Live data from Hacker News

Is Facebook Really Scarier Than Google?

nautil.us

271–280 of 360 posts

Re: Is Facebook Really Scarier Than Google?

#271

Earlier quoted context omitted.

I don't feel like many people would though, even if it was a one-time $0.99 purchase.

One time $0.99 won't justify this. Companies can make far more money per user from ads. It has to be a subscription model for it to be tenable.

Yes, and even fewer people would pay for that on a continuous basis like a utility. If there was a free alternative everyone would just use that one and suggest it to their friends.

Although I suppose there could be a certain exclusivity and elite aspect to such a network, similar to how Facebook differentiated itself with exclusivity by only being available at universities in the beginning. A network that markets itself as both for higher class people willing to pay monthly and free of ads/tracking.

Re: Is Facebook Really Scarier Than Google?

#272

Earlier quoted context omitted.

Uh, Google and Mozilla and a handful of organizations are converting all web traffic to HTTPS through subtle manipulation of standards and user interface. They even proxy mobile traffic from their search engine. ISPs can't even see what IP you're visiting because it's all fronted by CloudFlare. Even your DNS is being hidden by an HTTPS tunnel in the browser. Your ISP soon won't be able to see anything you do online.…

There's still SNI in TLS. Hopefully that changes in 1.3, but I don't know how prevalent that is yet.

Wait. Wait a minute.

I was planning literally this month to figure out how to implement domain traffic analytics on my LAN by using SNI via libpcap or similar (something tiny).

Will I have to configure a root cert and build a full DPI proxy to make this futureproof? >_<

Re: Is Facebook Really Scarier Than Google?

#273
Google surely can draw a network of its ever-logged-in users just with their ad services alone:

1. You visit a page with some Google ad at home.

2. Someone else at home visits another page with some Google ad.

3. Your browsers make the request through the same IP address, marking you as potential acquaintances.

With the frequency of this happening, they can put weights on the edges to mark how closely you are related to that other person. And since they can, they will. Otherwise, it would be the loss of this great opportunity for them.

I don't see Google being any less manipulative than Facebook, with all the capable tools they have now. They are ever-present, can track you almost regardless where you visit (ads and analytics), they can gather information about your profile/network, and most importantly, they curate the information presented to you (search and ads).

Re: Is Facebook Really Scarier Than Google?

#274

Earlier quoted context omitted.

> Because they're slow as shit, Tor isn't that bad actually for browsing, while i2p needs more love regarding speed. > difficult to configure correctly, For Tor: You just download the Tor Browser. Already pre-configured. > and make you more conspicuous to most three-letter agencies. Good argument for actually using them.

> You just download the Tor Browser. Already pre-configured. Isn't Tor Browser fairly bad due to it being a target? I'm not sure if that changed recently, but I recall seeing lots of "don't use the Tor Browser bundle". Tor also relies on exit nodes to exist, yet it's considered very dangerous to run one.

> Isn't Tor Browser fairly bad due to it being a target? I'm not sure if that changed recently, but I recall seeing lots of "don't use the Tor Browser bundle"

Just FUD.

> Tor also relies on exit nodes to exist, yet it's considered very dangerous to run one.

It's dangerous in some places to run an exit, not everywhere.

Re: Is Facebook Really Scarier Than Google?

#275

Earlier quoted context omitted.

The scandal is that third party companies like Cambridge Analytica harvested the whole social graph and used it to manipulate people. People know that Google and Facebook track as they are ultimately ad companies, and if you get a free service people know that your data is the product. We pay for ISPs to protect privacy, not sell it off, that is the big difference. Many people have seen Facebook tracking from the mar…

I'd like to expand this because I don't see it written much, and please correct me if I misunderstand: The scandal is that FB sold or allowed the ability for third party companies like Cambridge Analytica to harvest the whole social graph and FB sold or allowed the ability for them to use that same data on FB to target and manipulate people.

