News for hackers. Stuff that matters.
Re:New study: Google manipulates users into constant tracking by stickfigure on Friday February 01, 2019 @11:50AM (#19055760)
This seems pretty reasonable to me, although perhaps they named the switch poorly. My guess is that the product manager is trying to provide a single simple "privacy fiend" setting.

You apparently want fine-grained privacy controls ("save my personal locations, but not my search history"). It's reasonable to want that, but it seems awfully entitled (not you, but many others on this thread) to demand it. Especially since we've seen from past history that a multitude of fine-grained privacy controls tends to propagate the narrative "Evil corporation confuses users with complex privacy UI".

Product management is way harder than you are giving it credit for.

Re:New study: Google manipulates users into constant tracking by levesque on Friday February 01, 2019 @03:01PM (#19057905)
I only described my personal experience, I'm not making assumptions of evilness on Google's side, just pointing out that it feels a little restrictive. As somebody else pointed out in this thread, the information is still stored on Google's servers (favorite places, friends' addresses, even my home address is still on there), it's just all of a sudden no longer available when you disable history.

Sure product management is hard, I never said it wasn't. Google maps is wayy ahead of the competition in the field, with or without those history things (cue the whole Apple maps fiasco). But users will always complain about features they don't like, and they won't always give credit for what works.

Honestly I don't care that much, I mostly adhere to the idea that Google & co. don't give a damn about any user's individual data. To them, we're just numbers in a database, and a bunch of metrics to optimize (engagement, revenue, etc.). However, I do think that governments, once they know all this data is stored on centralized servers, are tempted to exploit it, for good reasons or not, which is why at some point I want to limit the gathering of data, when I find it is not explicitly warranted or needed. Case in point, Google Maps doesn't need a history of ALL the places I've been on a day-to-day basis to index my saved locations in it's search box.

Re:New study: Google manipulates users into constant tracking by StavrosK on Friday February 01, 2019 @12:21PM (#19056125)
This is disingenuous and unreasonable. The setting is unnecessarily restrictive. It's simple:

"Do you want us to store your data on our servers?"

Answering "no" should mean you store it locally, not that you don't store it at all. If you tell Google not to save your search history on their servers, they just disable it completely, when they could just save it locally.

This is nowhere near a product problem. The setting is per-account, so it's not like it'd be disabled on one device and enabled on another, so they'd get some weird UX, they just like crippling their app to force you to give them data.

Re:New study: Google manipulates users into constant tracking by creato on Friday February 01, 2019 @01:04PM (#19056629)
> Answering "no" should mean you store it locally, not that you don't store it at all. If you tell Google not to save your search history on their servers, they just disable it completely, when they could just save it locally.

I think you are ridiculously underestimating the complexity and issues that would arise with two separate mechanisms for saving the same kind of data, and an option to switch between the two. Do you download account-saved locations locally when changing to local storage? What if there are conflicting pieces of data?

I bet there would be a lot more outraged people about the bugs and limitations that would arise with this system than there are today with your complaints.

Re:New study: Google manipulates users into constant tracking by StavrosK on Friday February 01, 2019 @01:30PM (#19056929)
> Do you download account-saved locations locally when changing to local storage?

No, you delete the data that's on the server. That's the reasonable way to handle a revocation of consent from the user.

Re:New study: Google manipulates users into constant tracking by UncleMeat on Friday February 01, 2019 @04:46PM (#19059000)
Headline: "Big company degrades user experience for users who choose to opt out". I'm absolutely certain that there are people who'd get mad that their history is gone if they went from "collect it in the back end" to "collect it in the client".
Re:New study: Google manipulates users into constant tracking by creato on Friday February 01, 2019 @05:17PM (#19059296)
Obviously. I'm asking what you do before that. Do you just throw it away? That's going to piss people off. Do you overwrite the local data? That's going to piss people off. Do you try to reconcile it? That's hard to do and messy, and probably will piss people off.
Re:New study: Google manipulates users into constant tracking by StavrosK on Friday February 01, 2019 @07:51PM (#19060207)
I'm not sure I understand. Overwrite the local data with what? If you have history on the server turned on, then there's only one set of data, you don't need to overwrite the local data with anything. You just keep it where it is.
Re:New study: Google manipulates users into constant tracking by creato on Friday February 01, 2019 @08:08PM (#19060276)
> If you have history on the server turned on, then there's only one set of data, you don't need to overwrite the local data with anything.

This is a bad assumption and why you would ruin a product like google maps if given the opportunity.

Google already had this problem for years with Contacts. I had local contacts on my phone, and contacts stored in my google account. It was a complete clusterfuck of duplicates, stale data, etc. Years later after I had this problem, Google added a fairly complicated power-user tool that let you reconcile this data. And then years after that, the tool became not-a-power-user tool and wasn't bad.

But we're still talking years of issues, for data that is probably a hell of a lot less complicated than what maps stores.

edit: And that said, I still have issues every now and then with that contact reconciler tool, it doesn't always do the right thing.

