Hacker News new | past | comments | ask | show | jobs | submit login
California governor proposes data dividend (cnbc.com)
40 points by audace 13 days ago | hide | past | web | favorite | 45 comments





Honestly this is just political blame shifting, virtue signaling, and more empty promises of handouts in the face of California's disastrous finances. California takes more taxes than practically any other state to fund failed infrastructure projects, all the while allowing sky high rents in the SF, LA, and SD areas because the government is locked down by rich landlords protecting their investments. Not to mention the price hikes on toll roads, county sales taxes, and general cost of living while homelessness is still rampant. Our greatest achievement is still an underwater tunnel built in the 70s. I want to see Gavin do something to change California first.

Users are already getting value for their data - the product showing the ads. People have a choice in whether they want to use gmail and get a free best-in-class email client, backend storage, and worldwide access to their email. The trade is that they're participating in an ad platform that is required to build this infrastructure that is on a never before seen in the history of humanity scale. Of course Americans and Europeans think it's totally fine to charge a five to ten bucks a month for a service like that, even when it would mean forcing the growing lower classes, second and third world countries, and the tech illiterate to use unsafe and frankly dangerous alternatives. Ads are a communal investment that allows us to provide services to people who can't afford those services.


For the sake of discussion, lets focus on the idea itself, whatever the motiviations of the politician who's saying it.

>> Users are already getting value for their data - the product showing the ads. People have a choice in whether they want to use gmail and get a free best-in-class email client, backend storage, and worldwide access to their email. The trade is that they're participating in an ad platform that

FB are currently earning $50b pa from advertising, which is largely premised on FB's data collection on and off the platform. What users get is, by-and-large, very similar to what they were getting in 2012, when FB was making $5bn. That 10X increase in revenue (and 10X increase in FB staff/cost) hasn't gone towards making more/better product. It has gone towards making better (often more creepy) ad-tech.

At least with television there is competition, and that means ad revenue has to go towards making more/better programming.

SAAS has negligible marginal cost and often strong network effects. The value of an ad-platform and many uses of data (especially ad-targetting) also scales exponentially. This is leading to bad, monopolistic outcomes.


That's an incredible selective way to look at it. Google search has almost no competition. Social media has Snapchat, Twitter, TikTok, and there's really not a huge bar to entry. New social media apps rise all the time. It's also your opinion that their product hasn't improved over time, and your conjecture that that's where the money is going.

Also in television there is much less competition. The TelCos are the worst monopolies out there. Getting a competitive cable or ISP landscape is a laughable idea in the US.


> really not a huge bar to entry

Critical mass is a huge bar to entry.


Fully agreed. This is no more in ideas space vs a rogue single party communist state engaging in policy wanking and pandering to the worst elements in society while pretending to be helping the poor. This is a state where politicians work actively to protect the interests of richest rich while screwing up actually productive people.

I hope Google, Facebook and others will develop a diversification strategy and move out of this shithole.


I think morally it's the wrong direction, much like paying a dividend to victims of a polluting chemical factory to offset their deleterious health effects.

That said, putting a price on data may be what it takes to make companies take privacy and security seriously, simply because it might make it easier to argue standing in a lawsuit where data is leaked or mishandled. Similarly, putting a price on life seems insensitive, but wrongful death lawsuits motivate safety concerns.


> much like paying a dividend to victims of a polluting chemical factory to offset their deleterious health effects

Out of curiosity, why would this be morally wrong? To avoid reasoning-by-connotation, replace the word dividend with a word that connotes something positive (eg reparations): would you say that the functionally-identical ongoing payment of reparations to victims of pollution would be wrong?


Not OP, but this is how I see it. It's a bit like the daycare late fines experiment:

https://rady.ucsd.edu/faculty/directory/gneezy/pub/docs/fine...

Here's a digestible Freakonomics bit which discusses the same:

http://freakonomics.com/2013/10/23/what-makes-people-do-what...

When there was not a price on being late, the price paid was actually moral. That is, being late was a bad thing to do and so you should feel bad. However, when a monetary price was introduced, it was no longer morally bad to be late, as long as you paid the fee. The monetary-price replaced the moral-price, and it ended up that there were a lot of people willing to pay money where they weren't willing to pay morally. And so the truancy rate rose with introduction of late fees.

I would guess that that is what OP meant. If you introduce a price, you can actually remove even stronger restrictions that are in place based on morality. Because as long as you're being charged, you can assume the price is inclusive of your moral hazard.


> I would guess that that is what OP meant. If you introduce a price, you can actually remove even stronger restrictions that are in place based on morality. Because as long as you're being charged, you can assume the price is inclusive of your moral hazard.

This, but also that a dividend gives a mandate to continue.

I'd say that dividends and fines are distinct, in that a fine typically comes with the understanding that wilfully continuing what you got fined for will result in an even larger fine. Conversely, accepting money in return for some wrong committed against you would typically make it harder to get an injunction against that activity.


