Chris Peterson

A few days ago I posted about the new Facebook privacy “features”.

One of them – “Groups” – Facebook had described as such:

With Groups, users can essentially partition their interactions (passive or active) with Facebook and create multiple, customized Facebook experiences. For example, a user who participates in a “neighborhood” group can – with one click – view a newsfeed that is visible only to members of that group, post status messages that only members of the group can see, and peruse a list of profiles that includes only group members. This new functionality will make it much easier for groups (lowercase “g”) of friends to keep in touch and will likely accelerate the use of Facebook as a platform for organizing everything from bake sales to protests.

And I said this seemed “inoffensive enough.”

Well –

Facebook is being battered by critics who say the popular social network made a big mistake in failing to let people opt-in by default to its new feature that lets people form private groups around a particular interest.

The controversy reached a head on Thursday when a person created a group called NAMBLA, the name for a nefarious pro-pedophile organization, and started adding friends.

One of the person’s added to the group was well-known tech blogger Michael Arrington, who in turn added Facebook Chief Executive Mark Zuckerberg, Chester Wisniewski, a senior security adviser for security firm Sophos, reported on the company’s blog.

While not actually from NAMBLA, the group was formed to make the point that Facebook was wrong in choosing to let people automatically add their “friends,” and leaving it up to the added person to opt-out of the group.

via InformationWeek.

So Facebook can’t even let users click “yes, please let my friend add me to this group” before doing it.

How can one company fail at variations on the same thing so many times?

(cynic: because it is their intent to fail)

Grimmelmann passed along the EFF’s take on the Facebook privacy changed I blogged about yesterday. The EFF had a much better breakdown, critique, and set of recommendations – it’s a good read.

It’s been a month or so, so Facebook has announced some new privacy changes. CDT has the breakdown:
(continue reading…)

Must admit that my blood boiled a bit on the very first page of the CQR white paper at this:

“The days of you having a different image for your work friends or co-workers and for the other people you know are probably coming to an end pretty quickly” because of the dominance of social network sites — where people use their real names — and the extent to which information is now shared online, said Zuckerberg. That’s good, he said, because “having two identities for your- self is an example of a lack of integrity.”

(emphasis mine)

When I was in high school, I once gave a presentation on the Watergate scandal, in which I referred to Nixon as “for lack of a better word, a scumbag.” A few conservative teachers on the panel mildly reprimanded me, arguing that, whatever shortcomings Nixon may have had as a human being and political leader, surely there must be better, more precise, and more meaningful words than scumbag.

But sometimes, you have to call a spade a spade, and a scumbag a scumbag. And Mark Zuckerberg, for lack of a better word, is a scumbag. There are perhaps better, more precise, more meaningful ways to describe someone who, after citing as justification the evisceration of privacy of which he is the prime butcher, has the nerve to say that those who wish to keep their contexts intact have a “lack of integrity.”

I could go on, for some time, for how conceptually incorrect this is (but I think I’ve written enough about that in Losing Face), and for how absolutely richly pathetic it is for someone whose fame and fortune derives from a stolen business (itself founded in order to take petty potshots at a girl with the good sense to turn him down for a date) to accuse others of a “lack of integrity.”

But brevity is the key to writing as well as wit, and so I’ll happily settle to call him a scumbag.

I was (fairly extensively) interviewed for the CQ Researcher’s newly-published white paper on privacy and social network sites, authored by the inestimable Marcia Clemmitt. Unfortunately, it’s behind a serious paywall, so I can’t post it here – but there’s good stuff in there from the usual crowd, and hopefully it will serve as a useful guide to the sorts of folks who subscribe to CQR.

Which leads me to…

Google’s orkut gets the right idea:

Imagine Sarah, a 21 year old girl who just created an orkut profile. To get started, she adds her college friends to her friend list. They share photos, join communities, exchange scraps, discuss everything that’s hot on campus. A few days later Sarah finds out that some friends from high school are also on orkut and adds them: what’s better than keeping in touch with old friends?

Then Sarah gets her first job and adds her office colleagues to orkut: you can’t decline a friend request from your boss, can you? Sarah’s social network keeps growing. Even her parents, aunts, uncles and cousins are on orkut, and she adds them to her friend list.

The college gang, old friends from high school, office colleagues, family, everyone is in Sarah’s friend list now. Soon enough she will not be able to share anything with anyone anymore – after all, jokes and photos from the office party should be shared only with her work colleagues. Scraps and photos of her baby nephew at the family reunion should only be seen by members of her family. The plans for Saturday night and the photos of the parties she went to should be seen only by her party friends – Sarah does not want her boss or her young cousin to see those.

Just like Sarah, we all maintain different groups of friends, and the Internet was not able to reflect that. Until now, social networks treated people from different groups like they were all the same: they were all “friends”.

So we asked ourselves: does it need to work this way on the Internet? Can we reproduce our groups of friends from real life on the Internet? The answer is “yes!” Starting today, we will change the core function of orkut so we can share and interact with different groups of friends on the Internet just like we do in real life.

At least someone has the right idea.

Sorry for the lack of blog posts lately – the admissions cycle is starting, which means my blogging will probably become somewhat infrequent over the coming months.

When Facebook announced its places feature, you may have wondered “hrm, how long will it be before this undermines my privacy?”

Nick O’Neill at AllFacebook has some observations:

One feature that has attracted a fair amount of buzz is the ability for your friends to tag you in different places. That means you may not actually be somewhere, yet your friends will tag you as a joke and now you’re showing up at a random strip club.

…

While you may be fine with Facebook’s existing Places privacy settings, I know there are plenty of friends on Facebook who I don’t want to track my location.

