April Zero v3
aprilzero.com
aprilzero.com
I just can't seem to see the value in it; what actionable or information do you get out of this? I've tracked my movements with OpenPaths [1] in the past and had fun building a couple data visualizations out of it (I love the fact that others can request access to anonymized data collected by OpenPaths).
I just have no anecdotal or empirical evidence of users putting this data to good use. If anything, folks I've seen using Fitbit/FuelBand could use a bit more exercise and a better diet. People tracking their sleep activity tend to stay late and have poor sleeping habits (for example).
Do you really need a fancy dissection of your every move and interaction to even know what you should be doing with yourself?
[1]: https://openpaths.cc
"Who ran the farthest?"
"Who lost the most weight over the school year"
"Who is eating the best"
And allow competition groups in each between friends as well, e.g. A competition for all students to run or bike, but I can do an individual group challenge with just my friends too...
It might be a good way to get value and start setting good habits at younger ages.
To make it useful, there needs to be a carrot and stick. Some companies do this by way of competitions but then it becomes a resources / management issue.
The opportunity to add value is by intersecting the metrics. As a software engineer who likes to work in cafes, I'd like to know at which cafe I have the highest commit frequency. As a coffee drinker, I'd like to know at which cafe my resting heart rate is the highest (strongest cup!)
EDIT: however I agree on the first point you made. You can't improve something you don't measure. I'm merely pointing at the fact that as of now most tools/products in this space provide nothing actionable beyond the obvious.
I gave up on trying to understand why, and just accepted it.
(That being said, firing up Self Control[0] and a pomodoro timer provide a much bigger boost for me.)
And I say that as a person who absolutely loves pretty graphs. AprilZero's design is exactly what I dreamed of building for my own tracking. But again, it's not the data that matters, it's what you conclude from it.
That being said, I do think that having the data is a prerequisite for improving. The times where I've tracked my food intake, I've inevitably eaten healthier, partially but just knowing how much i'm actually eating.
> If anything, folks I've seen using Fitbit/FuelBand could use a bit more exercise and a better diet. People tracking their sleep activity tend to stay late and have poor sleeping habits (for example).
I strongly suspect that's correlation, not causation. Of course the people who need to lose weight are the ones more likely to buy a Fitbit.
They do marginal changes to their diet, go run for a couple more weeks and then end up going back to their former self. Same as the New Year's resolutions athletes who are gonna hit the gym in January and gone by end of February.
When you're driving a car, you're not staring at the speedometer, or constantly checking the gas meter, but they're still useful tools. Especially when things start to go wrong. If you're almost out of gas or in the redline, then it becomes much more useful.
The other thing to note is that this isn't finished. The new structure — splitting things up by brain, heart, core, fitness, travel — is something we designed to grow into. There are tons of new data points we want to add, which will hopefully improve the actionability. Maybe you already have a sense of how much you've walked today and the step counter may not seem that exciting, but that same logic doesn't apply for things like Vitamin D or Glucose, where having good visibility and software is much more valuable.
Each of the various sections has already had pretty significant effects on me as we've been building this. They are mostly very svbtle changes, like trying to walk more every day instead of taking Ubers after seeing how low my steps were, or realizing how much time during the day I was spending on Twitter after seeing that in the graph, or seeing how much weight I had gained in just a couple weeks and switching to ordering food from Sprig instead of Caviar. Seeing the map of where I went last month has led me to venture out of SOMA more and mix up my routine, which also improves some of the other sections. Theoretically I could've realized and done all those things without seeing any of the data, but it is unlikely.
Some things like heart rate I haven't really done anything about, but it is cool to just see it automatically come in every day. The lowest and resting heart rates are quite variable and seem to be much lower around days when I exercise properly, so that would be a fun thing to try to optimize for and some people have used it that way. My blood pressure also seems to be high, which I haven't really done anything about lately, but having that alert constantly there keeps it on my todo list.
I could see down the line how measuring continuously vital signs and blood work could help us detect or prevent illnesses or risky behaviors, but the complexity of its analysis will demand something a lot more intuitive than a dashboard. As of now though, these are all vanity metrics, just as useful as tracking the number of visitors landing on your front page.
