I have been using the Pebble Time smart watch for four months now and have posted several times about integration of step-data from the Pebble to other ecosystems. Most recently, I described my forays into using those steps as part of diet tracking. That application proved to be the proverbial straw that broke the camel's back.
Let me first say that I absolutely love the Pebble as a smart watch. Its ergonomics, comfort, user-interface experience, and price are all outstanding. The smart notification handling is especially well done and the open time-line concept was great, even if artificially chopped off after two days. On the watch itself, the activity and sleep tracking are also very accurate. The problem is that Pebble has basically provided no ecosystem for this data.
Other vendors have chosen to support Apple HealthKit as a secondary means of exporting and sharing health data. Pebble, on the other hand, has made that their only ecosystem (on Android devices, they use Google Fit for which I cannot comment). By going this route, Pebble provides no reliable system of record for my data. The only storage is in HealthKit on the phone, a platform which seems clever at first but which I don't think is not up to the task of managing data exchange. There is no on-line access to the information so data visualization is limited to the piss-poor dashboard provided by Apple ... worthless. Also, the whole social experience of comparing activity is precluded.
Data backup and restore is seriously flawed. I have been screwed by this twice now. In one case, my Pebble installation was so stuffed that I had to delete and reinstall it. In the process, I lost all of my health data and had to reenter it from a spreadsheet I had luckily exported manually a few days prior. The only true backup facility is to rely on the iPhone backup and to do a full restore of the phone. That means reverting everything back to the last backup point, not just the health data and certainly not selective data. This is not an acceptable solution. With Fitbit, Garmin, or other activity tracker ecosystems I have used, once the data gets to the cloud, I don't have to worry about it ... someone else takes care of it.
Integration on the phone just does not make sense as an architecture. An iPhone is a client device ... it is not a server and it sure isn't a workable event fabric. Integrating through HealthKit means that I have to: ensure that all of the relevant apps are launched, ensure that I tickle each of the apps during the day and do so in the appropriate order, worry about whether a data pull happens before the necessary data push, deal with iOS background eventing not working correctly. Thes are all problems that have plagued me during this experiment. Sorry, this HealthKit solution just doesn't cut it.
Though I had gotten my multi-hop solution working for end-of-year reporting of steps to my employer, the solution fell apart completely with diet tracking. In order to correctly plan my evening meals, I need to know how many calories I have available. Therefore, I need up-to-date step information pushed into MyFitnessPal. Unfortunately, when all was working smoothly, my steps arrived 2 hours late. Often, my steps would not arrive until the next morning. Even more frustrating were instances where several thousand steps never made it across. This was a case of pull-before-push and data reconciliation required un-installation and re-installation of apps on the phone. Are you kidding?!
Forget it, this integration experiment is a failure. Despite the woefully inadequate smart-notification support, I have switched back to using a Fitbit device. I don't want to cover my arm with devices, so the Pebble goes in the junk pile. My conclusion from this experiment is that, if Pebble wants to be a credible player in the activity-tracking market, it must create its own on-line ecosystem and it must forge partnerships with other key players such as MyFitnessPal to integrate their data in the cloud. As it stands, they just have a fancy pedometer, not a fitness solution.