Comment by photochemsyn | Hacker News Reader

Comment by photochemsyn | Hacker News Reader

Parent Full thread

Note that this is just records of some of the federal agencies, not all of them, according to the linked ACLU report:

> "Although the litigation is ongoing, we are now making public the records that CBP, ICE, the U.S. Secret Service, the U.S. Coast Guard, and several offices within DHS Headquarters have provided us to date."

The NSA is still vacuuming up all the metadata and a good fraction of the content from the main nodes where it put those fiber-optic cable splitters on the main trunk lines what, 20 years ago or so? Under the Patriot Act provisions, pushed through Congress in late October 2001 wasn't it? Just a few days after the Senate got shut down by those anthrax letter attacks sent to Daschle and Leahy (no, it wasn't Bruce Ivins).

Then you've got the backdoors into Google and Apple, the whole PRISM thing... I doubt they've shut any of that down. See Yasha Levine's "Surveillance Valley" for more on that.

https://yashalevine.com/surveillance-valley

It's not quite China yet, but I'm pretty sure that when our politicians and bureaucrats and their corporate masters look at China's system, their main emotion is one of envy.

Funny thing is, they started the surveillance in late-2001 to suck all the records up. Where did they get the computers and storage devices to do that all at once? They must have been installing equipment for months or years, especially since I am pretty sure they even built a new data center on Fort Mead for it around that time. Now, this is 2001, when building a datacenter didn't just involve spinning up 1000 AWS EC2's and opening the spigot to S3, so this type of thing would have taken some time.

So the question is: did they pass a law to allow data collection because of 9/11 and other attacks, or did they pass a law because they wanted the NSA to be able to collect this data using computer systems they had been planning for years, and used those attacks as a pretext?

>Funny thing is, they started the surveillance in late-2001 to suck all the records up.

It started long, long before 2001. Here is a 2000 press release from the ACLU about global surveillance done by Echelon, which had already been ongoing for years (back when the ACLU cared about privacy issues):

https://www.aclu.org/press-releases/privacy-advocates-concer...

By 1999 privacy advocates at already gotten together for an official "jam Echelon day".

>It was the latest in a long line of apparently futile online protest movements. On Thursday October 21 1999, internet users were urged to drop trigger words into their email so as to overwhelm Echelon, the massive surveillance project which is said to keep track of all electronic communications into and out of the USA, the UK, Canada, Australia and New Zealand.

https://techmonitor.ai/technology/did_jam_echelon_day_do_mor...

Ah right, I remember when Echelon Watch[0] was launched, and made the rounds in chat rooms I frequented at the time... Of course most people dismissed it as conspiracy theory stuff :)

[0] https://web.archive.org/web/20001018103147/http://www.echelo...

I would suspect it was already being done (in part, or in whole who knows) and the attacks were an excuse to "legitimize" it going forward.

It was definitely being done before 9/11. I know of one data broker who was under contract with a certain TLA in the 90's to perform hardware assisted rapid data scanning.

Remember the Clipper chip from 1993? It's not like they suddenly started it in 2001.

Insider Bill Binney tried to do responsible data collection. He got harassed. https://m.youtube.com/watch?v=JvAflFcpJFQ

I legit don't believe they can store all that data. Youtube alone creates too much data for them to process and handle.

They must be storing either a subset or only partial metadata.

I'm on the other side of the fence - I believe they can store all they need to. Backdoors into Google, AWS, and others were strongly hinted at if not fully exposed with all the leaks back in the Snowden days, so US Agencies may not even need to store some of the bulky stuff. But given the vast size and number of datacenters operated by US agencies, it seems likely they have a LOT of storage capacity. http://worldstopdatacenters.com/government-data-centers/

The NSA has an unlimited budget to store all the data they want. They have massive, billion-dollar data centers around the country (and the world?). The one in Utah is perhaps the best known.

https://en.wikipedia.org/wiki/Utah_Data_Center

Nothing is unlimited. If it cut into our tank budget, or anything old retirees care about, something would happen.

>Nothing is unlimited. If it cut into our tank budget, or anything old retirees care about, something would happen.

Nothing is unlimited, especially inflation. The idea of a "budget" when the FED can (and does) create trillions of dollars out of thin air with a few keystrokes is antiquated. They can - and do - devote whatever resources they desire for "national security".

https://www.usdebtclock.org/

If I was them, I would not back up YouTube, but I might carefully scrape and discard.

Hell, if I was a _benevolent_ surveillance program, I'd probably run routine searches for illegal stuff on YouTube, both to find it myself, and to make sure YouTube's tripwires are working.

There is so much low-hanging fruit in terms of "interesting secrets per byte"

Like, I could believe all SMS messages are stored for a year or so.

Some random source says, "Over 6 billion texts are sent every day".

If a text is about 140 characters, and you use a dumb image classifier to transcribe photos as "Nude woman", "nude man", "dick pic", "image macro", "guns", etc., that's only about 1 TB per day, right?

365 TB to keep all US text messages for a year? Maybe my source is wrong. That sounds low. But, it's just text. Maybe it's right.

In fact, the upper bound for all US keyboard input for a year must be below 4.6 petabytes.

