Strava

What does 'truly own your data' mean?

Relive posted saying it was as soon as they started introducing more social features so it could also be Strava trying to protect their position as ‘the cycling social network’.

Never quite got the point of Relive. Do people really sit down and watch their rides back like some kind of ITV Tdf stage intro?

Once...

To have control over how it is used. By agreeing to their privacy policy, you agree to them using your data for stuff, and not very well-defined stuff either. You no longer have control over what it is used for.

But you can tell them to stop using it and remove it? You have literally given it to them and said "process it" then you can remove it and that's not ownership? They're not taking your data as a backup service, they are going to process it to deliver the features that you thought you wanted when you uploaded it.

They specifically say not everything will be removed, and they'll keep using some data without having it associated with your name.

I don't actually mind though, that was not the point.

But don't you sign all that away in the terms of service? As in, Strava get to do what they want with it while you use the service and until you tell them to delete your data. You own it but you either allow them free reign or you don't in which case you pick up your ball and go home.

But you don't own any work product made using your data and other people's was the point i was making. That's what you said is valuable, and I agree. I think we maybe are talking about different things.

They have to say that because almost certainly they've used everyone's data in some fashion that makes it almost impossible to separate out. If they've got a count of everyone who used Road X and your ride(s) are included in that count, they're technically using your data but there's no way for them to remove you numbers from the count if they're only storing a total. So, saying they will have some anonymous data left over is a catch all for stuff like this. Sure, they could take the piss and just maintain everything but if they were found out the fines are large.

Yeah, that is indeed a problem. I'd have similar issues at work if I was unable to continue using some of the patient data we have - while we can remove that individual's data (that's what someone stores a data key for) from the anonymous-to-me database, there's loads of stuff we've already done with that data or based on that data that you can't just separate out anymore. Which is why the whole idea of 'owning your data' is a bit wobbly anyway.

@stelfox

But don't you sign all that away in the terms of service?

Kinda, partially, yes, as I already said: By agreeing to their privacy policy, you agree to them using your data for stuff.

The novelty soon wears off to be honest.

DCR’s analysis: https://www.dcrainmaker.com/2019/07/strava-relive-cc-app-platform-dispute.html

Ah ok, I thought I could only chop it down, so losing half of it. Will have another look. Ta.

That worked, hadn't spotted it. Ta.

That was a good read. Strava is really being greedy and does not understand they need the partners to Keep the platform competitive. The one thing they do Well is segment leaders , but they seem to think they Can become facebook. It just shows How clueless and lost they are about the future of the platform.

Do they even do that well?
There data cleaning/ management seems really poor.

Activities done in a car should be detectable: long periods travelling along roads where bikes are prohibited, travelling at 40mph or more without going downhill, unrealistic uphill acceleration without a subsequent drop in speed as fatigue kicks in... We shouldn't have to flag this.

Thanks, that is really interesting.

And now I'm wondering if I rest want to pay a few £ a month to a company which acts in this way.

The majority of Strava "features" do not particularly interest me, I just use it as a mileage tracker, photo journal and to compare my own fitness to previous years.

Exactly. They clearly don’t give a shit about the quality of their data.

Also yesterday I got an email stating I’d lost a KOM. Big news as I only have about 4. Only to click through the email to find I wasn’t leader on that segment and the person who had “replaced” me wasn’t either!?

There are signs that their code is probably a nightmare to maintain or add features to.

Occasionally I record a cycle commute as a run (either through rushing to get my watch started or just general brain fade). When I go to upload it to Strava I can change the activity type from "Run" to "Ride" before clicking the Save button, but Strava has already started to process it as a run as soon as it was uploaded.

I immediately get emails/notifications to say I've completed the monthly 10k running challenge, set various new PBs and segment records, etc. These slowly get cleared up but I bet it is the cause of so many false positives that people have about losing KOMs/etc.

They've known about this problem for years (the issue is somewhere on their public bug tracking system) but they've never done anything about it. More than likely it would require a considerable architectural change that is just too much work for them.

I only use Strava as a tracker and I'll never pay for it because it doesn't do anywhere near everything I want. What I really want is a combination of features from Strava, Garmin Connect, Runalyze, VeloViewer, Golden Cheetah, Training Peaks, and something that does proper swimming analysis (haven't found anything that does yet). Luckily I can code something to do everything I want and so I'm slowly doing this in the Copious-Free-Time(TM) I don't have.

I get what you are saying but doesn’t really explain my scenarios chain of events...

Sure, it was just meant as an example of how their current architecture is probably responsible for huge numbers of these kinds of problems.

They tend to kick off an email the second something happens (e.g. your KOM was taken by person X) rather than holding off for a short while in case of any further changes (or fixing some obviously duff data, etc) but by the time you got to look at it person Y had uploaded their ride and taken the KOM from X.

I remember taking a KOM off someone when I uploaded a whole load of my old Audax GPX tracklogs. They would have looked at it and seen that their 2018 KOM was suddenly replaced by someone with a ride that took place in 2009.

Well, you could argue that technically there are some flaws. To me it has worked reasonably well. Everybody who rides is there. Having sais that I think time could start to be ripe for open data solution. You just upload your rides to file service in the cloud and share only with service providers you like.

Just decided to backup my data from Strava (Settings - My Account - Download Your Account) in case they decide to continue these shenigans. If enough people do this maybe they will think twice in the future.

Is there a way to see KOMs you’ve previously held?
And also segments where you hold 2nd or 3rd etc?