Hacker News Reader

Comment by oblio | Hacker News Reader

> That’s because dossiers on billions of people hold the power to wreak almost unimaginable harm, and yet, each dossier brings in just a few dollars a year. For commercial surveillance to be cost effective, it has to socialize all the risks associated with mass surveillance and privatize all the gains.

> There’s an old-fashioned word for this: corruption. In corrupt systems, a few bad actors cost everyone else billions in order to bring in millions – the savings a factory can realize from dumping pollution in the water supply are much smaller than the costs we all bear from being poisoned by effluent. But the costs are widely diffused while the gains are tightly concentrated, so the beneficiaries of corruption can always outspend their victims to stay clear.

Spot on. Google's in exactly the same boat, it's just that they provide a few services which are legitimately useful (Search, Gmail), so they're targeted less right now.

How we will get out of this, I have no idea...

View on HN

The fix is not complicated. I have no issue with my search data/post viewing history providing me better results or better ads.

The real damaging data to people and society at large is a different set of data. It's the publicly visible counts next to every thought and utterance reinforcing misguided beliefs and behaviour up and down the food chain constantly. Any experienced shrink, psycologist or educator, marketing/PR expert knows applying the right amount of feedback at the right time is critical to how people process info.

Remove/delay/reduce the visibility of like counts/view counts/upvotes/retweet counts that are displayed and the world will be a different place overnight.

>I have no issue with my search data/post viewing history providing me better results or better ads. The real damaging data to people and society at large is a different set of data.

Hmm, I disagree. It seems to me that both sources of data are dangerous. Yes, showing "You have 10 likes" is bad for users as social media companies iterate their way towards addiction but your search data is a toxic asset beyond just showing ads. Let's say google leaked everyone's search history tomorrow. How many marriages are going to be ruined? How many politicians are going to resign? How many future politicians will decide to never run? How many firings and never-hirings will there be based on that history?

People should be able to live normal lives without being surveilled - by governments or corporations.

What if the opposite happened? What if we all suddenly had undeniable confirmation that shitloads of people watch mannequin porn, and so fuck it, why be embarrassed by it?

You have a very optimistic model of people.

Not necessarily. If literally everybody is open to the same level of scrutiny of their lives, how quick will they be to judge others, lest they in turn be judged?

I think its more likely that the judgement will spread in a highly non-uniform manner that reflected existing social biases with some people being vastly more negatively affected by having their secrets exposed than others.

Given equal access to the sources of shame of all three, would we really expect society to apply equal judgement standards to a mother, a billionaire asshole white guy, and a black guy from the inner city?

How is that kind of bias mitigated by privacy though?

My interpretation of the argument was that taking privacy away from those who had it wouldn't necessarily cause harm because in some aggregated way everyone would have something to lose and so people would be tempered against 'throwing the first stone'.

I think that notion is false because the negative effects of the loss of privacy will spread unequally and opportunity created for those to exploit the unequal consequences against their foes.

Very quick, it turns out. Judging others is a deflective tactic against being judged yourself. "People in glass houses shouldn't throw stones" is great spiritual wisdom but most people don't abide by it.

That's sorta my point. If no one had any privacy, then any time you judge others you open yourself to being judged. It's like punching someone in the face, you could do it but they'll probably punch you back.

So yeah, maybe you could make a big deal about congressman so-and-so having a diaper fetish or whatever, but then people will probably want to look into your porn browsing history.

Look at research into school bullies; it's a somewhat similar phenomenon to what you're envisioning.

---

Lisa: I can't believe I'm jealous of a baby!

Bart: Hey, so am I! When you're a baby, you can just spend all day rolling around on the floor. (sighs) I miss those days.

Nelson: Then roll, baby! (Bart rolls on the bus floor) Ha Ha! Floor Baby!

Lisa: You're laughing at him for something you made him do.

Nelson: Well… you're gay!

Lisa: People who call other people gay are often covering up their own latent homosexuality.

Wait I don't get the point of this story. Is it meant to show that by the story's end, Lisa has become the bully? That's how it reads to me.

They've just been punched in the face, they can't reasonably punch back. (Everyone has a plan until they get punched in the mouth).

