to our health).

Personally I think we need a way to measure our health rather than tracking metrics that can encourage unhealthy relationships with our bodies. Ideally we would be able to quickly assess how close we are to a healthy level of body fat, lean muscle mass, be it is too high or too low. Measuring how heavy / big you are feeds into "weight status" and "thigh gap" fixation. In a less-imperfect world we'd be competitive about how healthy we are, rather than silly and often unhealthy things like thinness, and weight.

If we're talking about fitness goals, generally, weight is actually a fairly poor indicator of progress. Someone can be 200 lb and another person can be 180 lb, and both can be the same height; yet the 200lb person might be stronger or more athletic, while the 180lb person could have very little muscle mass and a much higher body fat %.

Stated in another way, it's entirely possible that a person can gain weight and lose fat -- or even lose weight and accumulate fat.

The size of various parts of your body is a much better indicator compared to measuring weight (measuring one overall size isn't a great help either), although still not the best. Measuring body fat % would be best, but unfortunately, it's not very easy to do so -- at least not without special tools.

UPDATE 21/07/2015:

Hi all. I wanted to post a quick update based on some of your fantastic feedback.

As of today, I have released a new update (you should get it automatically without needing to update the app). This update includes the following:

- Ability to delete your account from the settings screen

- Improvements to the way measurements are saved (server side)

- Ability to save measurements in cm or inches (more to come on this, including an account settings preference for metrics)

I am aiming to change the login screen background this week along with a couple of other minor tweaks before I go away for a long weekend.

Thanks again Tom

Isn't there a big difference in health/success between a fat arm and an arm with muscles on it?

There is. There isn't a lot of point measuring arms as an indicator of body fat. But if you measure the right parts of the body, and you measure more than one, you can get a surprisingly accurate measurement of body fat - the US Navy Body Composition Assessment method gets to within 3% based on two or three circumferences and your height:

http://fitness.stackexchange.com/questions/1660/how-accurate...

There is, but I think if someone is using this app to keep track of their arm diameter, they are probably aware of that ;)

In theory, I like the idea of tracking measurements.

In reality, I can't take reliable measurements of myself. Even back-to-back measurements vary quite a lot because the tape isn't in the exact same position and orientation.

What about a smart belt that automatically tracks your waist size? Other measurements may be harder to get automatically.

Or just use an ordinary belt and look at how the belt-hole-number changes.

That's so 2014!

Belts typically go around your hips, not your waist. Clearly we need a smart corset.

I find the demo screenshot intriguing as I try to visualize someone with either a 14" or 20" waistline.

That demo measurement is for the upper arm. A waist that size would be disturbing, or an infant.

I may not be the target market, but the "social networking coming soon!" bit is a huge turn off for me.

Thanks for the feedback. What specifically turns you off about this idea?

The goal would be to provide support and motivation. It would be an opt out feature. You will be able to make your account private.

Any plans for iOS?

A way to add new categories to Health app would replace many of these kinds of applications.

Definitely! Unfortunately I don't have a date to give you at this stage.

did you use semantic-ui?

I did yeah :)

Reply on news.ycombinator.com