Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Searching Facebook drives home the privacy degradation problem

willmoffat.github.com

Post reply on HN
Print view View original

21–30 of 39 posts

Re: Searching Facebook drives home the privacy degradation problem

#21

Post by iamelgringo »

snewe wrote: post #10

Unless we assume that these people are completely ignorant of the public nature of their status updates, then this isn't a privacy violation. Is the argument that Facebook defaults to public and makes it confusing to change settings?

Unless we assume that these people are completely ignorant of the public nature of their status updates, then this isn't a privacy violation.

But that's exactly the point. Normal's don't get it.

Behavioral Economics studies show that when something is made a default, about 90-95% of people will keep that default. So, when something is made "Opt out" by default vs "Opt in", the vast majority of people won't bother to change things.

So, when Facebook changes this behavior, you can assume that going forward 90-95% of their content is going to go public.

Most normals that use Facebook aren't going to care until it starts biting them in the ass. While, many geeks instinctively understand the huge implications of everyone's status updates being made public.

Granted, you may not care, and I certainly respect that. But, at the same time, I have a hunch that many Normals, if they truly understood the full ramifications of what may happen might substantially change their behavior, or choose not to use Facebook at all.

Another issue, is this: Does making users Facebook posts public by default benefit the users? I think it benefits Facebook tremendously, but I'm doubtful how many users it's going to benefit to have their "rectal surgery" posts made public.

Re: Searching Facebook drives home the privacy degradation problem

#23

Post by natrius »

iamelgringo wrote: post #21
snewe wrote: post #10

Unless we assume that these people are completely ignorant of the public nature of their status updates, then this isn't a privacy violation. Is the argument that Facebook defaults to public and makes it confusing to change settings?

Unless we assume that these people are completely ignorant of the public nature of their status updates, then this isn't a privacy violation. But that's exactly the point. Normal's don't get it. Behavioral Economics studies show that when something is made a default, about 90-95% of people will keep that default. So, when something is made "Opt out" by default vs "Opt in", the vast majority of people won't bother to…

The opt-in/opt-out example I've seen the most is 401k contributions. If people have to opt-out of automatically putting part of their paycheck into their 401k, more people make contributions. The idea isn't that people are being tricked into it, although some people probably didn't read carefully enough and don't notice the part of their paycheck missing. The idea is that people are less likely to change from defaults in general. Most people know they should save, so actively saying "No, I don't want to save" by checking an opt-out box feels worse then leaving an opt-in box blank.

I don't think most people are mistakenly sharing things with more people than they intended on Facebook. When they made the privacy changes and presented everyone with the dialog that let you set all of your privacy settings, 35% of users picked something other than the defaults[1]. It was pretty clear what was going on. For a refresher, here's a video that walks through what the transition actually looked like: http://www.allfacebook.com/2009/12/how-to-use-facebooks-new-...

The only thing Facebook handled poorly in terms of privacy was Instant Personalization. It was... instant. Yelp/Pandora/Docs have my Facebook data linked with my account from the moment I visit the site, which may not be what I want. (I actually like it on Pandora and Yelp. I don't use Docs.) To make matters worse, it's annoyingly difficult to opt out of Instant Personalization altogether. I think this is what offended people the most, then the firestorm made people bring up peripheral non-issues that piled on and made the PR disaster worse. Beacon redux.

[1] http://www.readwriteweb.com/archives/facebook_brags_35_adjus...

Re: Searching Facebook drives home the privacy degradation problem

#24

Post by aristoxenus »

skolor wrote: post #6
aristoxenus wrote: post #5

Earlier quoted context omitted.

Let me put it another way: we all have a problem right now that we're trying to solve, which is that FB is screwing people. Your clever idea is screwing them worse, with no foreseeable gain for ethics either now or long-term that another demonstration couldn't provide. If you really care about the problem, deactivate your project and come up with something more clever that also takes the moral high ground.

This argument just doesn't make sense. Its trivial to make this search, the OP just put a nice wrapper around it. There are two possible choices he could have made: Something shocking, like "playing hooky", that could get people in trouble. In this case, people will get in trouble, because someone will find out they were playing hooky and report them. However, because people are directly being effected, a shit-storm…

To someone with the proper technical skills, it's trivial as you say. But until you do something like what these guys have done, it's dishonest (or geek-myopic) to say that it was trivially available before. When you package it up like this to inject into a receptive media channel, you make it "available" to an audience that would not have had access to this information before.

The rhetorical impact of choosing "playing hooky" over "weight" is obvious, and not the problem here. The problem is that someone whose main motivation was obviously geek cred and not solving a problem went straight for a collateral-damage campaign instead of one of the many other options available. Fixing the Facebook privacy policy is not worth endangering the livelihood of even one of these people.

Re: Searching Facebook drives home the privacy degradation problem

#25

Post by piguy314 »

aristoxenus wrote: post #24
skolor wrote: post #6

Earlier quoted context omitted.

