Knowing how many tracks you listen to a day is interesting trivia only.
Knowing how long you sleep for each night could be useful, but only if you have a problem with sleep and know, scientifically, what you're looking for. If you're consistently missing out on sleep and feeling ill as a result, you shouldn't need exist.io to tell you this. And if you're not feeling unwell or having sleep problems, then you shouldn't need exist.io to tell you this either.
Knowing the number of steps you take is also irrelevant unless it's part of some very slow burn exercise regime you're on, in which case you'd presumably count them anyway. If you're doing normal exercise, steps and heart-rate may be of interest, but only during the exercise period.
In short, knowing all of this info about yourself is not helpful. When it lacks greater context and, unless you know how to interpret it, is just a source of confusion and anxiety. At best, it's novel trivia.
What I find more important is tracking what you did in terms of achieving the bigger goals you have. But tracking might not be important here if you have a good system that makes you do work on regular basis to achieve those goals.
I have a choice on a daily basis whether to walk to work, catch the bus all the way or do some combination of the two. It used to be easy to convince myself that I was walking further than I actually did (things like "I only caught the bus half way today, that's almost the same as walking the whole distance").
But when I've got an app tracking me every step of the way, that's not possible. I can get to Thursday, check my app and realise that I'm several miles short of where I should be, so if I'm going to hit my targets for the week I'd better get walking.
If anything, I'd love to be able to capture more stuff about my life automatically - it would be a lot harder to convince myself that the doughnut I've been given doesn't count because it was free (or whatever my latest lie to myself is) if I've got an app telling me otherwise.
Most people lie to themselves about things like the amount of exercise they do, the amount of junk food they eat and the amount of alcohol they drink. If they could have the truth presented to them as cold hard fact, that would probably be quite an eye opener.
- Do I sleep better if I don't listen to techno music before going to bed?
- Does not exercising late in the evening help?
- How about not drinking coffee in the afternoon?
- Lowering the room temperature?
Sometimes the effects are obvious, but often they are subtle, and it's easy to delude yourself (for a while at least) that a solution is helping when in fact it isn't.
This is valuable information, but people are different, so the question is, to what extent does this apply to me?
Genetics might help predict this (rs762551 ?), but in practice I'll just need to test this for myself.
The notion that there aren't always one-size-fits-all solutions is well accepted (though often still not well implemented) in medicine (a.k.a. personalized medicine).
"Need at least thousands ppl's stats to get a small conclusion". - Not necessarily. It depends on the design of your experiment, the distribution of the data under question, and the desired margin of error. (Amongst other things)
But even if you only roughly estimate those things, you can gain conclusions that are valuable to yourself.
To make up an overly simplistic example: I spend two weeks not drinking coffee, two weeks drinking coffee in the morning, and two weeks drinking coffee in the evening. If the amount and quality of my sleep do not change significantly between the three experiments, it's safe enough to assume that I can consume coffee without influencing my sleep.
It's not a sound experiment, because it doesn't control for any number of variables. It has a wide margin of error. I have not really looked at the sample distribution of sleep duration/quality. And yet, I can make the assumption that drinking coffee in the evening will not affect my sleep.
The point is, it doesn't matter if it's 100% correct. It's roughly correct, most of the time - which is good enough to plan your life around.
It's of course not good enough to make any statements about the general impact of coffee on sleep patterns. But that's not what self-measurement tries to achieve.
By "embedded in context" I mean you need to not just know a larger constellation of interesting information (steps, tracks, mood, location, &c) but also have available information that connects the various data points meaningfully—"I'm blue so I listened to The Cure".
By "consumed experimentally" I mean that you have to have a lever or action which you want to use conditioned on the information gathered. Simply knowing is possibly boring or creepy, but being able to change your own behavior using that information is power. For instance, the opposite causal pathway as above: "I'm listening to The Cure, so now I'm feeling blue" implies a lever you could pull to feel happier. This drives you to want to experiment.
If you have both of those then you're driven to use data like this experimentally and that's pretty wonderful. To anyone who also agrees or is just interested in what you can do with data like this I suggest taking a look at Ian Eslick's "Personal Experiments"  which helps direct a user to collect data and invoke an N-of-1 experimental design atop the collection.
One practical application that I have found to Moves (one of the tools listed there, and the most used one) is journaling: not really detailing how much calories I have burned, bus simply where I was, and what I’ve done. The integration with FourSquare is great to help me remember “that great little restaurant I when to when I was in Barcelona… was that in June or April?”
In my experience — admittedly anecdotal and (I wish on everything that is precious) representative to no one — I torn my knee a year ago: ACL, the very bad kind. The pain… I spent a month in bed, convinced for the first two weeks I might not ever walk on my own (that wasn’t remotely possible, and sometimes making a jockish surgeon burst out laughing at your naïveté is all you need to feel oddly better). I still couldn't barely walk and had to: exercise, as soon and as hard as you can, is the only treatment. Moves really helped me track my every step there, and a dozen stepped counted at the time. I am emotionally quite fragile in general, more so then (because the accident had cost me a good client, delayed an interview with an amazing company, forced my long-distance girlfriend and I to cancel holidays together, ridden me to a windowless bed while Spring was finally out) and the pain… Without that bouncy blue dot proclaiming 360 steps was a ‘Personal record!’ I wouldn’t have gone through it nearly as well. Without the app to track on its own the time it took me to tumble on crutches to my very good, very busy and very time-conscious physical therapist (a block away, that is… 20 minutes the first days), I would have had no idea how to be there on time. Not representative, indeed. Bit not useless.
