In September 1987, a Beijing laboratory sent what became China's first email. The message, to a German university, read: “Across the Great Wall we can reach every corner of the world.” The development of internet infrastructure in the past few decades has enabled Chinese people to continue crossing the “Great Wall” and communicate with the rest of the world. But Chinese authorities soon threw up another wall to prevent the people from accessing information they deemed threatening to the Chinese Communist Party. In 1996, Beijing enacted a set of interim provisions for governing computer information, and in 1998, the Ministry of Public Security launched the Golden Shield project — a national filter that blocks politically sensitive content from entering the domestic network. This censorship tactic scheme has long been nicknamed the Great Firewall, and has undergone periodic upgrades since it was first introduced, given that people's efforts to cross the Great Firewall have been non-stop. Some describe the interplay between the Great Firewall and Chinese netizens as an ongoing “prison break”. Recently, Hong Kong-based investigative journalism platform the Initium reviewed the concurrent evolution of the Great Firewall and tools used to circumvent it. Below is a summary and partial translation of key points in the Chinese report. The first generation of the Golden Shield was a domestic filter that blocked specific domain names and server IP addresses: 「牆」就像一個邪惡的郵差，只要看到你的信封上寫上 Google 或 Facebook 等網站的地址，就會直接將信件丟棄。 The Great Firewall is like an evil mailman; whenever he sees Google or Facebook addressed on the envelope he will just toss the letter away. By using an overseas server, i.e. a proxy server, people could still go around the wall and visit any sites they wanted: 就好像你將一封希望寄給 Google 的信，先寄給一位認識好郵差的朋友，再由他幫你轉寄給 Google。 It is like you plan to send a letter to Google; you can first send the letter to a friend who knows a nice mailman and then the mailman can forward the letter to Google. The keyword-filtering system of Golden Shield was upgraded to detect the content of the websites that netizens visit, even if the internet connection is going through a proxy. If there is “sensitive content” communicated along the network connection, the Transmission Control Protocol (TCP) is reset: 這時候，「牆」就如一名資訊審查員，拆開你的信件，只要發現你提到敏感資訊，便將你的信丟棄。 In this case, the Great Firewall acts like an information inspector; it can tear open your letter and toss it away if sensitive content is found. Pathways for Censoring a Weibo Post. Graphic by Citizen Lab (CC BY 2.5) To prevent such inspection, Chinese netizens started using virtual private networks, or VPNs, to bypass the Great Firewall. Originally, VPNs were used by global corporation to protect business secrets; their internal communications traveled within a private network and were encrypted to make sure that no third party could detect the messages sent through the network. In other words, a VPN server outside China could safely transmit a Chinese netizen's visit requests to any third-party website. With support from the government, the developers of the Great Firewall finally managed to identify weaknesses in VPNs. They found that there are some obvious features of the commonly used VPN protocols, such as IPSec, L2TP/TPSec and PPTP, which often use specific ports. When processing the encrypted connection, it leaves a distinctive trace. Again, the Great Firewall was upgraded to detect such “irregular” connection traces: 這就好比「牆」雖然看不懂你的信件，但能從你信件內容中經常使用的筆跡和標點符號出現頻率，分析出你的真實意圖。 It's like even if the Great Firewall can’t read your letter, it can analyze the intention of the letter based on the handwriting and how often the punctuation appears. In 2011, the Great Firewall blocked two major VPN protocols, PPTP and L2TP, temporarily. But the economic damage of blocking protocols was huge. The Great Firewall since has been upgraded to slow down or reset individual VPN connections. VPN diagram, shared widely online. To solve the problem, open source developers from Github collectively developed a new tool, Shadowsocks. Same as a VPN, Shadowsocks is circumvention technology. It encrypts communication between the user and the website they wish to visit, but the connection is hard to detect for third party because Shadowsocks allows users to choose different encryption methods and appoint a random port. Moreover, Shadowsocks is an open-source project, so even if the original developer were to delete the codes published at GitHub under pressure from authorities, other developers could carry on the development and maintain variants such as ShadowsocksR and V2Ray.