I appreciate the car analogy, but if you car were as well designed as your body, you would not need a dashboard. Getting more sleep, controlling your food intake, getting more exercise, drinking less... It doesn't take a dashboard to know when you should act on it; your body let you know naturally. You get fat, tire easily, yawn, feel like your overate.
Knowing what to do isn't the hard part, your mom probably told you everything you should be doing since you were able to take your own decisions ("Don't stay up late", "Go play outside", "Eat your greens"). But as pretty as our tool is, I see this as useful as the tons of gadgets and fancy sport gears I see people wearing at the gym: a palliative which distracts us from the real hard work.
That is why you want these kinds of metrics, because shit like this can sneak up on you. And losing weight is way more work than gaining weight, at least for me.
1.) Show the data - This is all about selling people on collecting data and using the tool. Example: this product, or any other plain dashboard.
2.) Push actionable data - This is about training people to delegate decisions to the tool. Example: Reminding you to sleep earlier if you chronically undersleep, suggesting an evening jog to wind down from too-much-caffeine.
3.) Act on the data - Customers completely delegate the problem space to your product. Example: Placing Prime orders for food based on your fitness goals, proactively buying melatonin if you take too long to fall asleep, etc.
I don't know what the owner's vision is, but right now it's a very "Dribbble-pretty" product... from the experience and utility perspective, it's still a classic dashboard.
There's nothing wrong with that though! I really like the tool and hope to see it when it hits the more advanced features. But for now... Kibana is my persona-data dashboard.
I enjoy seeing projects like this though, because it means people are analyzing data. Without that, we aren't learning much.
Good job, Anand!
Of course privacy concerns will always be there, but I for one am ok with judicious use of data to further such advancements.
Because that's kind of the point of collecting them. I mean yes, making cool-looking stuff can also be the goal, but one should be explicit about the fact that those are two different goals. Currently, the Quantified Self / fitness toys (aka. wearables) talks a lot about actionable data, while really providing just nice looking toys, often dumbed down to make data not actionable.
I think most people figure it out after spending some time fascinated with pretty charts. At some point you finally realize that all you have are pretty pictures, and you're not really optimizing anything by just looking at them. After this realization, one is finally ready to search for significant improvements. And then comes the enlightenment - tools like dashboards and charts are crutches; the less you need them the better, because the ultimate goal is to have smart decisions, not data, and to minimize the separation between your mind and those decisions.
> Are metrics and stats in games actionable? No, but they are interesting nevertheless.
Usually. Games are good example actually. You can clearly identify the two things a game interface contains - visual clutter, and directly actionable information. Fast-paced, competitive games usually reduce the former and focus on the latter.
What I've noticed about them is that products are optimized to look pretty, not to be useful. Those two things are related, but it matters which you optimize for. Depending on which one you care about, you make different choices about e.g. visualization tools you use. To pick on AprilZero, since it's the topic of this thread, what I see on the site are:
- Charts without scales (or with very impoverished ones) - you can't read too many interesting things from them, whether about time or value of measured parameters. There's data loss.
- Gauge charts, which are cousins of a pie chart, i.e. the thing you use when you care about how nice something looks, as opposed to actually making it useful for learning something from the data.
- The weight chart that seems to be doing exactly the opposite of what it should; splines look pretty, but what you want to aim for is a rolling average, which acts as a low-pass filter, eliminating noise from measurements and revealing your weight trend. See how the line leads the data points? It should be lagging behind them. For an example of what I mean, see [0].
- Just data, lack of any attempt to infer a course of action based on it.
Again, the visual are absolutely stunning and I applaud the work. But now that we've figured out how to make data pretty, we need to focus on how to make it useful. For that, we need to start doing the hard work - understanding what choices we do have and how the data can tell us which decisions to make.