I know it seems like universal surveillance could level the field and make people less likely to judge, but in practice it doesn't work that way. You say "people will probably want into [the judger's] porn browsing history", but in reality, they don't. The preemptive strike generally wins, and counterstrikes generally look like defensive posturing. So removing all privacy just gives more power to the bigger asshole.

Just had a conversation about Amazon facial recognition with a friend last evening.

Imagine ten years in the future: you did something to upset the local police officer. Maybe you didn't pick up a can like he ordered. In any case, imagine he takes a photo of you with his body camera and then using Amazon's machine learning, they're able to find likeness of you doing "illegal" things and immediately write you a ticket. Resist more? Maybe Amazon can dig deeper and find out the faces you are around often and dig out treasure trove about someone.

I agree. Without universal enforcement, universal surveillance is pointless. This is why I was glad to see TSA selecting old women on wheelchairs with oxygen masks for searching while boarding onto airplanes because what we had before that was not random searching at all. It was profiling.

I'm amazed nobody in New York talks about NYPD reflective vests in a car's dash. Clearly, the car owner is communicating that the car belongs to a police officer to avoid a ticket. Anyone who does this does not belong in our police force. However, people just don't care about it.

It doesn't seem to work that way. When Ashley Madison leaked did people stand up and go "whoa millions of men are trying to cheat on their spouses, maybe we shouldn't be embarrassed about this"? No, they said "whoa there are millions of fucked up men."

What about when all those celebrity nudies leaked? Did people stand up and say "maybe taking and sending nude photos isn't something we should be embarrassed about"? No, they said "look at this crazy photo of <Celebrity>!"

No, we said, "whoa, there are millions more fucked up men."

Broadly speaking, prejudice can be and often is (much) stronger than family ties, even if the perceived misdeed does not impact the family.

How well do you expect things to go if you move these sorts of situations into a context without any kind of strong, positive emotional bond?

A liberal society works not because things are allowed, but because people don't know. (And are taught to not stick their noise into other people's business).

> How many marriages are going to be ruined?

Not many, considering marriages are less popular among young people nowadays.

> How many politicians are going to resign?

Not many, considering older politicians maybe don't spent their time on internet, and younger politicians are aware of the state of privacy on internet

> How many future politicians will decide to never run?

Maybe we will discover than many more people are "cleaner" than we thought, and will become influencers

> How many firings and never-hirings will there be based on that history?

That's a good point. However, the mega-corporation isn't 100% of the employment. Small enterprises and freelancer just want the work to be done and get pay, without wondering about who did what in their lives.

As another comment said, maybe watching porn and being interested in more than one partner isn't that bad after all. It only shows the dumbness of the artificial rules older authorities (such as religion, army, states, etc) imposed on crowd to keep control on them.

Edit : anyway, I support the idea of a private internet space for everyone, where people can express their ideas and experience with new concepts without fearing retribution. Words don't kill people.

I think that this is the solution. Remove the count and who of the likes. You can still show the fact that someone liked it for some stickiness but you remove race to post only to generate likes but rather to share a point of view.

    > How we will get out of this, I have no idea...

I don't know. It was a faustian bargain we all made. It was fun for a while, but now it is getting increasingly ugly.

How does one get out of a Faustian Bargain, anyway?

You have to Faustian Haggle.

A Faustian bargain implies that we knew the price. Facebook actively misled (and continues to mislead) us about what we were giving up to use its service.

I always get downvoted for FB criticism, perhaps by their employees, but I have evidence for what I just said: Zuck's many, many, many public apologies.

A Faustian Bargain is the pinnacle of deception. We had no idea of the "real" price in the early naughts. You are right about them misleading us.

I don’t think people really can feign ignorance. What did we think was going on? FB built this huge free product out of the goodness of their hearts? Did nobody stop to think about how Zuck became a billionaire and pays all these employees?

When facebook came out, the vast majority of sites _did_ (or were assumed to) operate that way.

Here's the problem: many people in tech circles, including HN, passionately argued that their main monetization technique was selling ads, that they were incentivized to protect user data. But it has become clear that while they were profiting off of ads, they were also essentially giving away the data they were supposedly protecting. EFF and others sounded the alarm as it was happening but many people here and in other tech forums couldn't believe it was happening in 2009. But now that it's all in the open we're seeing a bit of history rewriting.

    > Did nobody stop to think ...?

