Tracking how users use your own app is hardly malicious, in fact it's entirely in an effort to identify user frustration and solve it or otherwise improve the product. Tracking across different apps is definitely malicious, though.

Tracking how users use their own phones is certainly malicious. Regardless of whether they are using an application that you wrote, it is their phone.

That is not what is happening here. Users' on-screen taps inside the App Store app are being tracked. Not their interactions with any other aspect of the phone.

He/she is referring to parents comment

>tracking how users use your own app is hardly malicious

Where, pray tell, are the user's on-screen taps occurring, if not on the user's phone?

You can justify a lot of crazy behavior with "we need it to improve our product", that doesn't make it inherently ethical though.

That also doesn't make it inherently unethical, and so far nobody has provided evidence of that claim.

The Universal Declaration of Human Rights says something about arbitrary interference in a person's privacy not being okay, so there's that.

https://www.un.org/en/about-us/universal-declaration-of-huma...

Not convinced. By your logic, an nginx log with your IP in it is a similarly egregious and unethical privacy violation.

If you're going to commit to "all forms of tracking are bad no matter what" I'm going to need stronger evidence than a link a page on the UN's website.

Saving that IP to the webserver's log is rather necessary for it to function, whereas these detailed records of user behaviour can almost never be.

Just because someone uses an application, does not mean you "own" them and have a moral right to freely collect and do whatever you want with records of their behaviour. Also it's just weird.

Please explain how webserver logging is "necessary for it to function".

Perhaps you'll find this book interesting: Networking All-in-One For Dummies https://www.amazon.com/gp/product/B0921K9YTS

So I'll take that as you can't demonstrate it?

Interestingly, I started up an Nginx server and piped the logs to /dev/null and it worked just fine!!

Wow, turns out those logs weren't necessary at all. And by your standard, where tracking that isn't necessary for the website to function is unethical, that makes every single person running Nginx in its default configuration bad and participating in unethical tracking of users!

Because it doesn't seem like I'm going to get a response to this, I'll walk down the rest of the logic for completeness's sake.

I think we can both agree that "all unnecessary tracking is unethical" doesn't make sense and leads to insane conclusions like "all Nginx users with IP addresses in their logs are participating in unethical tracking".

So we're back to square one, where someone needs to demonstrate this type of tracking is actually unethical. So far we've got "all tracking ever is unethical" and "all tracking not necessary for function is unethical", but both of those seem to be dead ends.

I think the argument is a bit nuanced, but I see where the "tracking is bad" comes from.

I think everyone would agree that it would be very bad if eg. an instant messaging app logged message contents and allowed devs to look at anyones messages.

On the other hand, most people would not object to an app that sends an anonymous weekly counter of which buttons were tapped.

Anything between these two takes a lot of care and deliberation; on the one hand devs want useable data, on the other hand users don't want sensitive data tracked.

But it takes a lot of effort to make sure you don't accidentally log sensitive data. Eg. if you log spell check suggestions (to determine common typos) you might probably be able to use that data to reconstruct message contents.

Now, as a user, do I trust random companies to get this right? Do I trust random companies to not accidentally log sensitive data?

And the answer to that question is NO, I don't trust them. So if you ask me if I would consent to activity tracking, I would almost definitely say no, even though I don't think anything is wrong with anonymous usage tracking -- I just don't trust every company to get this right, so I'd rather have them not track anything at all.

> On the other hand, most people would not object to an app that sends an anonymous weekly counter of which buttons were tapped.

Considering how often companies misuse even this data...

Of course I agree there is nuance, but that's the point I'm trying to make. So far every argument I've seen against it in this thread is along the lines of "all tracking is bad" or "[overly broad category of tracking] is bad". Nobody seems to be willing to substantiate the more nuanced claim of why this specific type of tracking is bad.

For the record, I agree that this type of tracking should require consent (or at least be able to be turned off).