For instance, I could track my weight and things I eat, and draw pretty lines of macronutrients over time, etc. but what I really want to have is something that'll dynamically rebalance my diet to keep them at optimal levels. I want the software to tell me to eat X or Y for breakfast. And if I feel like eating something entirely different, to have it readjust on the fly. I ate too much fat today, so the suggestions for tomorrow will have less fat. An iterated version of this program could plan stuff in advance and generate me a shopping list, etc. You can probably imagine more possible improvements.
But note how in the example above, it turned out that I need exactly zero charts. The input I provide is measurements, but the output I care about is right decisions. Sure, I may like (or need, for something else) to look at the measurement data presented in a nice, visual form. Until my imaginary software is good enough, I might actually need to. But you can see how the focus shifts from just having a line chart to having a representation that will be useful to make decisions.
TL;DR: charts and dashboards are means to an end, not end in itself.
[0] - http://imgur.com/YcXK5af - it's a screenshot of a graph I made for myself when I went on diet a few years back; the blue line connects the actual samples, the green one is a rolling average with 7-sample window. Note how the green line lags behind the blue and how it filters out noise. Incidentally, this is exactly why the common advice given to people is to weigh yourself once a week, and not daily - because it's simpler to tell them that than to teach them about rolling averages and low-pass filters.
For instance, I use RescueTime to measure my productivity, Misfit to count my steps and measure the amount of movement I have throughout the night while sleeping, Spotify + last.fm scrobbling to capture my music listening habits, and a simple csv stored in Dropbox to track my caffeine intake (0 calorie energy drinks and coffee, mostly) and another to track medication intake and yet another to track water intake (in 8 oz increments) and one more to track alcohol consumption (usually measured in glasses of wine or 12 oz bottles of beer, but sometimes shots :P ). I used to track my mood periodically throughout the day, but this was of pretty limited usefulness, so I stopped.
What can I learn from this?
- One energy drink increases my productivity by about 20% as measured by RescueTime over the span of around 3.5 hours. - Working from a coworking space rather than my home office increases my productivity by approximately 25%. - If I consume any caffeine after 3:30pm or so, or more than one glass of one or more than two beers after 9pm, my quality of sleep suffers (evidenced by greater movement throughout the night and not feeling as rested the following morning). - If I walk more than 10,000 steps in a day, I generally sleep better (~10% less movement). - If I drink less than five 8oz cups of water throughout the day, my productivity and mood are lessened. - If I drink greater than ten 8oz cups of water throughout the night, I'm much more likely to wake at around 3am to go to the bathroom, which causes general grogginess through the first half of the following day. - Listening to music with lyrics before ~1pm decreases my productivity by around 15% (instrumental/atmospheric rock before 1pm with BPM greater than 100 increases my productivity by about 5%).
These may seem like inconsequential things, but I'd have never realized those things about myself without the data to back it up (or at least, it would have taken a long while and a good memory).
More importantly, let's say you're trying to lose weight. Without some way of quantifying your current state, it's impossible to know that you're making progress. That's why we have bathroom scales. Does it not then follow that if you're trying to increase the amount of activity you get throughout your day or be more productive or really anything, you need to be able to measure it?
Where we still have a way to go is tracking things at a smaller than day-level granularity — I really want to find relations between things that happen around the same time, like checking in to a specific place vs productivity levels, but that level of smarts is still a fair way off I think. Plus the "actionable" part of "actionable insights" is a whole can of worms in terms of computers telling people what to do.
All the same, exciting times.
[1] https://exist.io
I've always hated running (last one on the tracks in high school, always) often trying but failing to pick up the habit later on during my 20s but only since the first mobile running apps came about I've been running regularly and enjoying it (I still remember my first "runner's high" I had in 2011!). That said I'm one of those old-school arcade gamers who enjoy hi-score hunting so gamification was a huge driver for me personally, people are different - do whatever "floats your boat" :)
There is just one problem I have with the whole "quantified self" (as well as IoT btw) movement which is the fact that the internet is quasi broken, you don't want to "get managed" by any third party - be it your government, insurance company or anyone secretly observing you really.
Things we therefore should be focusing on: down-to-earth complex systems design, cryptography, operating system security, etc.
