Strava abruptly ends 3rd party data sync to Apple Health
dcrainmaker.com
dcrainmaker.com
And I wonder if the article is correct in that they are pivoting away from segments. That seems strange since segments seem like the most compelling feature.
And it apparently isn't valuable enough to charge more for ($5/month)
What are you referring to? That Strava has to store&compute more and more data from a user the longer they are using the platform?
That's not what DCR wrote, that part was clearly about user perception: Strava has grown into something bigger than just that app that declares you KOM when you go really fast.
They paired back things that weren't working, moved to focus on revenue (as they weren't going to be able to grow into the social network for all activities like they tried).
I think it's the right strategy, but this move makes no sense. It's very easy for them to manage duplicate data, and I'm surprised to see them suggest users set-up their garmin device to connect directly to Apple. This essentially makes it easier for strava to be cut out of the loop in the future if the social aspect isn't working for you.
My normal morning jog goes under a large bridge. This often confuses my phone's GPS, leading it to return a spurious point or two a kilometer away. Strava is completely incapable of detecting these spikes (accelerating to 200 km/h when running is normal right?) and offers no way to edit them out afterwards, meaning all my speed and distance records, progression charts etc are ruined by junk data.
Route building and personal heatmaps are some of my favorite features, but there's so much that could be added (like a simple search). To some degree their customers are spread out over so many market segments beyond the original biking and running that they started with. Each segment has their own needs and wants. Then there's all the hardware (watches, power meters, trainers, etc..) that needs to be integrated and supported.
I used it snowboarding and there's some way to track runs and push out the whole day of skiing as one post, but I couldn't figure it out, so I just polluted my own feed with these single run posts. Then for some of them I forget to start recording halfway down the run because once again thanks to the broken auto pause feature, so now I need to remember to start it up at the top of the hill in addition to already having to put on a snowboard binding. Then thanks to the auto pause and me forgetting about pausing the run, it records me shuffling about in the lift line and then riding 25mph on the nose on the gondola.
Or, if you happen to have an Ikon pass, the official Ikon Pass app (May work at ikon pass resorts without the pass — not sure?)
Both know the difference between a run and a lift. Slopes doesn’t quite get pauses. Not sure about how ikon handles it. (I don’t get too obsessive about it). Nice thing about ikon app is that it tells you which runs you took. (Slopes is seemingly adding this but it’s not prime time yet)
I posted it on HN but got flagged[1]. I still find it eye opening, a reminder that the code we write are products impacting the finite life of other people. Maybe we should have more modular, more open software as a standard so that other people can work on and deliver experiences that the original creators fail to deliver.
[0] https://web.archive.org/web/20200601152323/https://support.s...
Most apps/startups begin as an exploration of product/market fit. They'll try a lot of different things and use analytics (even simple server-side stats) to determine what people actually use in the app. Very frequently, you discover that the things you thought your users would want are actually only used by 0.05% of your customers. Eventually you have to start shedding rarely used features and limiting free plans, even if it makes the non-paying users angry.
Truth is, it doesn't really matter if you're losing someone who spend 5 years on the free plan but refused to sign up for the paid plan. They're not converting to paid unless they're forced to, and you're not gaining any money by letting them stay on the free plan for another 5 years.
> but where are all the new and cool features that should have come out over the last few years?
Cutting rarely-used features like this one could be a sign that they're trying to free up engineering resources to ship new features.
This is true, but it ignores the importance of network effects in products based on social media/interaction. I'd guess that a huge part of the attraction and user retention ability of Strava is the social aspect, which they'd be crippling if free users migrated somewhere else. As a secondary issue, Strava also benefits from the data generated by free users, though I have no idea what the value of that data might be.
The Strava/Garmin relationship is an interesting one: Strava has the social network, Garmin has the best devices [for serious Athletes, not casual users]. Garmin Connect is pretty cool in it's own right, but the Social features never really took off, which is where Strava plays and has a de-fact monopoly. Strava can't survive without Garmin, and Garmin benefits from Strava's content.
Garmin has pulled some 'power moves' in the past though with people it doesn't like... A competitor: Wahoo, who made cycling GPS computers, was cut off from inserting data into Garmin Connect and it left a lot of users out in the cold. Most serious cyclists will use Garmin devices, not an Apple watch, to track their rides as it seamlessly integrates with ANT+ sensors: power, cadence, wheel speed, heart rate, chainring and cog positions sensors.
