Why I just asked Memoto for a refund

I’ve just e-mailed Memoto and asked for a refund of my Kickstarter pledge. I still think it’s an exciting and interesting product, but I don’t believe enough people will be prepared to pay $108 (excluding taxes) per year to use it. It’s ironic, as I can see that one of the arguments for the high subscription fee is sustainability.

I’ve been excited about Memoto ever since I first heard about it at SSWC last August. I became a backer as soon as their Kickstarter campaign launched and I think the team is a bunch of brilliant people. But their recent announcement to charge $108, excluding any taxes, annually to use their cloud service is a step down the wrong path. I know that Memoto didn’t arbitrarily choose a price, that they have been crunching numbers and landed on $108 to build a sustainable company. I think that it’s a huge gamble on enough people being interested in logging all of their lives through the Memoto service, a gamble that I wouldn’t bet on.

People are prepared to pay for storage, software and other services in the cloud. Google, Evernote, Flickr, Dropbox are just a few. And you can combine hardware with a service. Look at Bodymedia users (an activity and sleep tracker for losing weight), who pay over $100 for the device and then $83 per year to use it.

I also know that there is a direct correleation between how useful a service is and how much you’re prepared to pay. To pay $25 per year for Flickr Pro is reasonable. Even $50 a year is reasonable, considering you get 1 TB of data. Paying Google $60 for 200 GB of data for your Gmail, Google Drive and Picasa accounts is reasonable. Paying $135 (which it will amount to in Sweden, as 25% VAT is applied) to store your photos in Memoto’s cloud? The discrepancy between “useful” and “price” is too big for casual users. Especially since you can only store Memoto photos there, nothing else.

There is an option not to pay for the service from year 2. In that case you can download the images from the camera to your own storage. You lose Memoto’s “momentification”, the service that puts all photos that belong to a separate moment together, so you don’t have to browse through the 1440 photos the camera takes during a 12 hour day. But you also lose any GPS data, as that is not stored in the EXIF data of the images. Without processing the images on Memoto’s servers, your images won’t be geotagged. This is the deal breaker for me.

Add to that that there’s big competition on the market. Autographer, a lifelogging device that works a bit differently is starting to ship July 30 (they might beat Memoto to market). It’s still unclear if they will charge you for uncrippled use of their product (I’ve asked them and will update with their answer) Autographer will cost more than the Memoto, £399, but will not require any subscription.  Google+ has a light version of momentification: if you upload a time lapse series of photos, Google+ will put them together into an animation.

Another reason I don’t think that Memoto will reach a sustainable subscription rate at the announced price has to do with function. I got to beta test an early version for a couple days, and I realized that what happens in front of me isn’t as interesting as what happens where I’m looking. The value of the logged data is higher if I find a way of mounting the Memoto on a baseball cap, rather than wearing it on the collar of my sweater. Another challenge that Memoto faces is the angle: when you wear it on your collar or on your chest pocket there is no way of fixating the angle of the camera. A lot of the photos I took during the day, especially while walking was at an angle similar to the image below:

The angle of Memoto when worn on the collar of the shirt

One solution is wearing it as a necklace, another is to wear it further down (requires a shirt or something similar). Or a wide angle lens for the camera. A fear I have is that many Memoto users might be disappointed by this, considering the amount of images of Memoto worn exactly like this there are out there. That disappointment will impact the readiness to pay the subscription fee.

So what do I think Memoto should do? After all, storage does cost. And if you use the camera for 12 hours a day it’s 1440 photos per day. Assume each photo being 500 k and you have over 700 mb of data per day. In a year that’s a whopping 250 gb. So year two you’d be pushing half a terabyte of data. Google would charge you at least $120 per year, Dropbox would take $499. Memoto’s subscription fee for lifeloggers who save all their data is more than fair.

While hardcore lifeloggers are an important target group, there is another one: the casual loggers. This group won’t care for the 8 hours of sitting at the office or the commute to work every day. They will want automatic capture of moments out of the ordinary: travels, dinners with friends, conferences. For this group, Memoto’s subscription fee is outrageously high.

I’d introduce two plans: one flat rate storage plan like the current. $108 excluding taxes is a great price. Let’s call this plan “Lifelogger”. And then I’d have the “Casual memorizer” plan, with limited amount of storage for $30 a year. Once you hit the storage limit, you can either delete (or move!) some of your images, or increase the plan at a premium rate. I believe there to be a lot more “Casual Memorizers” than “Lifeloggers” out there, while Memoto are putting their moey on that there are enough hardcore lifeloggers out there who will buy the camera and pay the annual fee.

I’m not ready to wage my money on it, and therefore I asked for a refund. I still believe in the product, I still believe in the team and if they change their subscription plans to accomodate casual lifeloggers, I’ll gladly come back as a customer.

UPDATE: I just spoke to the team behind Memoto, and they said that they’re still talking internally about different subscription models. Since there’s a chance they’ll go for a lower end, I decided to stay as a backer.

The drawing of the humans comes from Peter Kiselkov’s tutorial of how to draw a human body.