In general we need to be mindful of the cultural as well as technical details of the systems we've inherited and by knowing and understanding - essentially our history - we should strive for building on top of those systems, making them more secure while still staying true to their original idealistic designs (this still includes probably many more rewrites than we care to face at this time).
Gyroscope's pitch says you can share or instead keep everything private. I think a service like this would be a lot more appealing to me (and it is undeniably cool) if the data were kept encrypted on server, and only available on my end device.
I suppose the same thing could be done with some sort of multiparty destination encryption for sharing.
Until then, my data is just getting monetized for any party that will pay gyroscope. Any comments from your team, aprilzero?
But, you imply that's a tiny minority -- I take exception to that viewpoint. Many 'non-tech' types I talk to have a strong aversion to the idea of this sort of information going out broadly because they rightly believe they don't understand the ins and outs. So, the target market may grow quite a lot with a clear and understandable privacy story.
First of all, we're not selling any data, nor do we ever plan to. We do make money by offering a pro subscription--this new design is one of the features, along with things like custom domains and access to our Healthkit app. We've already gotten a few dozen upgrades today, which is really exciting and lets me confidently say things like "we will never sell your data".
We believe that sharing of this data is an important part of the experience for most people. Of course you don't have to, but for many people that is the first thing they want to do. Perhaps only with your family or a few close friends, and not the whole world, but a single-player experience that only exists on your device would be missing many of the experiences that we think are core to Gyroscope.
A small percent of users have made their profiles public, and many of those have set up a custom domain name and are very excited about sharing it and having a personal website that reflects who they are. Having the data in the cloud makes that use-case possible, though of course doesn't require it.
Our new mobile apps are also taking a slightly different approach, where all of the source data is private but allows for curation and customization, and you can export individual images with aspects that you want people to be able to see. I think that will be the right balance of privacy and control for most people, while still allowing sharing of things they want to highlight.
Making my account completely public has been an interesting experiment. When I first did it last year, I wasn't sure what to expect. Overall it has been very positive experience, and my current opinion is that more transparency is generally good. The location data is probably the most sensitive out of all the things we track. For other people, weight or age have been concerns, and we've added settings to fully hide those.
There are things I consider very private and wouldn't want anyone else to see — like my messages, emails, finances, etc. — but I am personally happy to release the data on the site, like what my heart rate is or how much I slept yesterday. That balance will vary for everyone, which is why I have been experimenting mostly on myself before recommending things to others.
The location data is probably the most sensitive, and we've been playing around with other themes that show different combinations of data and keep more of that private.
We might add other login options in the future, but currently we've found Facebook to be a great experience because we can boot up your basic account info (timezone, name, cover photo, age, etc.) without needing all that to be manually entered. It would be a terrible experience if the first 5 minutes on Gyroscope were just spent typing in basic facts about yourself. One of our big challenges with Gyroscope is how to passively power all this stuff without creating any more work for the user. Our goal has been to go through the whole experience without ever needing to use your keyboard to type anything, maybe with the exception of setting your username.
This wording makes it seem like that data sale starts as soon as your revenue stops making you happy
The point is that there's nothing structurally preventing you from selling data later. Unfortunately that's not a great reassurance, as much as one may want to believe in the project.
To be clear: I think this is an interesting undertaking and you guys are producing good stuff. This structural problem is concerning though, and deserves some serious thought.
E.g., data at rest for private accounts is encrypted with per-account keys based at some level on the user's login, and internally can't be decrypted until the user is signs in next.
I think it's interesting you're publishing your data, and I'm glad you're happy with what you're publishing -- it totally makes sense. User-defined privacy is an awesome thing.
It remains my proposal that you consider actually cutting off your own access to customer data unless they decide to publish it, essentially extending that ability you have while in control of the system to your customers. I think you can see from some of the comments in this chain that people aren't so trusting as you feel you deserve.
And, of course, if someday you or the company needs the revenue, then it might be nice to have already, when not under duress, made a not easily-revocable decision about the ways you'll choose to monetize user data. Or, obviously, it might hamstring your plans.
Talking about site itself, design looks stunning and impressive - lots of work in it. But as tool itself, I do not see much value in it.