Sure, we certainly "thought".

There were the mildly provocative arguments about facebook users being "the product". Like cattle. Many just assumed they were not susceptible to ads or strong enough not to be manipulated. Folks were thinking about this on an individual basis, assuming that yes, some dumbasses would get sucked in by Farmville but that they themselves were impervious.

Turns out it doesn't matter what any one individual thinks. Facebook is all about scale and knowledge. They're selling enormous social power to whoever forks over the money. It doesn't matter to facebook if the real customer is selling soap or authoritarian government.

I think only now are we getting wise to the full implications of Facebook and the like.

It was evidently to show you ads. Showing ads is lightly problematic, but not inheretly evil.

The privacy concerns were pretty much hidden from the public the entire time.

"Showing ads" is a phrase that covers a wide range of activity. Personally, I don't see anything wrong (and a lot of good) with a polite "hey, this thing exists" message. But once ads start trying to grab your attention, subvert your reasoning (often to your own personal detriment)... that's Evil. It's smeared out into billions of tiny bits, but that doesn't really make it any better.

Polite advertising combined with targeting is also potentially very problematic. Passive targeting (people who are interested in X do Y so I will advertise X near Y) seems fine, but it in this day and age it intersects with tracking... which brings us back around the circle.

I think the dichotomy was the language Facebook used to discuss their mission, not as profits but as a social company solely doing good. For example, in 2012, a book they gave to all their employees read:

"Facebook was not originally created to be a company, it was built to accomplish a social mission—to make the world more open and connected."

I wrote a tweet series critiquing this:

https://twitter.com/nemild/status/1006533287378968576

Some of us didn't join the SN or believe in it. But then Google took the low road and most of us thought they were benevolent.

From what I understand is that people want to feel good that they were not complicit in this all along.

Zuckerberg is the visible FB hood ornament, but people like Umanov are arguably still ultimately calling the shots

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

I was in high school when I joined. I have to say, I didn't think about it at all.

Then immediately doing the thing again for which he apologized - not to forget. It is called, plainly, being a notorious liar.

Did he say he wouldn't do it again? Or did he only say he was sorry?

You're probably being sarcastic, but I remember him saying "we have to get much better at this" about privacy. It was before new dark patterns UI, to accept tracking, that clearly violates GDPR, was rolled out.

It wasn't just Facebook. Most of the valley was effectively bought off in the Instagram acquisition, and even now much of the wealth is directly or indirectly linked to FB. There's a famous Upton Sinclair quote which neatly explain the seemingly strong pro-Facebook bias here :

> It is difficult to get a man to understand something, when his salary depends on his not understanding it.

>A Faustian bargain implies that we knew the price.

As someone who for years often pointed out the bargain, I am not sympathetic to your position. Almost everyone I know who knew the price did pay it. Often, it was not merely the lack of knowledge, but actively putting their fingers to their ears to drown out inconvenient voices like mine.

This is one of those cases where there were plenty of "internet wackos" sounding the alarm, all the way back to the launch of Gmail, and everyone ignored them. Even those sympathetic to the point for the most part just accepted it and moved on.

The doomsayers were prescient, and to say they weren't there isn't just unfair, it's actively hostile towards societal growth and detecting this as it happens again. The first step to growth is accepting fault, otherwise there's nothing to fix.

People seem to forget how god awful email had become with spam back around gmail was launched. Gmail from my recollection was the first service that really managed to deal with spam in a way that worked, was free, was simple and which everyone could use.

That's why people used gmail and it's why people still use gmail and why even the people arguing in this thread that google are evil are all still using gmail! (I'm generalising but I'm betting most do)

When it comes down to it, despite what people say they care more about getting good services at no cost to themselves than they do about their privacy.

>People seem to forget how god awful email had become with spam back around gmail was launched.

While I used spamassassin (not simple), my friend used Thunderbird which did a good enough job.

>That's why people used gmail and it's why people still use gmail and why even the people arguing in this thread that google are evil are all still using gmail!

In the early Gmail days, I didn't know anyone who used Gmail for anti-spam use. It was all about:

1. Space (1GB)

2. Nice interface was a distant second.

Most average people I know did not care about spam during those days. I mean, they cared and complained a lot about it, but getting anyone to change their provider (or software) to handle spam better was an exercise in futility.