…

One strange thing about Facebook Places is that despite controlling who can view your location information from within your profile with the previous setting, anybody who visits a location will potentially be able to view that you’ve been there before.

Nick runs through the ways to change your privacy settings. It’s worth the read, but here’s the short version:

• Go to the Privacy Tab and click “Customize Settings”
• Change your settings. For example, I disabled allowing my friends to check me in elsewhere, and noone can see where I check in.

I’m not big on the whole locations movement. Maybe you are, and that’s fine. But if you aren’t, Facebook just pitched you a curveball by opting users into the Places feature, so here’s how you opt out.

From AllFacebook, on a presentation by Google on a rumored new social utility:

The overall theme of the presentation was consistent: we have multiple groups and within those groups there are individuals who we have strong ties with and many more who we have weak ties with. There are also even temporary ties, like the person at the restaurant who served you food last night. While getting the system right on this is extremely difficult, the strong vs. weak ties is something that Facebook has yet to enable users to control.

…

If Paul Adams’ presentation is accepted as one of the primary perspectives of Google on social, perhaps the argument for Google’s new “Facebook killer” would be that there needs to be a more effective user-interface (UI) which helps users to control these various groups. Rather than dismissing it as a service for “advanced” users, perhaps the interface has simply not evolved far enough to give users the actual control that they want.

…

That would support the argument presented by Paul Adams in the slide below which states “If your privacy practices aren’t transparent, then you introduce doubt. Doubt leads to lower usage.” Only Facebook knows how great of an impact the latest privacy fiasco had on the company but it’s clear that Google sees this as a weakness.

If this is true, then Google has precisely the right privacy perspective to outflank Facebook on this issue. And they’re about the only company with the muscle to do it.

Two days ago, Facebook rolled out new privacy tools in a blog post by Mark Zuckerberg.

On balance, the privacy revamp represents a net benefit from where we were in the last few months. For example, Facebook has finally returned to users the ability to control basic information such as whether or not complete strangers can see your hometown.

Additionally, their new privacy dashboard (as illustrated below with a picture from the site):

Is a helpful way for people to begin to visualize what is available to whom if they select one of Facebook’s settings.

But – as is so often the question in policy problems, from health care reform to financial reform to Facebook – the question is not whether the reform is better than what we had, but actually “good enough” to be truly praiseworthy. By way of analogy, of course it’s better to throw a rope to a drowning man than to not throw one at all, but if he is 10 feet from your boat and the rope is two feet long, the effort may not be as laudable as it initially appears .

What more could Facebook have done? It could’ve made Instant Personalization Opt-In. It could’ve integrated some of the great tools like the Facebook Privacy Scanner or Zesty.ca profile mirror. It could’ve shown people what they currently are sharing and tweak it from there on the Dashboard, as opposed to simply giving them four options to pick from (although the four options are good for simplicity’s sake). Finally, it could’ve announced these changes in a big box on everyone’s News Feed – or forced them to visit the new privacy page the next time they visited the site – rather than hiding it in plain sight on the Facebook blog, which almost none of its users read.

I’m not pointing these things out just to complain about Facebook – I’m pointing them out to demonstrate how much Facebook didn’t do in their privacy reform. That doesn’t mean I don’t approve of the changes they did make – I do. But to understand the full context of Facebook’s actions, one must understand what they did and didn’t do. And in that respect, it’s still the latter that is far more striking.

There are many ways in which Facebook might know your phone number. The easiest (and most common) way is to give it to them, by including it in your profile so that your Facebook Friends can look it up when they need to give you a ring.

So your Friends have your number, but that makes sense – after all, they probably already had your number, or could’ve gotten it easily. And Facebook has your number, and while I don’t think it makes sense to “trust” Facebook anymore, I think users can trust in the fact that even Facebook is not so colossally stupid as to do something like sell your phone number to used car hucksters.

But who else has access to your number?

Think no one? Think again. Are you sure that only your Friends have access to your profile info?

What about those Groups and Events – you know, the ones that start “I dropped my phone in a pond, send me your numbers!” Have you posted in them? Do you know their privacy settings? Have the privacy settings changed since you posted?

The ever entertaining Tom Scott – creator of invaluable Internet entertainment such as StupidFight – has just produced the privacy equivalent of a horror film. It’s called – simply and appropriately enough – “Evil”.

As Scott explains:

This site randomly displays the private phone numbers of unsuspecting Facebook users.

…

There are uncountable numbers of groups on Facebook called “lost my phone!!!!! need ur numbers!!!!!” or something like that. Most of them are marked as ‘public’, or ‘visible to everyone’. A lot of folks don’t understand what that means in Facebook’s context — to Facebook, ‘everyone’ means everyone in the world, whether they’re a Facebook member or not. That includes automated programs like Evil, as well as search engines.

…

Evil uses the graph API to search for groups about lost phones. It picks them at random, extracts some of the phone numbers, and then shows them here.

Here is Scott’s screencast of what Evil looks like when it is working:

So what should you do? Scott says:

Go into all the “lost number” groups you’ve ever joined. Ever. Delete your posts. (You might want to try searching for your own phone number on Google, too; it might turn up in unexpected places.)

The thing to remember here is that the fault doesn’t lie with the users. That is to say, the fault doesn’t like with the users any more than someone can be faulted for stepping on what appears to be a solid deck only to have it collapse under their feet because it wasn’t built to code.

It is patently ridiculous and unreasonable to argue that all of these users wanted their cell phone numbers and names to be accessible to the entire Internet. But that’s what has happened, because of stupid, unsafe, and (indeed) “evil” design.

(h/t Ian Brown and PVN)