Moves app: great for life-logging, worthless for quantified self

The tag line for Moves app is “Let your iPhone tell you how much you move”. The sad truth is that Moves fails – miserably. It’s a bad idea to use it to quantify yourself, but if you see it as a life logging device for tracking exactly how you move, it’s actually the best one I’ve seen. But let’s start with the activity monitor:

I walk a lot. Last year I averaged 11,000 steps per day, and so far this year I’m averaging 15,000 steps per day. I’ve used several quantified self devices for tracking myself (here’s my run-down of Fitbit, Nike Fuelband, Basis and Philips Directlife) and currently I’m using two: Fitbit and Basis. So during my one week long comparison I had three data streams to compare. Here’s a graph showing the outcome:

Graph showing the number of steps during seven days measured by Basis, Fitbit and Moves app
Graph of number of steps measured by Fitbit (red), Basis (blue) and Move app (green) during a week.

According to my Basis, I walked 102,662 steps in total. My Fitbit says 100,223, or 2.3% less than the Basis. Now, you could argue about which one is more right, but I’d say that a 2% difference is acceptable. According to the Moves app, I walked 80,581 steps. That’s roughly 20% lower than the Fitbit and Basis.

Here’s a graph showing the difference per day between the Basis and the Fitbit (blue line) and the Basis and the Moves app (red line). As you can see, the Fitbit and the Basis have a bigger difference the more steps I’ve taken that day. On the 26th, when I didn’t move that much, the difference is very low. The Moves app, however, has the biggest difference in percent on that day, compared to the Basis monitor.

Graph showing the difference between Fitbit and Basis, and Basis and Moves app in percent
The blue line shows the difference between Fitbit and Basis in percent, the red line between Moves app and Basis in percent

Here’s the difference in absolute steps:

Graph showing the difference between Fitbit and Basis, and Basis and Moves app in steps
The blue line shows the difference between Fitbit and Basis in absolute steps, the red line between Moves app and Basis in absolute steps

The main reason for this is that the Moves app is not an activity tracker. It doesn’t actually measure your movement. You can place it in a bag or purse, and it will still “track your movement”, because it uses GPS. So in effect, where devices like Fitbit, Basis and Nike Fuelband actually measure your body moving, Moves app looks at the GPS to see how far you’ve moved and then calculates how many steps that could be (counting 75-80 centimeters per step from what I can gather).

When looking at the maps to find out why there was such a discrepancy between the Moves app and the Basis, I saw that the Moves app often cuts the corners, it looks as if I’ve been walking right through blocks instead of on the streets. When I compared the distance using Eniro’s maps and its drawing function, the walk to work yesterday morning was 3.9 kilometers while the Moves app reported 3.8. So while the Moves app cutting of corners did impact, the difference is not nearly big enough to explain the roughly 4000 steps that Moves was missing.

Image showing four examples of when the Moves app failed tracking
Four examples of Moves app recording cutting corners.

Maybe it’s the number of steps? Moves reported that my morning walk to work was 4,743 steps. But my Basis reports that during that time, I walked 3,888 steps – 1000 steps less. I compared other walks, and the mystery grew even more. The Moves app seems to be not too far off for each walk:

  • a 3.6 kilometer walk on Sunday: 4,412 steps on Moves and 4,623 on Basis
  • a 3.8 kilometer walk on Thursday: 5,069 steps on Moves and 5,439 on Basis
  • a 2.8 kilometer walk on Wednesday: 3,701 steps on Moves and  3,565 on Basis

The simple conclusion is that Moves only reports longer walks – as it uses GPS. All the daily activity you do indoors doesn’t get registered. All in all, the difference is too large for Moves to even be considered a toy when it comes to counting steps. When you add that there is no way of exporting your data and no web interface (the only way to access is is day by day through the app) it adds up to a quantified self device that places lower than Fitbit, Basis and Nike Fuelband. The only advantage of Moves app is that it is free, but on the other hand it does consume a lot of battery. If you think your iPhone’s battery life is short already, I wouldn’t recommend this app.

Another reason this is not a quantified self device is that it doesn’t care about your heigh, weight, gender or other factors that impact your energy consumption. Therefore, it cannot even estimate how much energy you use for the walk. In that regard, it’s less functional than the Philips Directlife.

But, I did say that it’s one of the best life logging devices? It is, if you want to track at what time you arrive at a certain place. I’m currently using Google Latitude and OpenPaths to track my movements, and while Google Latitude is great for current location sharing (see the right hand column for my current location) and OpenPaths is a great way to visualize your movements on a larger scale, Moves offers the exact time when I get somewhere. And it does that pretty darn well.

You can see what time you started, and what time you arrived:

A screenshot from the Moves app
You get timestamps for each trip/walk so you know when you left and when you arrived.

Here’s an example of what it looks like when you view the details for a trip. Here’s when I took the subway last Saturday morning.

Screenshot from the Moves app
I had no idea what time I got home last Saturday morning, after a night of partying

The best thing is that this logging happens automagically, without me having to remember to check in or start a certain app. I can see lots of uses for this, especially if Moves would let you access your own data.