None of the tracking apps that I’ve used have had the expected impact: some revealed to me how my psyche was at stake more than training, some that GPS technology sucks, or that good design is still direly needed… None (with the notable exception of Moves) I’ve used durably, but all have had a positive, eye-opening experience on me. Often, help me realise what you describe quite well. But experiencing it still beats reading about it.
Isn't that the service that exist.io is hoping to provide? It seems to me like their mission is providing people with greater context than just a bunch of data points. They provide the expertise in interpreting this kind of data so that people don't need to analyze it themselves.
Now, I'm not convinced that they'll be able to provide actionable ways for people to improve their lives, but I certainly hope they can! We generate a lot of data w/out much effort; if we can also use it to improve our lives w/out much effort that's a big win from my perspective.
Isn't that part of the entrepreneurial spirit: having the courage to build something even though you don't yet have all the answers?
Fitbit assigns me a goal of 10,000 steps per day. I can view the data and work to reach that goal. By being more active I burn more calories, allowing me to lose weight.
Gamification in service of nudging people towards healthier behaviors has value. It's a means to an end. Sometimes just having a concrete goal with a progress meter is all it takes to get a person to make a small but important change.
However, it seems to me that 90% of people doing anything in this space have the same idea: track as much as possible and then try and see if there's anything interesting there.
"44% more likely to check in when you climb more floors" -- big whoop! I suspect that this is exactly the level of insights that you are likely to gain from that sort of approach. You will discover that you are indeed less productive if you don't sleep well or that you tend to have headaches after drinking large amounts of alcohol.
Exploratory data analysis is appealing because we like the idea of discovering something completely unexpected and revolutionary but I think confirmatory analysis (the boring one where you form a hypothesis and set up an experiment to test it) is much more likely to be useful.
That said, I also just signed up for for Exist.io; the screenshots look great, and I'm curious to see where this leads!
What we're doing works a bit like a Kickstarter. We're asking supporters to back
us for AU$60.
Why charge for 1,000 places?
The maths on 1,000 × 60 means two co-founders working full-time for six months.
This is just enough time for us to get stuck in without worrying about the future
It's also a handy, low number that means we don't have to worry about scaling a
whole lot of data analysis just yet.
The name Exist is creepy. As if, you don't truly exist until you can pigeonhole yourself into a couple of statistics. I swear, everyone here is mind blind.
> SHARE WITH FRIENDS
Because highly personal data needs to be shared with your friends. Just because.
Actually the real feature here would be "SHARE WITH MOM" for those mothers who always call to ask "Are you getting enough sleep?"
I agree that tweets and tracks probably won’t help you notice any useful patterns. But I don’t see the name Exist as creepy. I interpret it rather as a way of saying that this is very passive software, and all you have to do to use it is to exist normally, as if it’s not there. Like existence is the context of this software, not the product they are selling.
I dare you to try talking to a stranger on the subway regarding a topic requiring any kind of methodical or critical thought. You're going to get a lot of glassy eyed vacant stares, even controlling for the fact that they don't know you.
Connect literally five other services for a broader view
Is happiness really found in measurable things? How am I supposed to measure how happy it makes me when my son grins ear to ear when I get home from work?
Happiness, as far as I'm concerned, is found in contentment, love and purpose. I don't see a way to measure any of these things. And I certainly don't need a website to tell me if I have them.
On the other hand, these products can help you achieve some "thing" that might make you happier. For instance, being fit might make you happy, and seeing how far you traveled each day may motivate you to walk further every day.
But the amount of steps you walk shouldn't define your happiness. And if you try to draw happiness from that, I would guess it's going to be a failed endeavor.
I had gotten into QS for a bit, but as others have said, once you start to use the products and follow the methodology, you start to see how what sounds great in theory doesn't work so well in practice.
That said, I hope they do well, because, why not. I just know that this isn't as interesting to me as it would have been 6 months ago, having some experience in the space.
"Quantified self" is a bit like trying to turn a quantum phenomenon into classical physics by taking precise measurements and bounding the parameters until some structure appears to emerge. You've got a theory then, but instead of describing the original phenomenon it's an image of the choices you made to get that data.
So busy quantifying your life, you forget to actually live your life. My 2 cents.
I imagine as more computer 'clothing' comes out things like this will become more common because more knowledge at no more time expense can't be a bad thing.
I'm really curious.
I think QS is probably more popular in communities like hacker news because there are a lot of people here that understand:
1. Computers think faster than you
2. Computers remember more than you
3. You can use the above to solve interesting problems.
Combine that with pervasive computing/surveillance and you enable yourself to remember arbitrary amounts of things and mine insights from them.