Keep in mind - this is during the days when IE 6.0 reigned supreme and had something like 90% of the browser share. Almost no one I knew was willing to change browsers, despite IE not having tabs, not having banner ad blocking or any other extension, and causing all kinds of Windows issues. If they're unwilling to change browsers, I can assure you they're unwilling to change email providers so easily. What tipped things in Gmail's favor was the extra space, when 50MB was advertised as large).

>When it comes down to it, despite what people say they care more about getting good services at no cost to themselves than they do about their privacy.

Agreed there.

> While I used spamassassin (not simple), my friend used Thunderbird which did a good enough job.

Err, what? SpamAssassin is a filtering system and framework, Thunderbird is a MUA. There were plugins to use ApamAssasin in Thunderbird, and there may have been some built in filtering in Thunderbird, but you'll have to do a lot better than just saying it was good to convince me it existed as anything other than a joke.

My job during the time Gmail launched and the years following was administering portions of an ISP, including mail servers and a separate set of SpamAssassin processing servers, as well as separate consulting work setting up SpamAssassin and a mail server for colocation customers.

I also used Thunderbird off and on for a number of years during this same period (when it wasn't Thunderbird, it was Mutt). No MUA without something like SpamAssasin (or some paid for service or package) was even passable during this time period, to my recollection.

> In the early Gmail days, I didn't know anyone who used Gmail for anti-spam use. It was all about:

Believe me, they did. People noticed it. ISPs noticed it. The ISP I worked for thought about shifting over all email for our customers to hosted Gmail, because the spam solution was actually better than SpamAssassin usually could do and it would offload a lot of work. It didn't happen, but it was at least thrown around as a possibility.

> Most average people I know did not care about spam during those days.

My experience fielding support requests during this time period points towards the opposite conclusion. People cared a lot more, it was much more of a problem then. Google trends backs this up.[1] It was bad enough that legislation about spam was being passed during this period, such as the CAN-SPAM act of 2003.[2]

I think perhaps you've forgotten just how bad it was, or were lucky enough to have an address that wasn't very badly affected. You were running SpamAssassin, so you would likely have been one of the better protected (and you could tweak your own rules, which might have helped if you did so).

1: https://trends.google.com/trends/explore?date=all&geo=US&q=s...

2: https://en.wikipedia.org/wiki/CAN-SPAM_Act_of_2003

The Bayesian spam filtering in Thunderbird was really good because it could be trained to the kind of spam you received personally. It was not a joke; I survived on that for my work email for half a decade before they finally bought an appliance to do the heavy lifting.

I guess it was, I'm seeing reports of it working well in the bugzilla for Mozilla from around that time period, so I'll concede that Thunderbird had filtering and it may have been fairly good.

I think my larger point, that Spam was a major problem of the time period, and that Google was seen as having very good spam detection, still stands.

I believe the standard process is to win a fiddle contest.

Touché !

All the violin virtuosi are probably on Facebook, so they can get us out of the bargain.

Paging Joshua Bell[0]

[0]https://en.wikipedia.org/wiki/Joshua_Bell

Be ye angels?

Nay we are but men!

If you win, you get a solid gold fiddle and out of the contract. If you loose you get a smaller silver fiddle and they kill one of your friends.

> it is getting increasingly ugly

This is what's scary. The ugly part is just starting, because we haven't even gotten to the point where their AI systems have super-human intelligence and can use all of your "whole-life/everything you've ever done" data against you.

How bad is this going to be in 20 years? And does anyone really think Google and Facebook will "self-regulate" sufficiently on this, without extreme outside pressure?

A good start is to get rid of your Facebook account and try to find alternatives to Google.

Sure, DDG is not quite there, but for most of my queries it's just fine.

Having "deleted" my Facebook account some four years ago was a real liberator, though.

DDG is definitely "good enough" for 90+% of my searches. I occasionally have to reword something or change a keyword. Maps and Gmail on the other hand...

I don't get why people recommend DDG. It's mostly a different skin on bing search results. Why is relying on bing better than relying on google? If DDG had its own engine, I could see the "they're the small underdog focused solely on this" angle, but as-is, you have the choice of Microsoft or Google.

There's that little thing that they don't track and bubble you.