(350 million people typing 365 days a year, 16 hours a day, 40 words per minute, 1 bit of entropy per character after compression, 8 bits per word)

with large datasets like bioinformatics you often compress the data by extracting features and building matrixes, then only keeping things that score above a threshold

Sounds right. I like to say we’re all on a list and it just matters where you rank on it. And the precision is only good when you rank high on it.

No different than google search results. The accuracy of the 7th page of results just doesn’t matter.

I legit believe they store everything they can because they've been doing exactly this for longer than anyone else. They are _the_ experts in metadata management.

> While the actual capacity is classified for NATIONAL SECURITY REASONS, we can say this: The Utah Data Center was built with future expansion in mind and the ultimate capacity will definitely be "alottabytes"!

https://nsa.gov1.info/utah-data-center/

I mean more than a few x00 someones have to actually know the answer to that question. Shocked it hasn't been leaked.

likely netflow data.

I recall some natural disaster in Thailand being blamed for a hard drive shortage at the time, when it ended up being the NSA buying them all up for their Utah data center.

Edit: anyone downvoting want to comment on why this is implausible or not noteworthy to mention? It was a topic of discussion at the time.

Source? Would love to hear about this.

I searched for "thailand hard drive shortage nsa" and only got...

2013 Nov, Backblaze's annual hard drive report still only blamed the flooding: https://www.backblaze.com/blog/farming-hard-drives-2-years-a...

2015, Tech Power-Up has a story about NSA malware hidden in hard drive firmware: https://www.techpowerup.com/209925/nsa-hides-spying-backdoor...

I don't recall where I heard it but found this in a quick search:

https://www.reddit.com/r/conspiracy/comments/li1pep/a_global...

It's not quite China yet

What do you mean with this, specifically? Do you mean that China has a more comprehensive data collection apparatus, that the Chinese government has easier access to commercially collected data, or maybe that they exert their control more overtly than the US?

Yes.

[citation needed]

And yet, even with all of this surveillance, we still have mass shootings every five days.

My understanding is that in the vast majority of these mass shooting cases, authorities are well aware of the danger but there's no enforcement mechanism for "probably going to do a mass shooting".

Yeah there is.

It's called not paying to bombard them with great replacement conspiracy theories all for a decade and instead paying twice as many teachers as well as getting dedicated mental health staff.

>replacement conspiracy theories

https://europhobia.blog

Yes. exactly like that one.

Judging by what is done (and what is not done) to remedy them, they sound more like a feature than a bug of our society unfortunately. People are able to storm political institutions when they are angry about an election, but just buy more weapons when their kids get killed. Go figure.

Only "nobodies" get shot. Funny how they never target any politicians, huh?

> Funny how they never target any politicians, huh?

They do. Here are a couple recent examples:

https://en.m.wikipedia.org/wiki/2011_Tucson_shooting https://en.m.wikipedia.org/wiki/Congressional_baseball_shoot...

Resolving mass shootings isn't the purpose of the surveillance.

According to https://en.wikipedia.org/wiki/List_of_mass_shootings_in_the_... there have been 337 mass shootings in the US so far in 2022, way more than one per day. One every five days would at this point be a big improvement.

surveillance doesn't seem to do much to deter crimes, it just makes investigation easier

I could almost support the panopticon if that actually happened.

Adjacently, to make the point: we warehouse 1000s or rape kits but don't bother to process them. Like, wtf?

Big Data completely flips the problem of investigation. Instead of identifying suspects, chasing leads, gathering evidence, and verifying alibis, with Big Data you just rule out everyone who does have an alibi. Leaving only those who don't have an alibi, thereby revealing the perpetrators.

It's been a while since I've actively cared about protecting privacy. So I don't know if any one's talking about how Big Data impacts legal concepts (balances of concerns) hammered out over the centuries. Like what is reasonable doubt and presumption of innocence when prosecutors can prove that every one else has an alibi?

When I tried to talk about this stuff with other privacy advocates, presumably savvy about policy and whatnot, they'd look at me like I'm paranoid schizophrenic. So if the experts are still coming up to speed, it'll take that much longer for the policy maker, much less the general public, to become aware.

Comprehensive surveillance makes blackmailing and controlling politicians easier.

>It's not quite China yet, but I'm pretty sure that when our politicians and bureaucrats and their corporate masters look at China's system, their main emotion is one of envy.

The US has a much wider intellegence reach and surveillance, not only on our own citizens but on the worlds.

> It's not quite China yet,

I'd call it a decentrailized surveillance system --- very robust against any attempt to dismantle it.

There are no "backdoors into Google and Apple." The government has to request specific accounts' data with a court order.

The NSA isn't "vacuuming up all the metadata and a good fraction of the content from the main nodes." It does full take in some countries with national security interest and collects data to or from some endpoints outside the US known to belong to foreigners with a court order.

We know all of this from the documents Snowden leaked and the documents the government declassified.

That "court order" is from the FISA court, which may as well be a monkey with a rubber stamp. We learned that from the Snowden documents too.

No, we learned that they were approved at roughly the same rate as other search court orders. Investigators aren't going to spend the time making an application for a court order unless it has a very high chance of being granted.

https://www.quora.com/What-percent-of-law-enforcement-search...

They only call it a court for marketing reasons. It's basically just an internal review process.

Reply on news.ycombinator.com