Okay, what if we say that safeguards Facebook put in place designed to detect API abuse were tripped, but after manual review Facebook permitted it to continue with the stipulation that it had to stop after the election? I don’t really see much distinction there but I will concede that point.
Once permission was granted, the campaign had access to millions of names and faces they could match against their lists of persuadable voters, potential donors, unregistered voters and so on. “It would take us 5 to 10 seconds to get a friends list and match it against the voter list,” St. Clair said. They found matches about 50 percent of the time, he said. But the campaign’s ultimate goal was to deputize the closest Obama-supporting friends of voters who were wavering in their affections for the president. “We would grab the top 50 you were most active with and then crawl their wall” to figure out who were most likely to be their real-life friends, not just casual Facebook acquaintances. St. Clair, a former high-school marching-band member who now wears a leather Diesel jacket, explained: “We asked to see photos but really we were looking for who were tagged in photos with you, which was a really great way to dredge up old college friends — and ex-girlfriends,” he said.
The campaign’s exhaustive use of Facebook triggered the site’s internal safeguards. “It was more like we blew through an alarm that their engineers hadn’t planned for or knew about,” said St. Clair, who had been working at a small firm in Chicago and joined the campaign at the suggestion of a friend. “They’d sigh and say, ‘You can do this as long as you stop doing it on Nov. 7.’ ”
For me this is deeply disturbing. Let’s not forget this article is from 2013. It seems the quote from the NPR article is in conflict with the statement here.
NPR Article:
One of the things, I think, driving people’s outrage is the indirect consent. A friend that installed this app agreed to stuff. And now I’ve been matched to my voterfile records. I didn’t directly agree consent to that. Not only that but they retained the data.
I have worked on this system on the api for many years, since its inception. I may have contextualized some stuff in the article with my own experiences. The graph API changed… I have no doubt this was a contributing factor. This is a quick and dirty mini-timeline:
2010 – graph api includes everything – even private messages on request/approval. Users “gave consent” but I’d argue they didn’t know what they were consenting to. The features were not secret. Prior to then, developers were not allowed to cache/store data for more than 24 hours. They did away with that in 2010. Facebook gave you a globally unique number for every person’s data you got, and that could be related across many apps if you had access to the data.
In late 2014 they changed their mind, and it to effect in April 2015. One change was changing the facebook user ID to be app specific. Zuck went on record as wanting to give users more control of their private info. (Whether that’s true… or the fear of FTC involvement or some major blow up…)
If you really want to split hairs, I think that the DNC had/has more access to more data from that era than CA had through their graph API access. That’s what they’re talking about when they say license violation because you can’t retain the data. But if they transformed the data in some way, they may be in the clear. I’m not sure and we need more investigative journalists to suss that part out of it. But to me, on paper, it looks like the DNC had a hell of a lot more access under the rules of the time than CA ever did.
Did you see what % of the voter file they were able to match to with indirect consent of your friends? Scary stuff.
Wow, this is a misleading statement. “The end of the use.” No, they saved the data and have been using it ever since at least according to the Obama campaign link article we used in our one tab:
“Where this gets complicated is, that freaked Facebook out, right? So they shut off the feature,” she added. “Well, the Republicans never built an app to do that. So the data is out there, you can’t take it back, right? So Democrats have this information.”
…
“They came to office in the days following election recruiting & were very candid that they allowed us to do things they wouldn’t have allowed someone else to do because they were on our side,” Davidsen tweeted._
That this data is now part of the records at the DNC has also been carried in several other news stories.
One of the more disturbing takeaways for me from the article was that the influence technology was up for sale for Caesars Casino. You should take a look at The Victory Lab book…
I suppose this is true. Though surely you would concede that there is an ethical problem when this is hidden far, far away from the more obvious-ish “deactivate account” options on your profile? Having to go to the help center to get to the link is kinda dishonest I think. They also don’t go out of their way to tell you if you do anything with the account it will be reactivated. I wouldn’t be the least bit surprised if that link changes to something else in a month or two.
Yes, yes it is. Very fine indeed. Do you think if Reddit was doing something that negatively impacted Code Nast’s bottom line that Advance would not intervene? Like you say they may be riding the wave of clickbait but our point was that if Reddit is for-profit it is easy to imagine they are just as happy as facebook to do whatever the heck they want with yo ur data.
The Channel 4 (thanks for the correction on that) video is mind blowing. It’s like something out of a B-movie where Bruce Willis is meeting with the bad guy before deciding for himself the only way to really take out the bad guy is to go mano-a-mano. To be clear I think the efficacy of their product is overstated and unlikely to be as effective as they tout. Sounds like sales talk to try to get a sale to me. Does that make it right? No. Is everyone involved here ethically bankrupt? Yeah, you bet.
The point here in all of this is not “I told you so” but to actually see past “Facebook.” Facebook has always been fairly transparent this is what they’re doing and this is what it is designed to do. If you have a problem with CA+Facebook, you owe it to yourself to step back and look at the entire ecosystem of ALL companies that operate this way and start to think about how we can prevent this kind of thing from happening.
What are the steps to nuclear desescalation here.
What disturbs me is that you have people like Zuck, Schmidt, the Mercers, the Kochs, etc so presumptuous as to “think for me” – every one of these folks are in a race to figure out who can most effectively target (manipulate? is manipulate too strong a word?) that segment of voting public everyone loses. It sure seems like folly to try and split hairs over which side was “weaponizing” the data for strictly innocent defensive or evil offensive purposes.
…oh god I’ve become one of those long news reply people noooooooooooooooooooooo
I would also add:
Civis Media Optimizer, as the software is called, is fed a list of kinds of audiences the advertiser wants to reach, which is then combined with data on consumer viewing habits from Nielsen’s panels and settop box data from Rentrak, the media measurement company bought last week by online measurement firm comScore. The software uses artificial intelligence formulas to help the advertiser reach its target audiences on the TV shows that will get the most reach and desired frequency of ad viewing for their budget.
Somewhere there is a quote by this Wagner guy that iirc says that social media was one of the sources they fed in to their system knowing what tv spots to buy because they couldn’t outspend the RNC. For super targeted TV ads. Because nothing appeals to emotion more effectively than a tv ad… this is the same guy that put the 15 million “likely to be influenced” people list together for Obama during that campaign in the article I linked for their tv ads…
(Even tho he’s not dead. But still)