I'm not sure it makes sense to model companies as moral agents in the same way you would individuals, especially when you're talking about unintuitive moral reasoning that leans heavily on individual feeling, like the Haifa daycare study. Hell, I thought it was a pretty common belief around here (and more broadly) to model corporations as amoral utility-maximizers, in which case a fine is the central way of aligning incentives (for non-criminal actions).

Even when modelling corporations as amoral entities, they still exist within a morally aware system. They are to some extent beholden to their customers, employees, and courts of law, which all make decisions of morality about their actions.

A dividend (or similar payment) generally improves the legitimacy of an action, whereas being fined for it does not.


I would say that the reparations are not morally reprehensible, as they are paying for damage already done. Unless of course, a company makes a decision that harms people, animals accounting for future reparations.

In the case of the dividend to pollute, you force people to choose between money/power/economy and them/their children having cancers and birth defects.

This is perhaps where the analogy breaks; ignoring DNA and credit, data may lack cross generational effects.


> Unless of course, a company makes a decision that harms people, animals accounting for future reparations.

To clarify, I was suggesting this case; that a company deliberately starts or continues a business which physiologically harms people, whereby these people are compensated as an ongoing cost of doing business, as opposed to a fine for desisted wrongdoing.

That is, I think, a truer analogy to what is being proposed with this data dividend. Granted, it breaks down at congenital disorders.


Good idea, bad implementation. This will legitimize and validate the data collection and nothing else will change. Will there be even less incentive for companies to secure the data they're collecting because we'd already have been paid for it? What should be targeted is how this data is: collected, secured, and used. Throwing a few dollars our way each year solves nothing.

> This will legitimize and validate the data collection

Such data collection practices are core to the business models of some of the most valuable companies on the planet. "Google does it" provides far more social proof for diluting one's users' privacy than "a California digital privacy law acknowledges it exists."


I agree with you. Both things can be true.

This isn't a new idea and though it is interesting and has good intentions, it doesn't solve the problem. As others have pointed out, this will encourage more data collection and less privacy. The exact opposite of what we want. Also, the users are already paid with free service.

The problem is that people value the free service more than their data. And companies value the data more than they value the service.

The tech and hacker community might value our data and privacy more than the service, but that's clearly not the case with most people.

Imagine if there was a company FaceTome that provides the exact same service FaceBook does. But FaceTome charged you a dollar for month and collect no data while FaceBook charges you nothing but collects your data. I guarantee you that most people would use FaceBook and FaceTome will go out of business as most people simply don't care about their data or privacy.


That absolutely was true prior to 2016, but now the post Brexit, post populist world has shaken a lot of people up. Before people didn't realize what giving up their data meant, and they still don't, but at least now they are hypersensitive to it. Hyper-targeting used to mean you didn't see ads for things you didn't care about, but now it seems to be a weapon for reaching vulnerable people.

Ads targeting isn't going away, but I'm not so sure a pay to use service wouldn't stand a chance in this current climate.


Good point. It's always possible that a paradigm shift can change the culture and open up the landscape to paid services. But from a purely anecdotal perspective I'm not seeing the paradigm shift. Instead, I'm seeing the opposite. People are made aware of the privacy violation and data collection and almost welcome it. They don't care if they are constantly monitored as long as they can find the nearest starbucks on their smartphones. Just my experience. but I'd welcome paid services and simply more competition. The centralization and domination of tech spaces by a handful of companies is something we all should be worried about.

Paying us for our data won't nearly compensate us for the loss of our democracy. Quite the opposite: it puts an exact price on how cheap that comes about.

People who are using hyperbolic untruths like "Facebook is stealing our democracy" or "your data is being sold" should be warned: non-analytical politicians may introduce regulation framed by these falsehoods. For example, a law regulating that Facebook not tamper in elections or sell user data would be a home run for Facebook, because they have done neither.

Speaking the truth in this instance is both more productive and also should frankly be more scary than the hyperbolic distilled versions that are oft thrown around: mining user data with AI systems to sell products to measurably alter human behavior is an unregulated industry and likely should be.

This "dividend" nonsense is transparently ridiculous since it misses the point of the problem -- that incentives here yield damaging societal effects -- but again, here we see the result of the "Facebook sells your data" meme/falsehood: politicians conclude if they just pay users for the data, the problem is solved, right?


Mining peoples' data to shape political discourse is not a falsehood.

Nonetheless, you are elsewise correct: mining user data to sell products to measurably alter human behavior is an unregulated industry and likely should be.


I agree, and paying people directly doesn't make a whole lot of economic sense.

But if it meaninfully increases the cost of using the targeted data, it could be an interesting trade-off.

Almost like a carbon tax isn't designed to raise revenue, a personal data tax could discourage the "use it by default" mentality, since it would come with a cost.


I agree with a lot of posts below. However, my absolutely biggest worry is that it will allow these companies to get even more personally identifying information about the user than they probably have today since they will have to compensate them monetarily.

