Hacker News Reader

> I don't think Apple/Google/Amazone or any other company should need to find ways to "punish" a bad actor. Nor should they be allowed to do so as this is a abuse of power and a intervention into politics and moral aspects companies should not do. Nor feel the need to do so. (I mean who decides what constitutes as a "bad actor").

Apple didn’t take any kind of moral decision here. It just enforced its policy for distribution of apps, just like it did for Facebook and Google last year. The decision is not about whether Clearview’s app or service is a threat to society or privacy.

Exactly. A fair number of companies try to use enterprise certificates as a way around Apple's App Store restrictions.

The big question is why this company didn't just distribute the software through the App Store. Is it a violation of Apple's App Store policies?

> Apple didn’t take any kind of moral decision here.

They most certainly did. This is both public virtue signaling and internal employee appeasement. They have good lawyers, so they had the lawyers come up with another reason, but it's purely a moral stance.

I’d expect someone making claims like that to have receipts for them...

You can disagree with Apple's policies, or the principles and reasoning of others. And I'm sure some of those people are these mythical, awful virtue-signalling hypocrites that people seem to hate so much. But what makes this virtue signalling/hypocritical rather than a values-based decision?

Maybe you think Clearview is fine to offer the service, but there are people who disagree based on moral, ethical and practical arguments - and this action seems in line with Apple's stated privacy concerns and prior enforcement of their enterprise distribution agreement.

I mean, I wish that were true. But Apple does a lot of selective enforcement and the app store is full of editorial decisions. I agree with the editorial decision here, and disagree with the HKmap.live editorial decision.

Until the wider non tech population can comprehend the danger of things like clearview AI, people who do understand the danger now will have to use tools such as secondary boycots to force change.

I guess the choice everyone in SC can make into the future is too avoid and essentially blacklist the people who work for, fund or facilitate the success of companies like clearview AI.

People are always talking about culture fit here. Moral fit?

But the government is the main customer of Clearview AI, so why would the government remove it’s own access to a tool?

Laws against facial recognition are futile (a bit like trying to ban tracking on web or encryption).

At most you can ban outcomes - persecution, discrimination, etc. But when every Guatemalan farmer can setup their own facial recognition startup - the genie is out of bottle.

> At most you can ban outcomes

You can also change incentives. If aggregating faces into a database makes you liable for the damage that happens when that database is used, the "Guatemalan farmer" (and any business that wants reasonably prices liability insurance) will go out of their way to avoid making that database.

No one is gonna go after couple of misdemeanours in a third world country.

Nobody is going to buy their Clearview or Palantir grade military/police tech from a third-world farmer, either.

> a third-world farmer

I think this is a fairly uncharitable view of software engineering in these countries. Many countries by surveillance technology from others.

The grand-parent post specifically mentioned a "guatemalan farmer", so my point was that people typically buy surveillance technology from bigger players instead of from small-time hobbyists and sole proprietors. There's nothing wrong with the software engineering in other countries.

> But when every Guatemalan farmer can setup their own facial recognition startup - the genie is out of bottle.

Not that familiar with Guatemala but I'm guessing they have internet there, a technically inclined Guatemalan farmer with money to burn and the marketing degree to sell snake oil could setup their own facial recognition startup 10 years ago.

The tech side of these things is neither impressive nor all too novel, the application side on the other hand has pretty significant impact on the society we want to build and live in. Just like GDPR for pushing back against invasive tracking I really don't see why a legislative approach is seen as ineffective in these discussions. Is that because of arguments like "somebody else will do it anyway" / "bad actors still exist"? That seems like relatively bad points to just do nothing. Everybody could rob a bank if they felt like it, that doesn't mean we shouldn't legislate against it.

> At most you can ban outcomes - persecution, discrimination, etc

If I remember correctly, a similar app was used in Russia by stalkers to find lookalikes to their fancy (public persons, ex-lovers). This app empowers all sorts of creepy behaviour.

I think public access to the images of a person should be vetted by the person in question on a case by case basis, but that would destroy their business model.

Yep, found reference: https://www.theguardian.com/technology/2016/may/17/findface-...

From the linked article:

> Kabakov says the app could revolutionise dating: “If you see someone you like, you can photograph them, find their identity, and then send them a friend request.” The interaction doesn’t always have to involve the rather creepy opening gambit of clandestine street photography, he added: “It also looks for similar people. So you could just upload a photo of a movie star you like, or your ex, and then find 10 girls who look similar to her and send them messages.”

Ok lets ban digitally approaching people because of their looks...

