Who’s That Woman in the Twitter Bot Profile?
After weeks of trying, I’d nearly found the real person behind a Twitter bot. It wasn’t the person who started the bot–chances are, that was just a computer program. Instead, I was hunting for the woman in the profile picture, the person whose identity had been stolen. The Internet is a big place; this isn’t easy to do. But I’d tracked the photo of a short-haired, punkish 20-something–used by @Arnitamj5, a bot calling itself Arnita Barayuga–to an abandoned MySpace profile of a Dallas woman named Elizabeth. She didn’t seem to have any other Internet presence, but I found one of her old MySpace friends on Facebook, figured out that he worked at a Dallas bike shop, and called it.
“So, listen,” I told him. “This will be the weirdest call you’ll get today.”
“Today?” he said.
“Probably all month.”
Then I explained: My goal was to draw a straight line from a Twitter bot to the real, live person whose face the bot had stolen. In the daily bot wars–the one Twitter fights every day, causing constant fluctuations in follower counts even as brands’ followers remain up to 48% bot–these women are the most visible and yet least acknowledged victims. And it’s almost always women, isn’t it? Bots are like a sorority party at 3 a.m.–a massive compilation of young, pretty faces who talk a lot of nonsense. But the women they portray are actual people, somewhere in this world. Who are they? And how were their photos dislodged from their original place?
This is a mostly pointless exercise, I knew: The story behind every photo would be different. And what would one of these women say–that she’s flattered to find her face spamming everyone on Twitter? Clearly, no. But it seemed worth doing, if only to tell one story, to have one answer. So I asked Elizabeth’s old friend: Did he still know her? He did, he said, though she’s since gotten married and changed her name. He promised to pass my message along. After four days of silence, though, I did more sleuthing and found her on Facebook under her married name. Then I emailed my plea: You’ve become a bot, Elizabeth. Can we talk about it?
Silence. Can’t say I blame her.
So I started over.
Bots are cheap. The company Buy Real Marketing will sell you 1,000 of them for $17, or 25,000 for $247–meaning the value of each is about a penny. And who’s buying them? Anyone. A brand’s social media manager will never admit to it, but chances are, gigantic companies have invested in this cheap form of image building. Why wouldn’t they?
Athletes definitely do it. A publicist for some major players–people at the top of their game–told me it’s common in his world. He once tried it himself, just to see what happens. He ordered the $17 package from Buy Real Marketing, via its website buytwitterfollowers.org. “They didn’t come in right away. I thought at first I’d been scammed,” he said. “But sure enough, within three days, they just poured in. It was exactly 1,000. To me, it shook the whole foundation. It made Twitter meaningless.”
The publicist gave me the names of a few people who also bought from Buy Real Marketing, and I dug into their followers. The bots were easy to spot–and these bots, no surprise, follow plenty of other celebrities and big brands. There’s no way to know if these were purchased follows or just pure coincidence, of course, but the list is wide-ranging. One bot from this batch followed Kelly Osbourne, former Formula 1 racer Tiago Monteiro, the Huffington Post, and an “Internet marketing consultant” named Trent Partridge, among 2,000 others.
If you click on a profile photo in Twitter, the photo will open in a tab of its own–and oftentimes will be larger, or more broadly cropped. I’d drag that onto my desktop, then run it through two image search engines: Tin Eye and Google Images. Each one scours the web for visual matches. After dozens of searches, a pattern emerged: Most bot photos had a long digital tail, having been posted on dozens of sketchy porn sites or blogs devoted to the barely legal. Occasionally, I’d be able to track a photo back to what seemed like an original source–like when a bot’s photo showed up alongside many others of the same woman, all posted to the fratboy site Barstool Sports. The site claimed her name is Aurora. But when I reached out, as was always the case, nobody cared to explain where the photos came from.
Then, finally, a reliable source: I tracked two bots back to the 2009 SUNshine Girls calendar, a lingerie showcase produced by the Toronto Sun. (I guess newspapers have to make money somehow.) The calendar only offered the models’ first names, and the paper’s photo editor wouldn’t connect me with them. But after a little Internet stalking–this is how reporting works, people!–I found a connection.
One of the bots, @Karriehga, which went by the name Maralyn Estes, showed a photo of a beautiful blond with dark eyes and hair poofed back like a Kentucky prom queen. This was Amanda the SUNshine Girl. And some clever Googling led me to a blog that included her full name. That allowed me to find her Facebook page, which didn’t list an email address, but did show that she recently clicked “like” on an events planning company. I figured that’s where she now works, so I called. Amanda, it turns out, was on maternity leave. “You can leave a message, and she’ll call you back in a few weeks,” her boss Darlene told me.
I didn’t have time for that, I said. Darlene asked why. So I began to explain.
