Noise infusion banned from statistical products published by Census Bureau • Hacker News Board

Re: Noise infusion banned from statistical products published by Census Bureau

#211

Post by ilyagr » Sat, Jun 13, 2026, 6:33 PM UTC

asolove wrote: post #13

The replies here arguing we should publish it all are wild in the worst kind of first-order thinking way. It’s a census: it just asks questions. If you start publishing and weaponizing the data against people with various attributes, they’ll just lie or not answer. And then you are left with worse than nothing: bad data people try to act on.

This might be the point. As long as they think the people who end up under-counted are not people this government would like to have voting power for the House of Representatives.

Re: Noise infusion banned from statistical products published by Census Bureau

#212

Post by esikich » Sat, Jun 13, 2026, 6:33 PM UTC

whattheheckheck wrote: post #70

conception wrote: post #63

Earlier quoted context omitted.

Remember “leftist “ and transgender activists are terrorists now. First they came for…

What's the actual antidote to this? 5calls.org?

Violence. But people are still pretty comfortable so here we sit.

Re: Noise infusion banned from statistical products published by Census Bureau

#213

Post by SmirkingRevenge » Sat, Jun 13, 2026, 6:34 PM UTC

p-e-w wrote: post #111

estearum wrote: post #102

Earlier quoted context omitted.

Voting and getting everyone you know to vote

Turnout in the past few elections was already extraordinarily high. Clearly this isn’t the answer. And neither is protesting, considering that 4 of the 5 largest demonstrations in US history happened in the past 10 years and achieved nothing.

The trend has been going that direction, where low turnout elections favor Ds and high turnout favors Rs. But only kind of... that holds when Trump is on the ballot. Trump seems to activate a segment of low-info, low-propensity voters who stay home when their guy isn't on the ballot. Things will probably get scrambled again once Trump is gone though.

And don't discount protests. It's crucially important to have big public and forceful displays of united opposition. The regime is unlikely to be toppled by protests, but they will weaken it.

That really matters.

In an authoritarian take-over institutions are the front-lines, not the masses. Think colleges, media, industry, courts, legal firms, local governments, etc. The dilemma those institutions will face is to follow rule-of-law or submit to authoritarian corruption. Authoritarians win when those institutions decide it's safer to submit than it is to follow the law. And when institutions (and the people within them) feel like they are twisting in the wind alone and nobody cares, they are more likely to buckle. Protest movements help reinforce the rule-of-law side of that calculation.

(The rise and fall of Orban is a great lesson on all of this)

Also see: https://essayx.substack.com/p/the-35-percent-rule-just-made-...

Re: Noise infusion banned from statistical products published by Census Bureau

#214

Post by brainwad » Sat, Jun 13, 2026, 6:35 PM UTC

throw-the-towel wrote: post #174

pelorat wrote: post #129

Earlier quoted context omitted.

Sure, in Europe we don't because we already have databases of all citizens, also recording attributes like race, skin color, religious affiliation or political leaning in a database is highly illegal, both for the government and for private use.

Wait, are you saying Europe doesn't have censuses?

In my European country (Switzerland), it's mandatory to notify the government whenever you move. There is thus no universal census and also no voter registration. There's still subsampled surveys though, for e.g. economic data, that might come by mail (addressed to you by name, because they know where you live!).

Re: Noise infusion banned from statistical products published by Census Bureau

#215

Post by nirava » Sat, Jun 13, 2026, 6:37 PM UTC

sherburt3 wrote: post #172

So "differential privacy" pretty much sounds like someone gets to modify the results of a census and how it gets modified is entirely up to their discretion. Seems like something that could be abused to achieve political objectives.

there are obviously measures in place to ensure the added noise is statistically homogeneous. the changes don't affect the final aggregates significantly, just enough to avoid saying much *about any individual person*.

know how you can buy "anonymized" data from data brokers and drill down until it's not anonymous anymore and in many cases point to the exact person? differential privacy would prevent that kind of thing.

