It’s been known for some time that very few women subscribed to adultery website Ashley Madison, which would suggest there wasn’t a great deal of actual adultery going on, despite the best efforts of men fleeced for hundreds of dollars in subscription fees.
The Impact Team hackers who stole and released Ashley Madison’s client list have said all along they’re angry at the company’s deceptive business practices.
A deep dive into the subscriber data by Annalee Newitz of Gizmodo suggests there were even fewer female subscribers than previously believed – in fact, she could only find evidence that about 12,000 out of 37 million total profiles belonged to “real women who were active users of Ashley Madison.”
Newitz’ detective work is a great example of how data mining techniques can uncover patterns in large volumes of information, with a relatively modest investment of human labor augmented by the enormous power of modern computers. Imagine this story was playing out 40 years ago with a mail-order dating company for married people who wanted to cheat, and the records were paper documents stuffed into folders. Of course, in that case, they would have been a lot harder to steal. There are lessons to be learned here about the fragile nature of privacy across the online world.
Newitz set out to establish whether the widely cited, and already rather pathetic, figure of 5 percent female subscribers kicked around in media reports corresponded to real people who were actively using the site. The key insight here is active users.
A subscription database entry with the female gender selected didn’t necessarily mean the individual described was a customer hitting the site and using its tools to communicate with adultery-minded men. Some of them could be women who created an account on a lark or a dare, and never went back. Newitz noted media reports that many of the others were mass-created by Ashley Madison employees to make it appear their site had far more female users than it really did – in fact, one former employee complained about being subjected to sweatshop-labor demands by management to crank out a thousand fake female profiles in Portuguese, to attract Brazilian customers, in just three months.
The payment transaction file included by the hackers wasn’t useful for easily weeding out real female subscribers, because women were given access to the site for free, but it turned out there were other database fields that described each subscriber’s use of the Ashley Madison chat and mail systems. Like most dating websites, tools were provided for subscribers to talk without giving away personal information, such as their email addresses. The vast majority of purported female subscribers never used these tools, or used only the bulk-reply feature of the email system to dismiss inquisitive messages from male subscribers.
Newitz also spotted an interesting fact about the Internet addresses associated with a large majority of the female accounts: some 80,000 of them were branded with the internal or “loopback” IP address common to all networks. In other words, they were created within the Ashley Madison network, almost certainly by employees of the company.
“This isn’t a debauched wonderland of men cheating on their wives. It isn’t even a sadscape of 31 million men competing to attract those 5.5 million women in the database. Instead, it’s like a science fictional future where every woman on Earth is dead, and some Dilbert-like engineer has replaced them with badly-designed robots,” quipped Newitz, before reaching the melancholy conclusion that the single most reliable data point to establish the high-water mark of real, active female users is that 12,108 of them used the “paid delete” feature and forked over twenty bucks to have their accounts permanently erased… which, of course, didn’t actually happen. They’re still in the database, with most of their information intact. The only thing their $20 bought is that one of the fields was changed to say “<paid_delete>”.
Newitz makes due allowance for the possibility that some of the fake-looking accounts could have been real women who checked into the site once on a lark, then never returned – which would be less of a scam than phony accounts created by company employees to pad out their subscriber list, but still isn’t much use to men looking for an opportunity to cheat on their wives. (As for those fake accounts, the Gizmodo piece reminds us that Ashley Madison covered itself with a finely-worded disclaimer about how “some” of the profiles were created “for amusement only,” which probably isn’t amusing to the men who shelled out hundreds of dollars to use the service, and now face personal and professional ruin due to the exposure of the subscriber data.)
One possibility not considered in the article: some of the men using Ashley Madison were gay or bisexual, and might have been content to communicate with like-minded men. Whatever other sociological conclusions can be drawn from this debacle, it stands as a dismaying example of dishonesty and foolishness… and a warning to people of far more decent inclinations that data is forever, and it tells tales.
There isn’t much evidence to date of a massive movement away from the Internet by people worried about their privacy in the face of headline-grabbing hacker stories – the Internet is just too damn useful for people to be frightened away from it, especially the younger generation that grew up with its power at their fingertips. Perhaps they’ll just get used to living transparent lives, clinging to the so-far vain hope that security forces will win their decades-long battle with hackers.