Comment by feelin_googley | Hacker News Reader

Comment by feelin_googley | Hacker News Reader

"As it had done consistently in its history, the firm, when faced with financial pressure, moved ever further in the direction of monetizing users` personal data. In this case, it finally went after the like button.

A little over a year after the IPO, on June 12th, 2014, Facebook quietly announced a change to its terms of service. "Starting soon in the U.S., we will also include information from some of the websites and apps you use," the company wrote. "This is a type of interest-based advertising, and many companies already do this."

...

In perhaps the creepiest example, Facebook applied for (and received, last year) a patent for a tool called Techniques, for emotion detection and content delivery. It would use the camera in your phone to take pictures of you as you scroll through content. Facebook would then use facial analysis to measure how much you did or did not like the content in question, so as to determine what kind of stuff to send your way. Ideas like this are what make Facebook, at times, feel like a giant blood-engorged tick hanging off your frontal lobe.""

Source:

https://www.rollingstone.com/politics/features/taibbi-facebo...

Oh wow, earlier this year, a few friends and I built a proof-of-concept of emotion-targetted content delivery in news feeds for a hackathon [1], as a way of warning people that this was plausible. Pretty wild to find out that Facebook actually holds a patent on the very idea that we were toying around with.

[1]: https://github.com/nwhacks-loki/loki

I did a similar proof of concept a while ago [1]. It outputs an emoji for an image or video frame based on facial emotion.

[1]: https://gist.github.com/shanev/c70099b3305e71c6a1232a42029f7...

This kind of stuff is already out in the field.

I worked at a company that designed physical installations for retail marketing and it’s not only known, it’s relatively easy to implement yourself.

Iirc there was an advertiser caught doing this when their device crashed and displayed logs including estimated age and mood

This one was on Reddit a year or so ago: https://twitter.com/GambleLee/status/862307447276544000

This is why the argument that "you agreed to this" doesn't really make sense from a practical point of view. The only thing Facebook would need to make this work is probably the generic Camera permission, which I believe is already used to allow you to take photos with the Facebook app.

This is the sort of trick companies like Facebook can use to mislead people. They give them a feature that they actually use and need and isn't too nefarious. Then they request the permission to make that feature work, and then they add other features silently to further track you and analyze your behavior in ways most people wouldn't even imagine.

How could anyone possibly know that this is what Facebook intended to use that permission for? Shenanigans like these are what make me think even the GDPR may not provide enough privacy protection.

GDPR allows you to remove the data if you discover they cross the line. Optimistically that encourages them to be good citizens about it so they can keep using that data.

The GDPR doesn't allow you to ask for generic "Camera permissions". You have to specify for what you'll use the data. If later you want to use it for something else, you need to ask for further permissions.

IANAGE (I am not a GDPR expert)

EDIT: I don't think you need to ask if it's obvious (e.g. asking for generic Camera permission may be fine if the user is using the "take picture for profile" functionality). But that only gives you the right to use the PII for that specific functionality.

To my understanding, IANAGE either, you are right. The purpose of data capture and processing must be explicitly acknowledged for the user. If you intend to use that data for another purpose, you need to ask consent. Plus, the user may revoke those permissions anytime. GDPR is going to be hard stuff.

There is an idea that I saw (I believe it was an HN comment some years ago) that would massively help privacy in the smartphone age. For every set of permissions an app requests, you can choose to allow or deny these permissions (as currently possible) or there would be a third option: provide a false set of data that imitates the data normally requested. e.g., instead of providing camera permission, the false data would return a black image with your camera resolution, and the app would function normally except that pictures would not appear.

I believe that this would be an insanely useful feature. I know that I would use it quite a lot.

It kind of reminds me of the secret codes that they show in spy movies- there is an authentication word that spies normally use when communicating with a contact, and a second alternative authentication word they can use when they are under duress (third party coercion) and need the contact to know this but interact under the guise that authentication proceeded as normal.

GDPR requires you to says simply and clearly, not only what data you are capturing, but also the purposes for which you are using it.

If they were looking at my face through my camera while I read their ToS, they wouldn't need a patented technology to know how I felt about their technology.

Well, they would probably think I am happy as I was laughing in tears and some mumbling was heard in a form of "people.. are... really using this... they are really using it..." ;)

Presumably, yours and GP's would be their entire dataset as far as reading the ToS goes. ;)

Who is sociopathic enough to follow through on an idea like this?

Shareholder value

If it makes that person money, then it's going to happen. Moral considerations don't factor into capitalism. Only raw maximisation of individual profit.

It's not a bug, it's a feature.

> Who is sociopathic enough to follow through on an idea like this?

On one hand, you are bound to find at least a few individuals that were born lacking the sufficient amount of human morality and/or empathy, among the 7 billion humans on this planet.

However, it's much easier to find examples if you look at organisms that by their very nature are incapable of relating to humans in that way. Such as cats or spiders or lizards. However they usually aren't in a position to follow through on the idea. But this lack of power also holds for most of the humans in the previous paragraph (even considering sociopaths are said to hold many positions of power).

So that basically leaves corporations as the answer to your question.

People need to live - not many are in a position to leave their job if they don’t like what they’re doing. Whether they morally agree with it is irrelevant - they have debts, a lifestyle, a family, and so on and so forth. This goes from the programmer up to the CEO level, and the shareholders (largely regular people through an index fund) aren’t close enough to the company/don’t feel enough ownership over it to demand it does something else. The other actors are those playing on the financial markets more directly, and again, they have a job of making as much money as possible, consequences be damned, or they lose their job/home/ability to support themselves and their family/etc.

This is capitalism at its finest - putting people into positions where they’re unable to express their moral preferences. And you can’t change this as an individual or minority group, as everybody grows up in a system that heavily encourages (and in many cases forces) that people put themselves into that position - as a minority actor, you can’t prevent the system from working.

Nobody would build these systems for free.

> People need to live - not many are in a position to leave their job if they don’t like what they’re doing.

It seems quite likely most of the developers at FB that work with these kind of features are in a position to leave. I mean, a face recognition expert at Facebook isn't really something that just falls out of your pocket.

They would however likely change to something similar though.

Living like a FB Dev is not the same as merely living (well or otherwise). You sound like you are trying to rationalize (not just explaining, but justifying) their behavior.

I would agree those in poverty are not in a position to leave their job (and even then, I'm willing to consider options). Devs in Silicon Valley do have a choice though. You are correct to identify capitalism as a fundamental problem, but you've conveniently minimized their privilege too far in this case. They don't get to say "capitalism made me do it." They really did and still do have an opportunity to be moral, costly as it may be to their standard of living. They are part of the problem because this just is their expression of their moral preferences.

They might not be able to prevent evil, but that doesn't mean they should actively participate in it. Just because you can't stop the Nazis doesn't give you license to be one.

I hold programmers responsible for what they program, especially anyone living above the poverty line. There is no justification for their behavior. It's plain evil.

------------------------------

https://philosopher.life/

There are (lots) of people who work in gun factories. This is nothing in comparison.

Not everyone working in a gun factory has that much liberty to pick and chose a job, maybe it's the only big business in town, but software developers aren't exactly starved for opportunities.

You're telling me someone who can get a job at Facebook is at risk of poverty and cannot find at least an employer who is marginally less creepy?

A better analogy might be the engineers who help design the guns, and the various other well-compensated, high-skill, and socioeconomically mobile individuals who facilitate their design and production.

No wonder Zuckerberg famously commented on how he puts tape over his cameras. :|

https://www.theguardian.com/technology/2016/jun/22/mark-zuck...

Reply on news.ycombinator.com