The following is meant as constructive feedback, and I hope you will take it as such even if it happens you disagree:

Your comments here consistently come across as highly adversarial: Assumptions and conclusions are liberally asserted, every sentence only attacks, there is no room for nuance, and attempts to find common ground are either mistaken for a new enemy assault or remain unrecognised entirely.

This makes for an uninviting and frustrating conversation, which is a real shame because you are clearly not a stupid person.

Admittedly my previous comment was not very constructive either. An unfortunate outing of my frustrations at the time.

Again, I post this with genuinely positive intent, and hope that when we next cross paths it might lead to a more constructive and enjoyable talk!

In this spirit, I have some constructive criticism for you. Comments like this (https://news.ycombinator.com/item?id=33528012) are so vague I am forced to make assumptions about what you're trying to say.

You post an extremely vague comment, I try to muster a response by making some assumptions about what you're trying to say. You post something that I think is obviously wrong, but I ask for an explanation just in case I am wrong or misunderstanding you, and you post this:

> Perhaps you'll find this book interesting: Networking All-in-One For Dummies https://www.amazon.com/gp/product/B0921K9YTS

I'll admit my comments were somewhat aggressive, but up until this comment (which, for the record, has changed my mind on this) you seemed to be participating in blatant bad faith.

If you want people to understand the nuance in your position, you have to actually say it, not just provide a random link to the UN Declaration of Human Rights. Your original comment seemed to be genuinely making the claim that 'all forms of tracking are bad because they are a privacy violation'. Maybe that's not what you meant, but I have to make assumptions about what you're trying to say when you don't actually say it and leave me to interpret what your quote and link are meant to say.

Also, for the record, I still disagree this type of tracking is unethical.

I'd love to hear the nuanced version of the argument why it's bad, maybe I'd be convinced.

You're taking a full black-or-white approach, which isn't helpful. Yes, an nginx log contains personal data (which is why a GDPR-compliant nginx log obfuscates the IP, absent of user consent). But tracking user interactions is much more invasive, which is why the parent posters complained about it.

I'm not taking a black/white approach, as a matter of fact I'm the one pointing out that doesn't make sense.

I asked for reasons for why this tracking is bad, and I got what basically amounts to "all tracking ever is bad", and thus I'm explaining why that "all tracking is black" approach doesn't work.

I think the argument here is that tracking how users interact with an app isn’t necessarily private information

Ah so first you use excuse to cut testing and do proper development by installing malware?

And then users "have to" provide you with information to "identify user frustration".

Why not to create apps that do not cause user frustration in the first place?

Because humans aren’t omniscient with perfect reasoning ability, and full knowledge of all possible user circumstances and needs.

How would you find out what causes user frustration?

Company internal testing can only go so far, especially in a small team.

Ask them/make it easy for them to tell you?

What is malicious about monitoring use to provide a better product in service of users?

Did you clearly ask the users if you can do it? If not, there's probably a reason why not, because users wouldn't like it if they found out they were being recorded.

Imo, as long as you clearly say "here's the choice: opt in or opt out, we are doing XYZ" in clear words, it's fair.

It doesn't really matter whether on each tap you fire a specific event ("TapOnList", "TapOnPurchase") or whether you log all app interactions (where the user taps, are they going around the app not finding something?). Or does it? Where do you cross the line?

If this app would be Facebook I would see your grievances - devs could see personal data. But let's assume it's an app where there's barely any personal data, or specific views do not have this logging enabled.

Since we are discussing an app store:

"pregnancy tracker"

"democrat donation app"

"cancer survivors forum"

What's malicious about reading all your emails to provide you a better email service?

The monitoring is the malicious part.

Why? The act of monitoring is not inherently malicious. The word malicious has a specific meaning and definition: an intent to harm. A life guard monitoring a pool clearly isn’t acting with malice and yet is monitoring. A store CCTV camera monitoring shoppers isn’t designed to harm but rather prevent harm. So monitoring a webapp may or may not be malicious depending on the intent and effect of how the monitoring is used.