Live data from Hacker News

An American Privacy Emergency

scottaaronson.blog

141–150 of 154 posts

Re: An American Privacy Emergency

#141

Earlier quoted context omitted.

> Federally mandated parental leave (paternity and maternity leave) polls at about 80% in favor with the US adult population. This is regardless of political affiliation, by the way. Democrat and Republican voters both support it. This sounds like the sort of thing where if you poll people yes/no, a majority of people support it, but then if you ask them whether they'll change their vote over it, the people who said…

Washington’s paid parental leave act has us put money into our Unemployment Insurance program in addition to the UI payments we already make. So the business owners don’t have to pay for it, they just have to allow it to happen.

What percent of the original 80% who supported it still support it when you tell them they're the ones who will pay for it though? (They would have been either way, but that sort of thing tends to affect polling numbers.)

Re: An American Privacy Emergency

#142

Earlier quoted context omitted.

That's not what I said or implied it is. I said there has been criticism of using differential privacy and linked to it. I also mentioned the reconstruction attacks on coarsening. Those two things can be true at the same time.

There are glaring factual problems with the florida lawsuit and recent critiques of the 2020 Census DP usage. (See 1.) Whether they're not aware or intentionally malicious, critics who claim the constitutional issue say that DP was applied to state-level counts. But disclosure avoidance (such as DP noise) was never applied to state-level population counts used for congressional apportionment---which is the constituti…

I wasn't referring to the Florida lawsuit and don't know anything about it. I was referring to the Alabama lawsuit which was dismissed without prejudice[1] because it was premature. The relevant constitutional issue is the redistricting balance since it has to follow the "one person, one vote" principle.

[1]: https://journals.library.columbia.edu/index.php/stlr/blog/vi...

Re: An American Privacy Emergency

#143

Earlier quoted context omitted.

You’re welcome to give all your money away first, then worry about taking others.

You a temporarily embarrassed billionaire? Paying their fair share in taxes (corporate and otherwise) is far from "taking all their money away".

The top 1% alone are 40% of the federal income tax receipts.

40% of the population pays 0% in income taxes, often receiving credits and refunds.

I’m all for everyone contributing: one side looks like it’s contributing 40% already. How much more do you think should be taken from them to give to the people that currently contribute 0?

Re: An American Privacy Emergency

#144

Earlier quoted context omitted.

You a temporarily embarrassed billionaire? Paying their fair share in taxes (corporate and otherwise) is far from "taking all their money away".

The top 1% alone are 40% of the federal income tax receipts. 40% of the population pays 0% in income taxes, often receiving credits and refunds. I’m all for everyone contributing: one side looks like it’s contributing 40% already. How much more do you think should be taken from them to give to the people that currently contribute 0?

[dead]

Re: An American Privacy Emergency

#145
post #97

Earlier quoted context omitted.

PII isn't a GDPR term

Personal data is though, and the two terms have relatively similar meanings.

US data privacy terms are US data privacy terms, not GDPR terms.

It's sloppy use of language by someone who works in the sector.

And businesses wanting to be compliant with GDPR should not be misled with US PII terminology

Re: An American Privacy Emergency

#146
post #66
post #31

What is the political goal behind this directive? I assume there is some completely non-subtle purpose, but I can't tell what it is.

The people currently in power want every single byte of data they can get on people who reside in this country so they can divide them up by arbitrary criteria and do with those groups as they please. Depending on the person, that might mean imprisonment, it might mean deportation, it might mean something worse.

Doesn't this mean such data analysis can be used by anyone against anyone else? This does not just mean one political group can do this analysis, any political group or foreign power can make use of the data if it is not protected using differential privacy. Has the military weighed in on this? Seems like the data could be used to attack police forces, cities, states, demographic groups, groups of almost any kind. It could be used to identify military forces, say those at secret bases. Turning off the privacy protections seems like tearing down a fence; makes it easier to take your horse out for a ride, but also makes it easier for the wolves to get in.

Re: An American Privacy Emergency

#148

Earlier quoted context omitted.

The 2020 Census had a number of issues and many of them ended up giving the Democrats more seats in the House than they should have. Oh, and the Census results were due at the White House in December 2016, but somehow they didn't reach the WH until January 21, 2017. Was Differential Privacy specifically involved? I don't think so, but I think it's being lumped into the mix and being blamed for the other issues.

No disclosure avoidance was applied to the state-level population counts used for congressional apportionment (i.e. how house seats are divvied up). There are plenty of valid critiques of the 2020 Census disclosure avoidance systems, but the Rs who claim that DP was used to manipulate state population counts to give Democrats more seats either have no idea what they're talking about or are being malicious. Also... th…

The Rs are just confused. The manipulation happened but they didn't need DP to do it.

As I said, it wasn't the disclosure avoidance that generated the most discrepancies. It was other forms of adjustments.

Re: An American Privacy Emergency

#149

Earlier quoted context omitted.

The 2020 Census had a number of issues and many of them ended up giving the Democrats more seats in the House than they should have. Oh, and the Census results were due at the White House in December 2016, but somehow they didn't reach the WH until January 21, 2017. Was Differential Privacy specifically involved? I don't think so, but I think it's being lumped into the mix and being blamed for the other issues.

> The 2020 Census had a number of issues and many of them ended up giving the Democrats more seats in the House than they should have. Do you have any source on that. For a census issue to have an actual impact in an election outcome, it must either be: - a lucky issue (flipping a very close race from one side to the other) - a very big issue Both can happen occasionally, but I have a hard time imagining “many of the…

I believe a few states would have fewer seats in Congress if it weren't for some last minute adjustments. It wasn't the actual election. It was the allocation of seats to each state that had the biggest effect.

Re: An American Privacy Emergency

#150

Earlier quoted context omitted.

> so they can divide them up by arbitrary criteria and do with those groups as they please You mean like by race, gender, and sexual orientation, and using statistical analysis to portray a group of oppressors vs. oppressed, perpetrators and victims? Exactly what was happening via official public policy for the last 4 years while the progressives cheered it on?

You won't convince anyone with "angry Facebook grandpa" style comments. Please explain to us in great detail how exactly "official public policy" did that, including the laws and changes that were pushed to do so. And explain why, if that were the case, conservatives are doing nothing to stop that. Why would anyone continue to do something bad, just because the people you hate started it?

All I'm asking is that we be consistent in our outrage.

If you don't like what's happening now, I understand that point of view. But to pretend it didn't happen already is mind boggling.

If for some reason you don't think 2020-2024 were a social morass of slicing up people by their "identity", and in many cases were in fact made official public policy by way of signed executive orders (you can google them, I'm not your personal researcher), then I don't know what planet you were living on during that time.

Post reply on HN