Empirically, it's probably a young person's thing, because people tend towards conservatism with age, and the technology required is going to be something young people are more comfortable with. But there are plenty of older people involved in QS.
Not yet. But they make calculations faster.
I use QS tools to define where some degree of normality is, so my simple monkey brain doesn't lie to me and I can see when I cross that line :).
Left to my own devices I'll end up eating out for all 1-2 meals a day, only move from the bed to my car to my desk, forget what sunlight looks like, pull 12 hour days and 4 hour nights of sleep (or stumble into a 48 hour cycle), IM only my closest friend every few months, get suicidal and quit another job because nothing matters and the world is cruel.
For me at least, quantified self is not about trying to treat myself like a machine, it's about trying to remember not to.
I find this sort of thing especially helpful when trying new medications/treatments. It's nice to have a chart that I can look over to help give me a better idea about what's going on with my body. "Oh, ever since I started this new medication, I get headaches when I don't drink at least 4L of water."
There's no way that I'd be able to keep track of my day-to-day symptoms without writing them down; the days often blur together.
Oh, and doctors love it, too.
The examples on the site include how many tweets you wrote (ok), tracks you listened to (erm... ok), checkins you had, and the max degrees Celsius for your location that day (what?).
In general, you optimize what you measure, but eventually it becomes a parody of itself.
But number of tweets, or number of checkins per day? Nah.. or maybe it's a play on something Oscar Wilde wrote.. "To live is the rarest thing in the world. Most people exist, that is all." Not that I mean to be snarky, I assume it's up to the user what stats to display and hopefully how to display them; but to get me excited, it would have to be the "Piwik version" of itself, if you catch my drift.
Makes me think of the concept that telling someone your idea while leave you less likely to act on it, as you'll feel like you've taken a significant step already.
I not so strictly measure my food intake, but I know enough now to correlate certain diet choices with my effectiveness in the gym.
Things I don't measure but would like to are various hormone levels and other health markers that could probably be worked out with a blood test. I take a number of foods and supplements that are meant to maintain a decent hormone profile - high testosterone, low oestrogen, cortisol etc.. But because I don't measure these I have no idea really if I am doing it right or not.
Another thing I really should measure is my sleep. Although obvious that if I feel dog tired I won't be as effective, I really should measure how much of a sleep deficit I can tolerate so I can balance my work/project time with sleep time effectively.
What foods and health supplements do you take to get that good hormone profile?
What foods have you found correlate to gym performance?
High fatty foods are supposedly good for testosterone production. Lower glycemic foods and higher protein is good for reducing cortisol production.
I have experimented with stuff like DHEA, Indole-3-carbinol and avena sativa for more testosterone and less estrogen. But as I haven't measured I can't quantify any difference it made.
Higher carbs seems to make me stronger, but also makes me fatter and less energetic post lunchtime. I tend to move between high fat and high carb depending on what phase of weightlifting I am in, strength building or fat reduction.
Something I haven't measured, but keep meaning to is how much of an effect total calories consumed has compared to the nutritional breakdown of those calories. It could be higher carbs makes me stronger because it is just an easier way to get more calories in.
If I could measure these things more finely rater than just speculate I could work out how to optimise my intake.
About the sleep app and uncomfortable having your phone in bed - turn on flight mode?
The main sleep app on IOS is usually $0.99 and insightful. Try it tonight. The chart after a night of sleep alone is worth it.
Full write up here: https://garrickvanburen.com/archive/how-i-learned-to-get-up-...
30k per founder for 6 months seems like quite a lot of money.
Maybe the startup founder's lot in life is to be poor while they are starting up, so could probably do it with less. But they are hardly overstepping the mark.
1. Pay yourself the barest minimum you can to survive, which may be 0.
2. Pay yourself just enough that you're not worried about money.
High founder salaries, especially in early stages goes against many investment theses.
I'm taking the founders at their word--they need the money to launch their product. It just seems like they could do it for less--however I don't have much detail so I'd like to understand the money breakdown.
"figuratively.io - a metaphorical API clearinghouse"
YC15 if you want to co-found with me.
Actually, even if the data isn't big, it's still _really_ hard to figure out causation. If exist.io does make it, It'll be because they figure out just ONE kind of reliable causative relationship they pull out of their data, and they figure out how too monetize that....
This is actually a pretty clever way to crowdsource THAT process though...
N times per day, for M seconds, turn on all my phone's sensors and take a short video capture of my computer screen.
Then upload that data (preferably to my own server, not yours).
Every couple days, I'll log on and tag the data. At 17:56 on 12/05, I was sleeping, working on X, talking to Y, going to Z, etc.
Sample 3-4 times a day for a year. Should get a really clear and interesting picture of your life after 3 months or so.
The only odd part is the backer model: unless they have over 1,000 users on the waiting list who have committed to buying it, it seems like they should have a prominent button to pay and start using it somewhere, rather than email invite.
So many assumptions.
I don't see this being taken seriously, probably because it is impossible, because there is too many data, about too many things, and each action we take generate some "data" that can be tracked in infinitely different categories, so stop claiming that you'll do this.
Having cleared that, I gotta say, it does look awesome and I wish you best of luck :) !