“Wait, wait, Amanda was a SUNshine Girl?” Darlene yelped, and started laughing. “I didn’t know that!”
Oh, boy. Sorry Amanda.
But after that, Darlene said she’d help me get in touch. I hung up, relieved. Then I looked at my computer screen, which still had @Karriehga up. It had just tweeted something, as these things regularly do. Usually they’re just snippets of text yanked from websites, just something to keep their profiles active.
This time, though, the tweet seemed like a warning: “Don’t spend time beating on a wall, hoping to transform it into a door.”
In the meantime, I contacted Buy Real Marketing. I expected this to be equally difficult, given the sketchy nature of what a company like this does. But its work is perfectly legal–in the name of viral marketing, big brands have done far worse–and so all I had to do was call a toll-free number and hit a few buttons. Then I reached a tired-sounding woman named Judy, who spoke to me on a scratchy phone connection. I identified myself as a reporter and asked to interview someone, but she volunteered herself for the task. So I asked her: Judy, who are the faces on your bots?
“These are not bots that we have on Twitter,” she said. “These are real people.”
Me: “So there are no bots?”
Judy: “No bots. Not even spam.”
Me: “I mean, I see a lot of what certainly look and function like bots. But they’re not bots?”
Judy: “They are real people. They just log in, like, once a month so they are considered active.”
Me: “I see. Are the profile faces them?”
Judy: “Yes, exactly.”
Me: “So, the pictures of the people who are on a…”
Judy: “Some of them are. We can’t really control them. These are real people, and they have their choice of freedom on what picture they place there.”
And that’s all she was giving me.
Amanda’s email showed up the next morning: “I heard you contacted my employer Darlene yesterday and would like to talk to me. I’m interested in knowing what this is all about.”
She gave me her number. I called immediately.
Amanda lives in Bowmanville, Ontario, just outside of Toronto. Her husband is a police officer there. The night before, as they puzzled over Darlene’s message to call me, her husband began telling Amanda about all the facial recognition software that’s becoming available to law enforcement. It freaked her out.
Truth be told, she’s been trying to distance herself from the SUNshine Girl thing. (We’re helping out by not publishing her last name. That’s one less Google result to worry about.) It’s not that she’s embarrassed; back in the day, she even did live promotions for the calendar. But these days she has to worry about what employers think. Darlene doesn’t care–thankfully–but Amanda used to work for the government. She figured it was best not to flaunt her past.
And now, this. In the past day, I’d found five other bots using the same photo of her.
“It’s kinda of creepy, to be honest with you. The whole thing,” she says. She’s on Twitter but rarely uses it, and had never heard of bots. “I’d like to find the source and tell them to stop using my photo, you know? Because you never know who’s going to see it, and I don’t have control over what someone’s saying. That could ruin who-knows-what.”
I told Amanda that she could report the bot as spam, and hope for the best. She said she’d do that, but that she likely wouldn’t do any more. After all, what’s there to do–sue? Sue who? She doesn’t even own the photo; it’s the Toronto Sun‘s property. But she appreciated knowing. She thanked me.
Four days later, Amanda’s bot @Karriehga was still live. It tweeted, “Let’s commit the perfect crime… I’ll steal your heart, and you steal mine.”
To say nothing of a face.
Via Fast Company: http://www.fastcompany.com
One in five Americans read an ebook in the past year, according to a 3,000-person 
In Apple’s imagined 3-D desktop, the user’s files are organized in piles in 3-D space like in a real office. Some may be pinned to the wall, some are alone, others are in stacks (and there’s no reason that expert users may choose to keep things “real,” as in this virtual 3-D environment you could stack things on the ceiling, or leave them floating in the air). Though this sounds like a simple re-organization of the way we all work with files in folders now, the actual innovation is that instead of having to do something arcane like double-tapping on a file to use it, Apple’s guessing it may be more intuitive to gesture at the file. Something like zooming it into the center of the field of view to work on it, then pushing it away while you consult another document. Graphical transitions would reposition and re-size the image representing the flle in real-time, making interacting with this kind of desktop feel even more fluid and natural than current tricks like symbolically dropping an unwanted file in the trash.
BARCELONA: Adobe Photoshop Express is a great, free and highly usable iOS app for quickly fixing up your digital photos, but it pales in comparison to the desktop version of Photoshop. Last year Adobe introduced a solution: Photoshop Touch, but chose to bring it to the 
The 
Kindle Cloud Reader, Amazon’s HTML5 app that lets you read Kindle books both online and offline from your browser, is now available for Firefox.
Apple is bringing speech recognition to the masses with its new iPhone 4S, equipped with an intelligent assistant named Siri. It’s a major differentiator for the new 