Re:New study: Google manipulates users into constant tracking by StavrosK on Friday February 01, 2019 @08:15PM (#19060320)
I have no idea what you're talking about. I'm talking about having server history turned on and then turning it off. What's the synchronization problem you see with that? I have no idea why you're talking about contact syncing, and if you mean that you need to sync locations like you sync contacts, then how is that a problem Google doesn't already have?
Re:New study: Google manipulates users into constant tracking by almuric on Friday February 01, 2019 @11:45PM (#19061142)
Most people have multiple devices. If it's all local, how do you keep it synced? I have a work PC, home Windows PC, Macbook, linux laptop and chromebook in addition to a phone. How do I sync across devices at all times? New contact is created on one device but how does it get to the other devices if there's no common server? New work location; new restaurant; new anything? Or are you assuming everyone only has one device.

I'm sure you've got it all figured out. You should go work at google and implement it.

Re:New study: Google manipulates users into constant tracking by StavrosK on Friday February 01, 2019 @11:53PM (#19061171)
I don't know what you aren't getting about this: Maps already has to implement sync. Disabling location storage on the server means that they don't have to implement sync. Each device has its own history. I can't explain it more simply.
Re:New study: Google manipulates users into constant tracking by chrshawkes on Friday February 01, 2019 @02:52PM (#19057810)
It's not that complex, especially for Google. They could use localstorage, cookies or whatever, no reason the users data has to go to Google to make personalized experiences possible.
Re:New study: Google manipulates users into constant tracking by rifung on Friday February 01, 2019 @01:22PM (#19056833)
> Answering "no" should mean you store it locally, not that you don't store it at all. If you tell Google not to save your search history on their servers, they just disable it completely, when they could just save it locally.

This is nowhere near a product problem. The setting is per-account, so it's not like it'd be disabled on one device and enabled on another, so they'd get some weird UX, they just like crippling their app to force you to give them data.

I work for Google, opinions are my own. I could go on and on about this so apologies for the long post.

Obviously I am biased, but I think you're being a little too cynical.

People's habits largely depend on the incentives given to them. There is no incentive for the Maps team to take in more data, if anything there are incentives to have less data because they'd have to go through less security and privacy reviews.

What they DO have an incentive to do is to get more users, since that largely determines how much funding your team gets. Having users share location data, admittedly at the expense of privacy, provides a superior experience.

Each team has limited funding/headcount, so obviously they would use their resources as much as possible to try and get more users. Supporting the use case you described, which is not used by the vast majority of users does not align with the incentives given to them.

Recently, Google updated its values (I can't find it externally but I hope you'll believe me). One of these values is "Respect the User". I personally think that your use case would fall under this umbrella. Ignoring whether the specific feature is the best use of engineering time even for this value, the reality is that values mean nothing without incentives.

And there aren't incentives. You don't get a great performance rating for following the values, you get one for impacting some large percentage of users or improving the performance or conversion rates.

Even if you had a team of monks, completely driven by their morals and not career-driven whatsoever, this would not be sufficient because if they didn't hit performance goals then the team would just get defunded, unless the person above them was also the same, and the person above them, and so on.

Re:New study: Google manipulates users into constant tracking by StavrosK on Friday February 01, 2019 @01:40PM (#19057064)
> Obviously I am biased, but I think you're being a little too cynical.

Very possibly.

> Supporting the use case you described, which is not used by the vast majority of users does not align with the incentives given to them.

I think you have causation wrong here. Users leave search history enabled because Google makes it unnecessarily hard to use the app with it disabled.

> Recently, Google updated its values (I can't find it externally but I hope you'll believe me). One of these values is "Respect the User".

I have no reason to disbelieve you.

> Ignoring whether the specific feature is the best use of engineering time even for this value, the reality is that values mean nothing without incentives.

If the values mean nothing without incentives, they aren't values, are they?

> You don't get a great performance rating for following the values, you get one for impacting some large percentage of users or improving the performance or conversion rates.

So what you're saying is "there's no incentive for allowing the user to have a good experience using the app without sending us data", or did I get that wrong? That doesn't sound awesome.

> Having users share location data, admittedly at the expense of privacy, provides a superior experience.

Especially when you don't bother providing a similar experience without people's data.

Ultimately, this isn't a question of "which would make Maps better, if users shared their data or if they didn't?". That's not your choice to make, so it's a question of "the user doesn't want to share their data, do we bother to make the experience good for them or not?"

If your answer to the latter is "No", I don't care if your values say "The User is the Alpha and the Omega, thou shalt have no other priorities before the User", it's a red herring because you aren't following your values, you're following the money.

Re:New study: Google manipulates users into constant tracking by rifung on Friday February 01, 2019 @02:06PM (#19057350)
> I think you have causation wrong here. Users leave search history enabled because Google makes it unnecessarily hard to use the app with it disabled.

That's a fair point. Nevertheless, most people on the team were not there when this feature was first introduced and we can only work from where we are now. To undo the status quo is something few are in a position to do.