If someone actually wanted to achieve political objectives by tampering with census data, there are better means than tampering with homogeneous statistical fuzzing.

Re: Noise infusion banned from statistical products published by Census Bureau

#216

Post by shiandow » Sat, Jun 13, 2026, 6:37 PM UTC

foolfoolz wrote: post #68

i have such a hard time reconciling stuff like this: > The census bureau decided to adopt differential privacy for the 2020 Census and: > The consequences will be dire for utility or for privacy, and possibly both. It's hard to understate this point: future statistical releases will either be useless compared to past ones, or they will be incredibly unsafe so we took the census for centuries before this point, and it…

As far as I recall they did have some measures in place. Differential privacy just made it a bit more robust.

Arguably the defaults for differential privacy are too robust but that is a different story.

Re: Noise infusion banned from statistical products published by Census Bureau

#217

Post by t-3 » Sat, Jun 13, 2026, 6:39 PM UTC

kgwxd wrote: post #167

deaux wrote: post #138

Earlier quoted context omitted.

> Turnout in the past few elections was already extraordinarily high. In a sense, this in itself is the issue. It's long-term _worse_ to vote for the "lesser of two serious evils". This extreme "long-term pain for short-term gain" attitude is what's gotten the US to where it is. If in 2016 of 2024 even 20% of the dems would've stayed home or voted third party, the DNC's continuous forcing of awful corpocrats with zer…

> Every time I've explained this I've gotten instantly downvoted Because it's dumb. People don't want to hear dumb ideas, or take the time to try and convince someone that would spend however long it took to type that, apparently multiple times, without realizing it. Throwing away votes will never be the reasonable thing to do. I know you don't want to hear that, because it's too painful for you to admit there's no s…

Please, explain to me how voting for a candidate you don't like is not throwing away your vote, but voting for a candidate you support is.

Re: Noise infusion banned from statistical products published by Census Bureau

#218

Post by Bratmon » Sat, Jun 13, 2026, 6:39 PM UTC

swiftcoder wrote: post #112

Bratmon wrote: post #92

Earlier quoted context omitted.

But this article is about a decision to damage the census less. If you value an accurate census, you should be celebrating!

TFA lays out why things don't work that way. If you erode trust in the privacy of census responses, an awful lot of folks will have to start lying on their census

I think the TFA was very light on evidence that people's desire to respond to the census is increased if the government fabricates the results later

Re: Noise infusion banned from statistical products published by Census Bureau

#219

Post by Bratmon » Sat, Jun 13, 2026, 6:42 PM UTC

This is a rare occasion of the Trump administration getting something right.

Why even do a census if you're just going to synthesize random data as the last step?

Re: Noise infusion banned from statistical products published by Census Bureau

#220

Post by SmirkingRevenge » Sat, Jun 13, 2026, 6:42 PM UTC

mullingitover wrote: post #176

deaux wrote: post #138

Earlier quoted context omitted.

> Turnout in the past few elections was already extraordinarily high. In a sense, this in itself is the issue. It's long-term _worse_ to vote for the "lesser of two serious evils". This extreme "long-term pain for short-term gain" attitude is what's gotten the US to where it is. If in 2016 of 2024 even 20% of the dems would've stayed home or voted third party, the DNC's continuous forcing of awful corpocrats with zer…

> Every time I've explained this I've gotten instantly downvoted without a single reply making an argument against it, Ok I'll break it down for you. > If in 2016 of 2024 even 20% of the dems would've stayed home or voted third party Parties cater to their bases, and putting yourself out there as an unreliable voting bloc is exactly how you get your demands ignored. > The whole idea of "third-party voting is a comple…

> Parties cater to their bases, and putting yourself out there as an unreliable voting bloc is exactly how you get your demands ignored.

I wish more people understood this. Instead, there's this mistaken notion that you give away leverage by supplying votes. It's literally the opposite.

Your coalition will have more influence and leverage within a party by supplying votes, not withholding them.