I have scales at home to measure my weight, and sports watch to track exercises. Also whole tracking is way too detail. I can see my health trends in no less than a week, but usually in a time scale of months.
Also this tracking sucks my time, attention, etc., and it probably won't tell anything if I have some scary illness growing inside me until I can tell it myself, without any tracking.
Finally, most of devices used have accuracy problems, if you want to track your heart rate in detail you must use a chest band, not a wrist one.
All in all it's just a very very cool toy. It needs too much attention from user. Too many devices to manage/charge/connect/wear, etc.
This is pretty cool and fits in well in 'Hollywood'. Now I want to find a project that I can actually use a similar UI on haha.
Tron Legacy: http://www.robscanlon.com/encom-boardroom/
Guardians of the Galaxy: http://www.robscanlon.com/about/
Wargames: http://www.robscanlon.com/github-wargames/
It's interesting how certain techniques seem to come up over and over again... like they almost always use complex intro/exit animations for every element and additive blending over dark backdrops.
And there's (as best I can tell) no goal setting nor recommendations for meeting goals based on prior successes and failures.
That's still the big gap in self-quantification. In the mean time we're just making prettier charts.
https://instagram.com/p/_H22dosEM9/
I have been using this product as soon as I could join, then joined the company few months after.
There is definite improvement in my health and focus. As a weightlifter, it's especially useful and awesome to see after a work out. Or whatever you do to keep you active. And if you haven't started make a goal of taking a walk or light jog! That's streaks keep your motivated to improve!
Check out other stuff here at https://gyrosco.pe/myusuf3
Give it a shot and use the data to improve, set short term goals and strive towards them, the differences you will see are incredible.
Other than that... Very cool looking design. Looks like it would be interesting to try. I like the integration with other apps and sensors. Lots of potential.
We want to build this as something that anyone can just go and sign up for and start using. I would love to get more athletes using this. We recently released a running app powered by the Gyroscope data — (see gyro.run) — and it's been really exciting to see people use it to share marathons, training, etc. We still don't have a good solution for things like gym training, which are hard to passively track, but that's something we're definitely thinking about.
Just wanted to say thanks for building this out. It'll be great to be able to see custom metrics in the dashboard which the user can enter themselves, and yes, the app can remind the user to enter them. This came to my mind since I can imagine people wanting to track body fat% or other specific body measurements which can't just be picked up by a tracking device, nor can be found in Apple's Health App.
I'm interested in joining this, however I don't have such a wrist device. Which tracker would be best for receiving the most accurate data? Is there a recommended tracker (Fitbit vs. Jawbone vs. etc.) that provides more data, so this app can track more things for the user?
Our new healthkit app (gyrosco.pe/app, currently in beta) loads steps from the iPhone, and we also integrate with other apps like Moves which also run on the phone.
The idea is you can get started within just a few minutes and not need to buy anything. We can also backfill the last year of data so you don't start with an empty site. Then you can start to add hardware to get more data you care about, like a wifi scale or heart rate monitor.
There are many other apps that also write to healthkit, like sleep tracking apps Sleep++ / Sleep Cycle, which will then show up.
http://www.robf.com.au/wp-content/uploads/2011/12/Aliens-A2S...
This application reminds me about a mobile application that is similar with this.
An application that called "Exist" [1]. Exist tracks everything in one place. It tracks your fitness data and other services connected with yours.
[1]: https://exist.io/
Anyhow, if anyone has good info that would help me understand this, I would appreciate it.
It is really impressive site.
In other words it suffer a little bit from the hacking in movies syndrome. The real interesting application is the continues monitoring by the system not by me. That kind of defeats the purpose.
On a different level, it's very much not clear from the site itself what hardware you need to have to get all the various metrics it's displaying.
I am not sure this is your target market, but using TrainingPeaks as an integration will solve a lot of problems for athletes and coaches.
TrainingPeaks is the number one tool today for serious athletes, especially ones that have coaches and teams.
Looks like someone's been working on the dev version :)
I did like the places/travel pages.