> So what you're saying is "there's no incentive for allowing the user to have a good experience using the app without sending us data", or did I get that wrong? That doesn't sound awesome.

Correct.

> If your answer to the latter is "No", I don't care if your values say "The User is the Alpha and the Omega, thou shalt have no other priorities before the User", it's a red herring because you aren't following your values, you're following the money.

I think we are in agreement.

I don't think anyone is following the Values because the incentive structure doesn't align with it. I would like to believe it takes time for these changes to manifest but frankly I'm skeptical it will ever happen.

I'm not trying to convince you things are great or even okay. I just would like to dispel the notion that everyone or every team at Google is scheming together to get as much of your data as possible. Google, like any company, is made of individuals who largely are there to make money and whose actions are driven by the incentive structure. As it stands, the incentives are largely around getting more users and there is very little incentive to create privacy focused solutions (customers don't really seem to care.. or know..?)

Re:New study: Google manipulates users into constant tracking by StavrosK on Friday February 01, 2019 @02:55PM (#19057846)
It does, indeed, sound like we are in agreement. I'm not saying everyone in Google is evil and wants the worst for the user, but that Google doesn't really care about the user and that shows.

I guess everyone in the company is complicit to that, with a share of responsibility proportional to the amount of power they have, as with any organization...

Re:New study: Google manipulates users into constant tracking by librish on Friday February 01, 2019 @06:10PM (#19059683)
I think it's more accurate to say Google doesn't care very much about privacy concerned users since they represent a very small portion of the user base.
Re:New study: Google manipulates users into constant tracking by HeWhoLurksLate on Friday February 01, 2019 @11:13PM (#19060991)
> One of these values is "Respect the User".

The language here bothers me to the core. User? That's the same term drug dealers use to describe their "customers", who are nearly all captive to them or others who peddle the same thing. Does it get the point across? Sure. However, there are better ways to word this sentence, such as "respect the people who use our services" or "respect your customers", though the last one ought to read as "handle product with care".

Re:New study: Google manipulates users into constant tracking by ocdtrekkie on Friday February 01, 2019 @07:22PM (#19060080)
I feel like you may have entirely dismissed the issue that location-based advertising is quite possibly, the most valuable form of advertising Google can conduct, and Google has massive incentives to collect location data. Statements like "there is no incentive for the Maps team to take in more data" feel like they completely ignore your employer's primary business model.
Re:New study: Google manipulates users into constant tracking by darren_ on Friday February 01, 2019 @08:00PM (#19060242)
So I in fact work on the maps team. There is absolutely no _personal_ incentive for me (or anyone on my team) to help Google collect more location data. I will not get a bonus for it, I will not get an increased perf rating for it, any benefit I would get from it (increased stock price, I guess?) is so completely abstracted away that it simply does not register.

There is an incentive to have increased usage of the features I'm in charge of, many of which are intrinsically tied to (or superior with) location access. I don't believe that's the same as incentivizing the collection of location data, because it's not our real goal - if Apple would provide it I would absolutely love to provide users with a 'one-shot' location feature (it's probably too clumsy but the idea is when the user hits e.g. the directions button, the OS pops up a permission box and if they agree then maps gets one and only one GPS fix and then the permission is back off).

Re:New study: Google manipulates users into constant tracking by ocdtrekkie on Friday February 01, 2019 @11:29PM (#19061071)
That's fair, I suppose. I just suspect resistance would appear from up above if anyone tried to make location access significantly less required.
Re:New study: Google manipulates users into constant tracking by dzader on Friday February 01, 2019 @12:44PM (#19056365)
It's simple: "Do you want us to store your data on our servers?"

Why don't you just build your own mapping software then if its all so easy?

Re:New study: Google manipulates users into constant tracking by StavrosK on Friday February 01, 2019 @01:33PM (#19056957)
Because then I wouldn't get inane comments from random commenters online and my life would be substantially worse.
[dead]
[dead]
Re:New study: Google manipulates users into constant tracking by reaperducer on Friday February 01, 2019 @02:37PM (#19057672)
Yet not only are you so certain that it's hard, you're willing assign all sorts of benevolent motives to Google.

Google would change the rate at which your terminal cursor flashes if it could find a way to monetize it.

Re:New study: Google manipulates users into constant tracking by stickfigure on Friday February 01, 2019 @11:27PM (#19061056)
I do not try to hide my lack of authority, which is why I transparently couch my words with "it seems" and "my guess".

I don't need to speak with authority here. If you want to stoke the flames of moral outrage, the burden is on you, not me.

Re:New study: Google manipulates users into constant tracking by StavrosK on Friday February 01, 2019 @11:44PM (#19061138)
Yeah, but all you said was "if you didn't work on X, you can't have an opinion about X", which is a completely useless and trivially wrong statement.
Re:New study: Google manipulates users into constant tracking by stickfigure on Saturday February 02, 2019 @08:54PM (#19067281)
I would phrase it more like "if you didn't work on X, you can't have a strong opinion about X".