Interesting to see Strava cut off Apple... I'm guessing it has to do something with preventing them from developing an alternative to the Strava social network.
> Garmin is the only one I actually trust currently, as they've shown 0 willingness to monetize your health information.
What about the concern that the company is bought out in the future? Or that it may be sharing data already, with government agencies, etc? Or companies that it works with?
I honestly can't even imagine the criteria whereby a corporation could be 'trusted' with personal information!
It boggles my mind, that people think about this - and plainly you do - but come to the judgement that its ok for corporations to have this personal data!
Apple had a chance to make on-premise data storage a thing. They had AirPort, which combined with a Mac Mini could provide full iCloud functionality from the home. iCloud itself could have simply been an encrypted offsite backup, like Tarsnap with no options.
Instead Apple built a system that still puts clear user data within their reach.
> When your device is locked with a passcode, Touch ID, or Face ID, all of your health and fitness data in the Health app — other than your Medical ID — is encrypted and inaccessible by default. Additionally, if you are using iOS 12 or later and turn on two-factor authentication, Apple will not be able to read your health and activity data synced to iCloud.
The two factor authentication is the mechanism for key distribution such that HealthKit data can be synced and migrated between your own devices, without being decryptable by apple.
To me, it comes down to me being fine with this data being considered public. If others don't want to share it, that's totally fine and I understand their perspective. For me though? I couldn't care less.
Congrats, you sold some data that says I'm fat, out of shape, and don't like riding my bike on busy roads.
I can only imagine anyone buying the data does so only if they believe they can make more using the data. I don't think I know of anyone doing this now (not sure if there is regulation preventing this), but its not hard to imagine a eventuality where your health data/work out information can be used by your health insurance provider.
Imagine a scenario, your bmi is higher than normal and is trending higher in the past few months, so that increases your risk for x and hence your insurance premiums go up. Would you care then?
> What about the concern that the company is bought out in the future? Or that it may be sharing data already, with government agencies, etc? Or companies that it works with? > It boggles my mind, that people ... come to the judgement that its ok for corporations to have this personal data!
I'm not OP. I chose Garmin precisely because I don't think it's ok for corporations to have this personal data.
I chose Garmin because I can use it without needing to share any* of my data with a third party, including Garmin itself. I save the workout files to my computer via the same USB cable I use to charge the watch. There are various non-cloud apps I can use to view and analyze these workouts, if I actually cared to do so beyond the "fastest 5k" etc that the watch tracks automatically.
For this, it does not matter to me whether Garmin is bought out in the future because my watch works just fine as-is, and cannot update its OS without my explicit permission. I'm unsure what data Garmin would be capable of sharing because I have given it none.
* of course, speed/time/location data is obtained by pinging GPS/GLONASS satellites, but the watch can and does record my workouts quite accurately in non-GPS pedometer mode.
You could of course access the .FIT files directly on your device via USB, so maybe that's what you meant. Their hardware is pretty good for working off grid but you need other software to do anything with the data it creates unless you're connected to their servers.
Isn't that to be expected, if you're using their cloud services? The point is that Garmin is (or was, I've had my watch for a few years) designed in such a way that the cloud services are not a requirement for full use of the device.
Yes, you can access the .FIT files directly via USB and then sync with other apps, such as TurtleSport, GPXsee, etc.
For the most part, yes, though I don't think it is particularly unreasonable to expect that you could view some of your recent activities in the app without having to phone home, given how small the .FIT files tend to be.
Syncing with other apps is also something I would expect could be done without phoning home, assuming the other app is capable of parsing your .FIT file directly (which Strava is). In that case all the app needs to do is fetch the .FIT over Bluetooth and pass to the third party app using whatever phone API.
I’m a happy Garmin user but I know a lot of distance runners that have switched to Coros. They love the battery life w/o sacrificing GPS accuracy.
I guess their huge data breach that they didn't notify and spent weeks fixing doesn't phase you, but it punted me out of their system.
Strava -> Apple Health, still works
Garmin/Fitbod/3rd party app -> Strava -> Apple Health, no longer syncs.
IMHO this is better for me because now I don’t have duplicates in Apple Health and can sync to both services. But to each they’re own I guess.
BTW and a bit OT, I find it very impressive that Strava can retroactively create a leaderboard going back years for a newly created segment, meaning that they must evaluate potentially millions of nearby activities for overlaps, often in just a few minutes. That's a hell of a query. Anybody know of more information on how they do it?