They do have their own engine.

Not for regular search results.

https://duck.co/help/results/sources

What is a "regular search result"? Your link says:

"DuckDuckGo gets its results from over four hundred sources. These include hundreds of vertical sources delivering niche Instant Answers, DuckDuckBot (our crawler) and crowd-sourced sites (like Wikipedia, stored in our answer indexes). We also of course have more traditional links in the search results, which we source from Bing, Yahoo, and Yandex."

I use DDG because I like its !bang searches and I genuinely prefer its Instant Answers and page layout over Google's.

The last part of your quote is what most people associate search results with.

I don't like to comment on downvotes on my posts but seriously? Do people really think that the actual links when you search are not the primary search results?

Good question. (I didn't downvote you!) Reading DDG's FAQ, it's not clear to me if "more traditional links in the search results" means the search results (after the Instant Answers) after exclusively sourced from Bing, Yahoo, and Yandex or if they mean "mixed in" with links from the other sources.

Is the web too big for a new search engine company to compete with its own crawler? Or are Google, Bing, Yandex, and Baidu the only search engine companies big enough to crawl the web effectively?

DDG themselves elaborated a bit about that in their previous Sources page:

> Bing and Google each spend hundreds of millions of dollars a year crawling and indexing the deep Web. It costs so much that even big companies like Yahoo and Ask are giving up general crawling and indexing. Therefore, it seems silly to compete on crawling and, besides, we do not have the money to do so. Instead, we've focused on building a better search engine by concentrating on what we think are long-term value-adds -- having way more instant answers, way less spam, real privacy and a better overall search experience.

Look up filter bubbles[1]. The problem with a search engine that (helpfully) tailors its results to you, and keeps improving that tailoring over time, is that it will cut you off from results that others would see (unless the others are exactly like you). In the short term, having better results is great, but in the long term you are in a bubble.

While DDG does use Bing (and others), the DDG searchers are getting the same results because Bing can't track and tailor your results.

Many also like DDG's bang[2] which easily lets you direct your search elsewhere. For example if you search for "example" and don't like the results, then "example !g" sends it to Google. I like it for slightly more obscure stuff - eg "example !homedepot" does what you'd expect. For us python folk, it is nice to do "example !py2" or "example !py3" depending on the codebase.

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

[2] https://duckduckgo.com/bang

It took me forever to figure out that their bang feature exists. Not that I didn't want the privacy stuff eliminated when using search, but a filter bubble is REALLY useful for solving programming problems. Straight DDG gives terrible results for engineering questions. Especially when they are edge cases. Bang is rad. They should really make it more obvious when you're using the site.

The problem DDG has is not understanding the search terms. eg "usb 3 voltage" will get you lots of results as though you had left out one or two of the words (there are far more pages about "usb voltage" than "usb 3 voltage"). You know the 3 is critical, but they don't. A general rule of thumb is that 3 or more words in your DDG search dramatically worsens the results. (Amazon search is even worse where two words begins the spiral of spectacularly bad results.)

That said DDG often does get reasonable results. And if you don't like them, just add in !g to see what google gets. (Conceptually I like to give DDG first crack at a search, and then fallback to google if DDG didn't do well.) What I have noticed is google's results are getting worse, as sites are getting better at gaming google.

Meh, to look outside your filter bubble you can just open google in an incognito window. I do that from time to time, and for web search, filter bubbles don't have much of an impact imho. I mostly search for technical things though.

> It was a faustian bargain we all made.

I didn't make that bargain, but I think the real problem these days is that it doesn't matter if you made it or not. All it takes is one weak link in the chain and suddenly you're under surveillance from a company you had no intention of being involved with.

I don't, for example, use GMail but if I communicate with people who do then my mail falls under Google surveillance. I don't use Google Drive but if I do business with a company that does then some of my data has been forked over to Google and I'm unlikely to ever know about it. I don't use Facebook but if I appear in a photo that someone else uploads and annotates then Facebook has a photo ID of me without my say-so.

It's difficult to stay clear of surveillance companies when you're constantly being undermined by the unintentional (or, perhaps more precisely, unthinking) actions of even just a few people. My data's being bought and sold every day and there is no way I can control that, opt out of it, or even get a list of companies who have my information or what they're doing with it.

