Breaking into someone's e-mail can be child's play for a determined hacker, as Twitter employees have learned the hard way — again.
For the third time this year, the San Francisco-based company was the victim of a security breach stemming from a simple end-run around its defenses: A hacker guessed the password for an employee's personal e-mail account and worked from there to steal confidential company documents.
The techniques used by the attackers highlight the dangers of a broader trend promoted by Google and others toward storing more data online, instead of on computers under your control.
The shift toward doing more over the Web — a practice known as "cloud computing" — means that mistakes employees make in their private lives can do serious damage to their employers, because a single e-mail account can tie the two worlds together.
Stealing the password for someone's Gmail account, for example, not only gives the hacker access to that person's personal e-mail, but also to any other Google applications they might use for work, like those used to create spreadsheets or presentations.
That's apparently what happened to Twitter, which shares confidential data within the company through the Google Apps package that incorporates e-mail, word processing, spreadsheet, calendar and other Google services for $50 per user per year.
Co-founder Biz Stone wrote in a blog posting Wednesday that the personal e-mail of an unnamed Twitter administrative employee was hacked about a month ago, and through that the attacker got access to the employee's Google Apps account.
Separately, the wife of co-founder Evan Williams also had her personal e-mail hacked around the same time, Stone wrote. Through that, the attacker got access to Williams' personal Amazon and PayPal accounts.
Stone said the attacks are "about Twitter being in enough of a spotlight that folks who work here can become targets."
Some of the material the hacker posted online from the Google Apps documents was more embarrassing than damaging, like floor plans for new office space and a pitch for a TV show about the increasingly popular online messaging service.
Twitter says only one user account was potentially compromised because a screenshot of the account was included among the stolen documents. The value in hijacking a user's account is limited, as those attacks are mainly used to post fake messages and try to trick the victim's friends into clicking on links that will infect their computers.
Sensitive Twitter documents were filched, though.
The hacker claims to have employee salaries and credit card numbers, resumes from job applicants, internal meeting reports and growth projections.
Stone said the stolen documents "are not polished or ready for prime time and they're certainly not revealing some big, secret plan for taking over the world," but said they are sensitive enough that their public release could jeopardize relationships with Twitter's partners.
What the attacks on Twitter show is that websites don't need to get compromised in the traditional sense to put its users and employees at risk.
Hackers don't need to find a vulnerability in the site itself, or plant a virus on an employee's computer, to sneak inside.
The easier approach is much more low-tech: All they need to find is an employee who uses weak passwords for his or her e-mail accounts, or has security questions that are easy to answer with a little information about the person.