For example, they will/may start asking for address, phone number, real DoB and SSN for tax reporting purposes since this would be an income. If I have multiple gmail/FB/... accounts I may have to drop some or provide the same information for those, basically removing any doubt that I'm the owner of those and so and so forth.

Moreover, since this won't be limited to Google & FB, almost any service which collects data, which would be everyone, would potentially have to ask you for this information before they let you create an account. It may lead to the complete loss of anonymity on the Internet.


> For example, they will/may start asking for address, phone number, real DoB and SSN for tax reporting purposes since this would be an income.

As I see it Hertzberg's proposal that Newsom announced would be the state collecting on behalf of the recipient. I would hope this could be done without providing identifying information.

The alternative proposal that companies pay users directly for their data is the one we should be wary of, as you're correct - companies would need your info to pay you. It was put in the article as a byline but didn't have any supporting commentary.


Four years ago I wrote about requiring companies to pass along x% of targeted ad revenue sales to the users: https://thelocalyarn.com/article/judicious-change

I still think this is a good strategy for the purpose of snuffing out the incentives for surveillance business models.


We do get x% of targeted ad rev. just isn't passed as USD, and instead paid by giving a free product.

Exactly. Your data is a currency. You spend it at FaceBook. You spend it at Google.

What I think should happen is, there should be some protections regarding our data. I should be able to say, ok, FaceBook, I wish to no longer use your services so you need to stop billing me (collecting my data).


Yep, people here don't seem o understand the concept of barter...

Oh, and wait until you have to declare your "data dividend" in your tax declaration...


Right, and I would like to force it to be passed as USD whatever else happens.

Before a company will pay you for your data, you will need to prove beyond doubt your identity, by providing that company with all the id and data you have. “Sorry, you havn’t provided enough data to validate your ID yet”, is the get out clause in that business model. The journey to hell, is paved with good intentions, and there are preists all along the way telling you which way it is.

Or, it will allow them to validate what data they have on you and charge more for it!

If they're going to do that, I'd leave the disbursement to the state.

That'd be easier for companies (just send one check), and provide a privacy preserving proxy between the payment and payee.


That's an incredibly good point. Good enough actually to inspire a modest little conspiracy theory in my head that it's actually a disguised push from the data giants.

Another serious counterargument: when a company is making billions from billions of users, how much could one reasonably expect? Revenue per unique user is tiny.


This kind of echoes what Jaron Lanier has been proposing. I think they get at something true, in terms of moral sentiment, but I don't understand how it would be implmented.

How do you set prices? How does this lead to better incentives and outcomes, rather than a scheme where people can get $4 a year and the industry gets a moral blank cheque.

Currently, our Google & FBs data is being monetized mostly via ad-targetting, or via (for example) optimizing the FB newsfeed algorithim so that you spend more time on FB, where they can put that ad-targetting data to work.

Google's captcha-powered self driving cars demonstrates a point, but at least for now, the value of data is mostly ad-related (or business/political/military/police intelligence, which overlaps with ad-targetting in worrying ways).

Paying consumers to be advertized to ... isn't that a black mirror premise?

As of now, I'm more inclined to the "make-the-data-public" direction. For democracy concerns, for example, it would be valuable if the public could


So, has the time for project Xanadu (https://en.wikipedia.org/wiki/Project_Xanadu#Original_17_rul...) come?

If so, do we have the technology?


Wait until people find out that the cost to implement, monitor, support and regulate something like this will be higher than the dividends they'll be getting...

bing pays you rewards, which for me adds up to like $20/yr. I'd hope there would be more competition between search engines so they compete with benefits for users. For most queries google isn't that much better, use DuckDuckGo for privacy or other engines.

People who see me work invariably ask: Why do you use bing?

And I invariably answer, in my best Idiocracy: 'Cause they pay me every time I do!


This would just normalize and turn surveillance capitalism into a casino. It would create the illusion that ordinary people win, except the house actually always wins, and in this case the house would always win by a lot.

Wow, that guy looks a lot like Mitt Romney.

> “California’s consumers should also be able to share in the wealth that is created from their data,” Newsom said from the State Capitol in Sacramento.

This is such an odd sentence to read, post-GDPR, because it sounds as if consumers have no idea what their data is being used for, and/or have no control over how it's processed, and just have to subject themselves to it.

The GDPR literally starts off by recognizing personal data and the protection thereof as a fundamental right, and regulates how others must subject themselves in order to process it.


> post-GDPR

As pervasive as the GDPR is, a Californian in California isn't subject to it, and outside of people interested in tech, would make an assumption they probably know little about it?


I think you misunderstood my point, which was that we're clearly seeing a massive global trend towards privacy and rights thereto. The GDPR is a manifestation of this trend, and has further sensitized consumers.

Cambridge Analytics happened at a time when neither processors nor consumers were yet very sensitive to the topic of privacy. This has changed dramatically, and the GDPR is not the cause thereof, but a consequence thereof.

Of course the GDPR is not binding in California, but it should be at least thought-provoking in California.


Double entendre on processors.



Applications are open for YC Summer 2019

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: