Facebook: "We can do anything we want with your content. Forever"
consumerist.com
consumerist.com
We all know that removing something which has been posted to the web is like removing the proverbial drop of food coloring from a swimming pool. With their new TOS, Facebook is stating something that's been true all along: They're running a swimming pool. Once your content is exposed to its userbase you can't take that content back.
I am reminded of a passage in J. Michael Straczynski's screenwriting book where he talks about how important it is to avoid sending unsolicited manuscripts to movie or TV producers. They will send the manuscripts back unopened [2] and refuse to deal with you again, because:
One individual I encountered had written a spec Terminator 3 script, hoping to either send it to producer James Cameron and later sue Cameron because this person was sure this would be the next story or tempt Cameron to sue him, which would force Cameron to read the script, after which he would be so thoroughly blinded by the script's brilliance that he'd buy it instantly. (No, I'm not making this up.)... Unsolicited manuscripts are the constant nightmare of any producer.
Perhaps Facebook has become large and old enough to have had its first encounter with such a copyright troll. The solution to this problem, as employed by every producer in Hollywood, is to require a release form before you'll read anything.
---
[1] Note that I am not a lawyer. The lawyers in the audience are welcome to explain that I'm wrong.
[2] "It will be returned in a larger envelope, unopened (although there will often be a tiny tear in one of the corners of the original envelope, made by a secretary to verify that it contained a script)." -- JMS
In France we have somthing supposed to ensure data privacy called "CNIL" ( http://en.wikipedia.org/wiki/CNIL ). Is there some equivalent in the US?
Meanwhile, if you're saying that it would be great if we had a regulatory regime that recognized the reality of the Internet -- where anything you publish effectively belongs to the world and cannot thereafter be retracted or centrally controlled -- amen to that! But that's not the world we live in. We live in a world full of lawyers who mail C&D notices to indie filmmakers if they happen to photograph a building or a sign or a person without getting a signed release first; a world where a lot of modern art can't afford to legally exist because it costs too much to clear the rights to all the components of that art. Under those circumstances, I find it hard to blame Facebook for throwing up their hands and demanding a waiver before you publish anything on their platform.
The net and the web were built like that, but now why are we building "jails" within the compelling interactivity of the "web 2.0"?
There are numerous interesting protocols to build upon (openid, the extensible XMPP..., etc..), and startups should use them and respect their users data. Concerning social networks there seem to be some alternatives (Elgg, OpenSocial APIs) but could a startup build something more open and still profitable on those?
The fact that social networks gain value with users means the solution isn't to "break up" a site like Facebook. That only worsens everything for users.
So all their solutions - like Facebook connect - will stay them-centric.
Facebook's technology and the terms of use of its API attempt to strike a balance for issues like this. Interoperability is fine on paper, but FB isn't just a website; it's a part of the plumbing of people's real lives.
("abuse" and "unfairly" in quotes because there is a lot of complex legal code defining these things, but I think this is the general idea.)
When you combine that with the fact that most customers just don't care -- not until after they've got a lot invested in your site, at which point you're probably already a big success -- it's not surprising that the sites which evolved into big winners did so by keeping themselves fairly insular. Why spend money and time developing features that aren't important?
I'm not holding out hope for distributed versions of things like Facebook. Not yet. It's not unimaginable that the user community will eventually embrace such things, after they've had a lot more training in the downside of walled gardens, but I think we're way too early in the Web 2.0 adoption cycle for that. Right now I'd settle for a baby step: An export API that lets you dump all your data out of a service in some sort of JSON or XML format. That would be unusable for 99% of people, but at least it would let them back up their data and hand it to some third party that might be able to read it for them.
The problem is that nobody cares about Atom. Nobody cares about Media RSS or FriendFeed. The people that use services like that are in the slim minority, because it takes too long to understand, it takes too long to set up, and the biggest problem with open systems is that you don't retain tyrannical control over the information, which hurts your interface.
I've been kind of an outspoken critic of FriendFeed for this reason. It's kind of the culmination of these "open aggregation" sites. The problem is, when you put them all together, you wind up with a slush pile. You get nothing. A bunch of tech-obsessed people will find value; the average user will find nothing.
The theoretical solution is to reverse engineer every single site. That is to say: when I get tagged in a photo on Facebook, then show the tagged photos in the interface. If a friend uploads photos from Facebook and Flickr, then I see both of them - but not as a feed; I see it in a unified photo place. Furthermore, I need to be able to interact with those photos simultaneously. Say I'm looking at a photo gallery that contains photos both from Flickr and Facebook. The system would have to detect duplicates, unify the interface, and make it so that I can both tag the photo and add comments Flickr-style. It would also probably have to detect which comments are coming from what site, so that when I post my reply comments, it doesn't seem like I'm messed-up or out-of-sync - it could add two comment streams, but that's cluttered and messes up the interface.
Now imagine this but for every single site a person would use. Integrated Tweets, Tumblr posts, Facebook notes, possibly Last.FM songs. Most optimally, integrated private messaging between the sites. And the challenge is not getting all that data - as you say, it's publicly available. The problem is unifying that data and producing a comprehensive grasp of it for the user. Facebook can do that easily, because they have complete closed-door control of that information. They know exactly how you'll be commenting on things, they know the actionflow for uploading photos and tagging people in notes and posting items and whathave you. With any attempt to duplicate that openly, you lose that absolute knowledge unless you're brilliant and figure out the one unifying thread between every single site you connect to, and that's a pretty damn tall order.
The people who use FriendFeed as a social hub do not understand what people use social networking for. I've heard this argument a lot, that FriendFeed is a new-wave social hub. It's not. For most people, social is about personal connections. And that's not the same as "I know the person uploading this photo." It's "What was Mike up to last night?" "What is Christie saying about Valentine's Day?" "Who's writing what?" You could argue that FriendFeed shows all of these, in a way, but it misses this unification that I'd say is necessary. You can see photos and Twitter posts and blog entries, but you don't have this interface that says "This is Facebook, this is your portal, you can do anything you want to from this site, here's everything from your friends." In some ways it's more primitive than FriendFeed, and that's the point: people don't like complex systems. On FriendFeed there's a barrier between joining and doing. Not true on Facebook.
This is going to be the problem with OurDoings, too. I hope you guys get a good userbase, because you have a neat service, but at the same time you're largely using other service's photos, or that's the impression that I get. You're even importing from other photo aggregation sites, which I find silly. Now, the problem is that for most users, if they're putting their stuff online, they'll stay there. You use Twitter if you want to post short blurbs, you use Flickr to put up your photos, and then if you want to read other blurbs or see other photos, you stay there. I mean, stories get told on Facebook all the time. Not "I remember when this was taken: such and such a thing happened," but "Man oh man was I drunk." That's the story that people care about, and that people will reminisce about.
tl;dr, exporting data might sounds snazzy but most people just don't care enough.
Usability. Take a look at the Yahoo research on OpenID for an example of just one standard:
http://developer.yahoo.net/blog/archives/2008/10/open_id_res...
Asking why people are flooding Facebook is a great question, one that often doesn't get asked by geeks, probably because we assume the problem is simple human flocking behavior and people herding towards where the rest of their friends hang out. Nope, that's a necessary but insufficient condition - why did that crowd form in the first place? Just try using myspace for a day and it's painfully obvious.
Easier to make money with a walled garden than encouraging agriculture.
Some day a 21st Century "Diary of Ann Frank" might start with notes made on FB. Won't FB want to take advantage of print/advert/movie rights. $$$
"Power corrupts. Absolute power corrupts absolutely."
Sites really aren't evil. You've got to trust me on this, most sites aren't saying "How can we potentially steal a million dollars in the off-chance that one of our users posts up a movie script as a note rather than, say, sending it off to an executive?" Rather, they're saying, "Look, we're getting a lot of people who're being dicks about how we keep their data so that if they want to return their account magically comes back, and we think that it's better to keep that data than to remove it, because that way the people who delete accounts stupidly can come back and be happy they don't have to start again. At the same time, it's true that we're retaining data we shouldn't retain. So let's just add to our TOS that we're allowed to do that, since most people won't care, and since it means we can provide a useful service that'll help more people than it will piss off."
If that bothers you, bad news: Damn near every contract you sign nowadays contains that clause.
The better terms of service will permit you to at least cancel the contract after a term change if you do not consent. My CC contract had that, and I bet it's because it's required by law. But I haven't seen that many other places.
Now, this might sound like a horrible clause. And it probably is. On the other hand, contracts have a real need to be mutable, and for something like the Facebook TOS, you're basically dealing with millions of parties to the contract that simply can't be meaningfully "informed" about the nature of the contract or the nature of the changes. So, it's not just a problem with evil, greedy companies, it's also a problem with the fact that people are essentially incapable of understanding the TOS (due to either time or a literal inability to comprehend) and our legal system simply requires some sort of TOS to exist. The problem is complex and does not admit a simple solution.
Once you start to participate in something like Facebook, you really don't own the interactions you have via it. If anyone does "your network" does, but since that isn't an entity in any meaningful sense, Facebook the corporation must do.
Facebook has one of the easiest, and certainly cheapest, methods to inform users of new TOS agreements. In contrast to your CC company, they don't have to physically mail an updated document. If the TOS change, the user can simply be presented the new terms upon logging in (the next login after the change, or course), and they have the opportunity to agree to the new terms or not. This is how some sites have tackled this problem in my own experience.
If they wanted to be really fancy, they could even highlight where the changes are via change bars or text formatting. I wouldn't expect that, however.
Also, I'd imagine that as an engineer, "Create the most optimal terms-of-service change page" is a bit of a drag.
Also also, if Facebook isn't intending to do this maliciously, then if they highlight stuff like the new lines Mashable's jumping on, they'll get a lot of paranoid users for nothing. If they are doing it maliciously, it's in their best interests not to let users see what's happening.
I think they figure it's not worth bugging users about it.
Go read your credit card TOS. Go read your cell phone TOS. Go read the TOS of your favorite ten websites large enough to have a legal staff. Go read the EULA for your favorite ten pieces of commercial software. I'll wait. You'll find that clause is everywhere. I know this, because I have read them.
Not to mention I then go on to talk about why they have legitimate reasons to exist and shouldn't be considered simply as "power grabs"; I explicitly repudiate conspiracy theories, with reasons.
What more do you freaking want?
My specific point was to say that it would be extremely _easy_ for Facebook to have notified it's users of a change. My credit card, mobile phone and utility providers can all seem to do it via snail mail whenever there is a change. Whether they choose to do it for whatever reason is another discussion.
Seriously, you think a million bad cameraphone pictures of people getting drunk are worth $15Bn?
There is also the element of trust. What you describe is a sensible thing to protect against, but I have that gut feeling that the new policy is covering a new approach FB may take. Something feels wrong with having this wording of this policy, so I mistrust the intentions behind this new policy until evidence to the contrary emerges. I'm not holding my breath.
They wanted the press to keep a spot on Facebook!
Facebook makes it very easy to share photos. What if one could still share photos with friends on Facebook, but these photos were not uploaded to Facebook's server?
An idea could be to make an app that will upload deliberately crippled (low resolution, watermark), yet recognizable, pictures to FB, and a high-quality copy on Flickr, so tags work, but when you go the to picture, there's a note "I won't let FB own my stuff, see a high quality version of this pic on (Flickr-link) that I own. Posted by CoolPhotoApp."..
Never underestimate an angry customer...
Facebook is secure already, and nobody cares that it's closed. If it was a shitty site that was also closed, then maybe an "open social experience" would come in handy, but Facebook also happens to be the best-designed popular site that's ever been made, it runs oily-smooth, and it has five hundred bazillion users.
Never underestimate an angry customer...
Like the 10 million angry customers that protested when Facebook redesigned and lost its little guy logo? Or that protested that you can't change Facebook to other colors? Because I have friends that joined those groups, and they're still on Facebook. "Mildly irritated" customers can be underestimated, for the most part. They don't do anything.
Never underestimate an angry customer...