If you make it accessible to everyone then you make sure you've got solid audit trail tech. Definitely notify face owner they've been looked up and by whom. A bit like LinkedId profile visitors work already.

The government's definition of a bad actor is different to yours and mine.

You can lie, cheat, steal, spy, torture and kill, so long as you do it on behalf of, and for the benefit of, a government.

I don't think there are any governments that have that policy. For instance, there are complaints when someone gets polonium poisoning, in spite of people thinking it's because of a government. There are complaints about the occupation of Iraq, from governments. And so on.

There are complaints, but polonium poisoning is a tool of a particular government. The perpetrators who carried out the murder did not face justice, they were given shelter by said government, and very likely rewarded for a job well done.

Other governments use drones to bomb people in various countries with impunity.

If a private citizen does any of this sort of murder of their own accord, they go to jail. The difference between government and private citizens is that the former can exempt itself from the consequences of these heinous acts. That is what is meant by the term “monopoly on violence.” [1]

[1] https://en.wikipedia.org/wiki/Monopoly_on_violence

Legal self defense is a thing, at least in my jurisdiction. So I don't think the government has a monopoly on violence.

Also, your opinion on what is murder is arbitrary. All your criticism really is is, "I don't like the definition currently used". That doesn't contain any reason why it should be modified or how or describe any principle.

The fact that governments can circumscribe the killings they do as “not murder” and have regular people believe them is the quintessential expression of the state monopoly on violence.

A monopoly on something doesn’t mean exclusive, 100% godlike control of it. It just means the legitimate provisioning of that thing falls under the authority of one entity. Copyright is a government granted monopoly on the right to copy a creative work but that doesn’t make it impossible for others to make copies, it just makes it illegal.

"The fact that governments can circumscribe the killings they do as “not murder” and have regular people believe them is the quintessential expression of the state monopoly on violence."

A state, and words and concepts, only exist insofar as people believe in them. You're only saying "murder is [defined]", not anything about governments. If you pretend you are making a statement about government or politics, it suggests that somehow a different system could achieve a better and different state. Everyone defining legal killing for themselves is not better or worse, it's a contradiction in terms - an absurdity.

So you expect the same government that is eager to invade your privacy and rights to turn around and protect your privacy?

I trust government power a lot less than corporate power.

Some corporations also invade privacy, like some governments do. And some protect it.

Just playing devil's advocate here:

"Nor should they be allowed to do so as this is a abuse of power and a intervention into politics and moral aspects companies should not do."

Your justification for apple not removing them is that it brings them into the realm of making moral decisions which companies should not be making. Yet people call for Clearview AI to make the moral decision of not selling facial recognition tech. It's a two way street. Why should apple not be allowed to make this moral judgement, yet Clearview should have to? Shouldn't Clearview just stay out of moral decisions, and as a company just try to maximize profits?

Apple owns the marketplace here so a different set of considerations apply to them compared to individual app makers.

Like you I've also instinctively always been opposed to task private companies with civic duties but in particular seeing the direction both the US and China are heading in terms of privacy I'm begrudingly almost feeling better with companies taking on stakeholder roles.

I don't really have any illusions that most of the entrepreneurs aren't full of it when they talk about values but nonetheless it's incredibly sad to see how strongly the very largest governments fail in advocating for civil liberties.

The notion that Apple should not be “allowed” to keep some software out of their App Store is nonsense. It saddens me that a person who claims to care about freedom also wants the government to forbid people from making editorial decisions on their private platforms.

You might have a point if Apple did not control the only App store for iOS devices.

"only App store for iOS devices." which was created by Apple. Just like Playstation, Nintendo and Xbox.

Consoles are problematic for the same reason.

Yes and they have a near total monopoly with their [checks notes] 20% global handset marketshare.

You can't just go to Samsung, Huawei, Xiaomi, Oppo, Nokia, or anybody else and get one!!

Don't deny that lock-in effects are a thing. Not that it Apple's practices less grossly immoral.

Amazon sells reckognition which competes with clearview

you sound like a shill for the likes of clearview etc.

The state was never about protecting people but protecting its masters (politicians, bureucrats).

It's a dangerous precedent to force distribution companies (in this case, Apple) to be able to decide to deny service to a vendor.

We want Apple to be able just to get rid of a bad actor.

Can people inside Apple theoretically abuse this power? Yes. However, it's way worse if Apple is forced to allow bad actors in the platform.

I'd say we might talk about better notarization approaches for app (mobile and non-mobile alike). However, getting rid of any notarization is a step backward.