If this is the information age and if information has value, then my information is my asset and I want control of it. Without the ability for a person to assert sovereignty over their own data, how can there be a balance of power between the individual and large organizations seeking to exploit everyone's data?

     > ... it doesn't matter if you made it or not.

Yep, like in the film "Rosemary's Baby". It was Rosemary's husband that made the deal with the devil but it was she who ended up carrying the child of Satan.

Spot on. Opting out of the weaponized, automated surveillance machine is not really possible.

I do block FB, Google and a bunch of others at the IP level. There are fallbacks, but I do end up leaking when they stand up new IP blocks, and the smaller surveillance shops are impossible to keep up with.

I also pee in their datasets sometimes, but that's a more experimental/entertainment activity. And I'm playing with some other ideas not ready for prime time.

> how can there be a balance of power between the individual and large organizations seeking to exploit everyone's data?

Regulation. That's what you do with antisocial behavior, negative externalities and worse. The transnational nature of the net adds complications and freeholds, but the big boys absolutely could be brought to heel, if people wanted to. Politically, the winds seem to be blowing in the other direction in the US, and the GDPR has been argued to death here.

But that's how the situation improves. (It will never be "fixed", absent an event that destroys the ability to run networks.)

Any more info (eg. tools + source) on how to pee in their datasets?

I think a very effective way is to get much better control over your cookies, when you allow them to be set, under which conditions, on which sites.

I use a combination of CookieBro and Umatrix on Firefox for this. Cookies are only allowed for sites I actively want to log into. For sites that require cookies to function, periodic auto-deletion via CookieBro handles those.

I shouldn't overstate what I'm doing - this is largely an exercise in reversing what they're doing. And when I figure it out I might have a little fun by way of verifying I figured something out.

Tools - I use Charles Proxy and perl with WWW::Mechanize (substitute your recreational language of choice). Sometimes JMeter is helpful, but that's kind of a rathole if you haven't had reason to subject yourself to it before.

"How does one get out of a Faustian Bargain, anyway?"

Well, Goethes Faust got away pretty well, beeing gods favourite .. and a grammar exploit in the bargain.

(the bargain was, Mephistopheles serves him, until he says some specific words (Verweile doch, du bist so schön) ... and in the end Faust says, now I feel like I could say "Verweile doch..." Mephistopheles thinks thats it, summons his demons, to prepare for it. Faust dies, beeing old ... but since god was the all powerful judge and Faust his favourite, ... well that was enough to save his soul. And the demons had to go away accomplishing nothing)

So any lessons for the real world? I don't see any...

edit: maybe to give more context: Faust made the deal, because he was not content, mainly about the limitations about understanding the world, but also in general. And the words of the bargain mean something like, now I am content. So Mephistos part was to make him content, or even too full of this world. But it is somewhat complicated, beeing Goethe.. but worth a read, at least in german. Don't know, if the translations are good enough, because much of the greatness of the book, is the powerful language and verses, which at least I could never translate adequately.

Let's hope Zukerburg isn't a fan of Marlowe then.

Legislation. Vote for politicians who care about these things (increasingly many do, especially the younger ones). Or stand up and run on a platform of providing a safer internet for all of us, free from the kind of tracking that we take for granted.

I agree it may not be easy. But we desperately need more tech-savvy politicians in these technological times.

So after a long article about Facebook, the first comment is about Google.

If Tesla kills a driver, it's Tesla and Waymo.

If Facebook is selling your data, it's Facebook and Google.

Why don't we stop pretending and just say we all hate Google, irrespective of who is actually at fault? That will save us trouble of having to find any evidence proving our claims.

The article is about the dangers of large companies tracking users to such a high degree that it's dangerous for us, profitable for them. Facebook is an example. Google is also an example. Personally I'm more worried about Google because they have all my e-mails and everything my phone does. I don't have Facebook anymore.

It's not about hating Google, it's about recognizing the power and ubiquity they have, and how for many in HN, it dwarfs Facebook in terms of potential personal impact.

Yeah. He is right. Anyway, thank you. I couldn't read an article so long to communicate so little, so I just went into the comments here. Thanks again.

Lots of people come to the comments without reading the article. I wonder how many of your downvotes came from people who did just that.

Reply on news.ycombinator.com