It is nuanced a bit so some background: Cambridge Analytica did violate Facebook terms of service as that wasn't truly allowed to pull down the whole social graph, but the protections against pulling friend data without their knowledge in the Facebook OpenGraph APIs weren't truly in until v2 around 2013-2014.

I know this because we used to do lots of Facebook apps/games and back then, once someone gave you access to their information, you could get all their friends and all their information and recursively pull down most of the social graph for public information. Most games were just using it for friend names, if they played the games, invites and competing with friends but there were bad apple apps out there harvesting it all down. The facebook app revolution was partly due to the data element and was open for many years.

It always surprised me how much data could be pulled, it is part of the reason Zynga was so effective as well and attracted some oligarch money. Part of the reason Facebook started locking it down is game/apps were getting more adept at pulling all data and Facebook was scared someone would become a social graph competitor, so they locked it down mainly for their own needs not really privacy.

Who knows if Cambridge Analytica had extra access beyond that to get to profiles that weren't public, but most profiles were public by default back then and people only put information online that they wanted to share publicly without as much expectation of privacy. Over time people for some reason started to trust that Facebook was protecting their data but still had the friend permissions access hole.

Back in the late 90s and early 00s people were very against sharing any real info on the web with sites previous to and like Facebook, it slowly changed as the appearance of privacy was added but truly it was still wide open if even one of your friends gave access to the app until Facebook v2 OpenGraph. With the OpenGraph v2 friend lockdown changes, you could only get a friend ref id only available to your one app that wasn't their actual facebook id and was different per app, and you could send them an invite but not pull their data until they agreed which it should have been all along.

After that change it was an era of tons of invites on Facebook and companies like Zynga threatened to leave and did try to build their own, it also shut down many Facebook game companies that could no longer get the numbers, many moved to mobile that was still wide open. Zynga was given special privileges by Facebook for a while due to this where others didn't have that access, others may have also had those special privileges. Facebook transition to mobile took a long time and some people even thought Facebook wouldn't be able to make the leap. At that time, the app/game companies on Facebook considered it Facebook killing the viral nature of some of those apps/games which was ultimately good. It was a huge mistake for Facebook not to separate app/gaming from your personal info and friends but that was the product then, they should have allowed people to setup app/gaming profiles that other app/gamer users could friend each other and not pollute your main friends list and pull all your social graph data. Games were a bit of a trojan into your social graph due to the setup back then.

It is possible that Cambridge Analytica had other access to non public data but as I mentioned, most data was public by default then and in a way CA was late to the game, many companies probably had people internally that could pull it down and possibly even from data centers, Facebook eventually built their own data centers. Then there is the whole side where the NSA had any access they wanted or needed as well to both public and private data, who knows if that was exploited or not. Cambridge Analytica used their data for nefarious purposes against the ToS of Facebook but that was bound to happen because it was the move fast days and security was an afterthought. In theory you could still have a network of apps that combine to get people to give you access to their data and friends approve it as well but most of that has moved to mobile rather than facebook apps as that is easier on mobile now and people moved there including Facebook themselves.

Really this whole adventure was spurred by the Web 2.0 era that people were being social and sharing more online and it was democracy online, more public, previous to that it was very limited. Web 2.0 launched this site, reddit, Facebook, Google social products, comment systems etc. So I think there was a temporary time where it was the Wild West of data mining and people sharing more than they should with the expectation of privacy because Facebook was a walled garden and people thought it made their data safer. Turns out that was not the case if they didn't specifically mark it private.

Many of these issues still affect mobile but that is getting better, however the Facebook apps probably pull more from mobile to build the social graph than they ever could on the web including calls, audio and other things that mobile allows you to do as it is native and not sandboxed like the web. Sandboxing via web browsers was huge back in the day because people were so worried about their private data and hacking, that went away for a while, Web 2.0 happened, mobile happened, data was misused, now it will tighten to more private/permissions again and has been for the last few years. Ultimately people knowing that data you put online or when you use apps isn't private is probably a good thing as the good that will come of all this. We might get to a right to your own data Bill of Rights amendment or similar one day.

