Twitter has enabled secure hypertext transfer protocol (HTTPS) for all its users by default, meaning that traffic on the micro-blogging site is now encrypted, providing better protection against man-in-the-middle attacks.
HTTPS keeps the session cookie encrypted throughout the log-in session, preventing the information from being intercepted. Twitter, which made opt-in HTTPS available to users for the first time last March, said that it is particularly important to use the encrypted protocol when accessing Twitter over an unsecured Internet connection, like a public Wi-Fi hotspot.
“HTTPS is one of the best ways to keep your account safe and it will only get better as we continue to improve HTTPS support on our web and mobile clients,” said Twitter in a blog post.
Users still have the option to turn off HTTPS through the Account Settings page.
The move was welcomed by security firm Sophos, which said that using public Wi-Fi hotspots to access Twitter without enabling HTTPS could allow a hacker to “sniff your session cookie”. This means they could post tweets as you, or read your private direct messages.
“Don't imagine that sniffing session cookies from unencrypted connections is rocket science,” said Graham Cluley, senior technology consultant at Sophos. “Tools such as Firesheep have made it child's play in the past for anyone to access the Twitter or Facebook account of someone close by if they haven't taken the right precautions.”
Cluley pointed out that actor Ashton Kutcher's Twitter account was hacked during the brainbox TED Conference last year. The hacker accessed Kutcher's account over an unencrypted Wi-Fi connection and posted pro-SSL graffiti in his name.
Google became one of the first major web communication companies to adopt HTTPS across its sites in January 2010. The Google Plus social network has there for had HTTPS turned on since launch.
In the case of Facebook, however, HTTPS is still disabled by default, despite giving users the option to enable it a year ago.
Meanwhile, research published by the Electronic Frontier Foundation last year showed that the SSL certificate system that underpins web security is far from trustworthy. Ultimately, this means that users may not be able fully trust HTTPS connections. However, until schemes like DNSSEC come online to prevent website spoofing, they have no choice but to do so.