This argument just doesn't make sense. Its trivial to make this search, the OP just put a nice wrapper around it. There are two possible choices he could have made: Something shocking, like "playing hooky", that could get people in trouble. In this case, people will get in trouble, because someone will find out they were playing hooky and report them. However, because people are directly being effected, a shit-storm…

To someone with the proper technical skills, it's trivial as you say. But until you do something like what these guys have done, it's dishonest (or geek-myopic) to say that it was trivially available before. When you package it up like this to inject into a receptive media channel, you make it "available" to an audience that would not have had access to this information before. The rhetorical impact of choosing "play…

You sound exactly like somebody who opposes full disclosure of security bugs because you think it makes systems less secure. Shining the light of day on problems is the only way to solve them. Full disclosure is the only way to move big companies, it keeps them accountable.

Re: Searching Facebook drives home the privacy degradation problem

#26

Post by aristoxenus »

hamstersoup wrote: post #14
rictic wrote: post #11

Earlier quoted context omitted.

We briefly added these measures, but after thinking it over we've decided to disable them. From a technical standpoint they're a fig leaf. This isn't a complex server-side app, it's a minimal UI on top of the JSON results Facebook returns for these searches: http://graph.facebook.com/search?q=control+urges&type=po... . This is frankly the least scary use of this data. For example, it would be trivial to start crawlin…

Here's an example of the fig leaf in action: http://willmoffat.github.com/FacebookSearch/?q=HIV%20test&#3... We decided the redactions reduce the impact without actually offering any privacy at all. (That's up to Facebook and the users).

"Full disclosure" is a discussion about the ethics of publishing an exploit. Publishing exploits is customarily done in a descriptive manner -- I've never seen it done by publicly sharing the spoils of using that exploit.

Did you try publicizing it without the full identifying data available and measuring the response? Did you consider a strategy of escalating outrageousness, instead of going straight for this course of action?

What about automatically contacting the affected users first, and attempt to rouse them to action?

I'm sorry to be so harsh in a public forum, but when someone takes it upon himself to say that the affected lives are going to suffer for a good cause, then he'd better accompany the resulting campaign with a very thorough -- and thoroughly-vetted -- piece of argument explaining exactly why the ethical balance is in his favor. Two guys deciding they'd get more pageviews by going with plan A, and leaving the moral debate for blog commenters after the fact is not a thorough vetting.

There were already numerous forces at play which could potentially result in FB getting things straight. Your app won't have accomplished anything that wouldn't otherwise have been accomplished, except perhaps to cause a few more people to suffer.

Re: Searching Facebook drives home the privacy degradation problem

#27

Post by Chirono »

rictic wrote: post #13
Chirono wrote: post #7

Earlier quoted context omitted.

That seems rather over the top. I can't really see this ruining anybody's life. It isn't letting me search for specific peoples comments. I don't know any of these people, nor am I likely to meet them. This website reminds me of being outside, with people, and hearing little snippets of conversations from strangers. Nobody goes to a park and then complains the council that strangers where able to hear their conversat…

I think that Facebook has failed in two clear ways. They do not do a good job of indicating how public an action will be. This is true across the internet, but it's a real problem on Facebook due to the strong ties to your real identity there. They change the rules far too often. If you had a very strong grasp of Facebook's privacy policy two years ago (or even six months ago) that information would be worse than use…

Thanks, that makes sense. When you add to that the internet's characteristic does of indignant shouting about anything and everything, than I can start to see why these reactions are coming up. My grasp of facebook privacy has always been pretty simple: anyone can see anything I do, except one on one messages. I know this is /way/ over simplistic, but it would explain why I don't feel scandalized that other people can see my wall.

Re: Searching Facebook drives home the privacy degradation problem

#28

Post by GabrielBen »

bloodnok wrote: post #17
skybrian wrote: post #16

Earlier quoted context omitted.

So how do you find out whether they're ignorant of that? A good way is to search for things that people probably didn't mean to reveal publicly. If a large number of people don't know what they're doing, that's a UI design fail. It doesn't matter whether they should have known. (I suspect Twitter has the same problem. People should know it's completely public, but I suspect many don't.)

When I talk to people about it, there seems to be a general apathy around the privacy aspects. Most are of the opinion "why would someone other than my friends be interested in me and what I did on Saturday night?". I think the first wake-up call comes when their parents join facebook. Otherwise it's being turned down for a job without knowing why, getting fired for bragging about throwing a sickie, being harassed/st…

Im about to move out with my girlfriend, and my realtor checked my background with Facebook. I only knew about this because that guy turned out to have gone to high school with my boss, and my boss told me.

Even though im a programmer, the only thing i do in facebook is add friends, it already affected me to just have an account there. The effect was neutral to good..this time.

Re: Searching Facebook drives home the privacy degradation problem

#29

Post by rmc »

dschobel wrote: post #8

http://twitter.com/#search?q=hooky I know it's trendy to hate FB right now but don't underestimate the ignorance factor in all of this.

A big difference is that from the start Twitter was public whereas Facebook was private. Then Facebook changed it to my public by default. Since most people don't change from the default, people are using service (Facebook) that they thought was private, but is now actually public. It's the change from 'default private' to 'default public' that's the biggest thing.

Re: Searching Facebook drives home the privacy degradation problem

#30

Post by rmc »

snewe wrote: post #10

Unless we assume that these people are completely ignorant of the public nature of their status updates, then this isn't a privacy violation. Is the argument that Facebook defaults to public and makes it confusing to change settings?

Initially Facebook was private by default. It's only recently that it's changed to 'default public'. Lots of people are using a public thing, when it used to be private. It's the change that's the most important thing.
Post reply on HN