Ultimately Facebook was not necessarily nefarious in this, companies like Cambridge Analytica that exploited Web 2.0/mobile and social networks to use that data against you, rather than just serve up ads, is where things went too far and thus the backlash. Facebook since v2 OpenGraph has been privacy/security conscious both for them to protect the social graph data and to create trust in users.

Now ISPs are getting in the game with removing privacy protections with their new law and they don't care about consumer trust as much, that is the scary one.

Re: Is Facebook Really Scarier Than Google?

#276
post #92

I know Google is profiling me and harvesting my data. But it's not using me to harvest my friends information. I know Google is targeting ads at me - and they maybe AB tested. But those ads are from real companies or organisations - they aren't fake bots or astroturf groups algorithmically designed to tell me what I want to hear. I'm sceptically watchful of Google, I feel I have a social contract with them where they…

> those ads are from real companies or organisations - they aren't fake bots or astroturf groups algorithmically designed to tell me what I want to hear. How do you know this? As far as we know Google was also used to spread misinformation by Russian entities in the elections [1]. Correct me if I'm wrong, but Google invented targeted advertising. Hell, this is the company that boasted about A/B testing shades of blue…

I'm in the UK not US so I'm n to totally sure what sort of fake news US electors get.

However during Brexit and UK elections I get political ads and memes via FB. Google don't push that stuff at me.

Google if it (rarely) pushes politics at me pushes links from political parties or identifiable campaign groups. So if it is lies or fake then I and others can hold them to account.

FB often seems to be pushing fake stories - sometime started by fake users - actually started by who knows who. So Propaganda can't be held to account.

Actually I think Twitter problems are more similar to FB - except they maybe lack the deep profiling of social networks that FB have.

Re: Is Facebook Really Scarier Than Google?

#277

Earlier quoted context omitted.

> No Maps Here Maps or OpenStreetMap > No Search I’ve switched to a mix of Qwant and DuckDuckGo, depending on question > No GMail I’ve been self-hosting all my mail for 4 years now, works fine thanks to simple solutions such as this tutorial https://workaround.org/ispmail > No YouTube This is the largest issue, but most of what I consume is either music or streams, and most of the streams are also on twitch, and the…

> I use seafile with automated upload and a custom-built photos UI frontend to provide a photo gallery for me, and to allow me to share photos. I use the tensorflow pre-trained example model for object classification to tag the images, it’s good enough. Possibly the most hacker_news.txt comment in this thread. "Why rely on [popular consumer service] when you could just build your own photos frontend and then set up a…

That’s why I said, for normal people this is insanity, but this is Hacker News - no one on this site has an excuse for using GMail. And even the photos part – every week we have another Show HN of someone building some new project with the pretrained tensorflow models.

But yes, my setup is something I don’t expect most people to build for themselves – instead I’d hope that NAS vendors would include similar functionality, and people would start using those instead.

Re: Is Facebook Really Scarier Than Google?

#280

Forget scary. Let's talk about utility. Facebook gives you distraction but Google gives you information (YouTube, Google Search, GMail, Maps, Voice, Fiber, Project Fi, etc) and is more restrictive about how it lets partners use its platform to interact with users.

This 100% Google build genuinely useful products, and use the knowledge it gains about you to show (generally somewhat useful) ads, and generally stays out of your way. Facebook on the other hand desperately tries to manipulate you into consuming as much time as possible on their platform at almost any cost, using lowest common denominator tricks.

It's just kind of a false choice. I use FB to keep in touch and chat with friends, find out about local events, follow various interesting people, etc. google tried to do the same thing with Google Plus and failed. But I certainly don't feel manipulated by FB, I feel they are providing me with a useful service.
Post reply on HN