Even if it is a couple orders of magnitude larger than I think geographic partitioning can keep the volume small enough to easily fit in RAM.
Parallelism along the location, customer and activity dimensions makes this an easily reducible problem.
Not for a segment that doesn't exist. When I create a segment, it goes back over all nearby activities since the dawn of Strava time. That's the part where geographic sharding helps. Then, it has to check each one to see whether it traversed the segment without interruption or deviation - not within one meter but AFAICT within only a few. Does further sharding/tiling help there? Yes, and I'm sure that's part of how it works, but I'm interested in exactly how they apply those techniques and then solve all of the remaining problems - undoubtedly including those that neither you nor I have thought of. As you yourself say, there are probably some clever choices involved, and that's the part I'm curious about. It definitely seems like a "meaty" enough topic to fill out a blog post or two, and the world needs more such IMO.
I was actually looking into this myself, and a few parts of strava are not that hard. For me the overview of my rides (including avg speed and distance is important), I quickly implemented a webpage with javascript and leaflet (map viewer), and I could present that data pretty quickly using geojson. I had an offline program which converted gpx traces to geojson, but I'm pretty sure I'm able to read gpx in javascript directly. The harder part is generating the gpx trace (which I used komoot app for), as this involves matching your trace (gps positions) on a map instead of using the data as is.
* the database of "segments" created by users
* the leaderboard data for each segment
* the network-effects of being "the place to record data"
But a self hosted tool still can do a lot for own analysis. For tracking the own progress, the own training plan.
And after all this, your user base consists of the people who didn't want to pay for strava.
I guess you might be able to do interesting dashboards with grafana.
Progress is slow as a solo dev who contributes to it in their spare time, but I’m happy enough with it that I could archive my Strava account and switch to it fully.
It seems that Strava agrees with me - which is a bit odd.
Even for route planning (years ago) which I would not say Strava is fantastic at, it still seemed worse.
YMMV
Disclaimer - my site. It’s very focused on quiet routes, so will generally prefer a hilly road if the alternative is a traffic-infested one. Europe/North America/AU+NZ only at present.
UX observation: clicking to create a start and end was relatively intuitive, but it took a while to find "Close route" to reset it (and be able to click for start / end points once more).
How do you determine quiet-ness? Is it just by the grade of the road, or do you have some kind of data feed for quietitude?
E.g. in the UK, some A-roads can be very quiet despite being technically major, whereas some B-roads or even ungraded can be very busy, e.g. if they are used as shortcuts.
No.
Google Maps directions for bikes is a good start but you have to take the routes with a big grain of salt (even with that Google is still better than every other app I've tried). I street view the route to make sure there's a least a good shoulder and re-route appropriately.
My workflow is 3 monitor setup. Komoot in the middle, actually planning the route. Strava heatmap on the left to see where people often travel in an area. Gmaps on the right to see streets and street view.
It's still a pain to plan, but it's _so much easier_ with komoot.
Used by ultra-racers to plan multi-day events.
My only problem is, that when I select 'road ride', it tries to get me on a bike path. That would be a good thing for some people.
Even more importantly, Strava is a platform where friends can encourage each other, and celebrate each other's accomplishments. I greatly appreciate the "kudos" I get from others, and that was especially true after I had a setback some years ago. I have a friend right now who had an even more serious setback, and I'm glad I have a way to support him as he logs his rehab activities. It's the only forum where we're connected (he's not very online the way I am) and it enriches both of our lives.
I know a lot of people think Strava is just about getting KoMs to satisfy ego. I submit that it's often projection or bitterness from people who don't get what they want out of it. Seems like a miserable way to be.
This was once nice for me, but at some point it was obvious that many simply kudos everything.
However for a while it was quite a motivation to go, just to keep the public image of myself active. ;)
This one:
1. I don't use Strava because I am doing things for myself and not the approval of my peers
I shall shelve this along with my other HN faves which are (paraphrased):
1. I am remote because unlike everyone who goes to the office I have a life with hobbies and I don't like having friends only from work
2. For those of us who own a phone for productive reasons and not for social "clout", iPhones are terrible
3. If you want to do work and not just sit around in coffee shops posting about work, you would avoid a Macbook
Very enjoyable.
My impression of Apple has been that it's a mixed-bag for privacy, etc., so I'm actually not sure.
For privacy, certainly they have been a crusader. For open standards, considerably less so.