This post was originally published in Arabic on Tarek Amr's blog Kelmeteen .
We have all heard about the Egyptian Ministry of Interior's tender to purchase programs and applications in order to monitor social networking websites and blogs and to survey public opinion. As a matter of fact, even if we disregard the above bid, a significant technological progress has been achieved in analyzing written texts – translation software and search engines are best proofs of the above. Accordingly, technology to monitor the Internet and analyse on-line content has become available to the public, as well as security agencies, marketing companies and the press alike.
However, the human mind didn't cease to exceed technology. Hereinafter let me enumerate some simple methods that Internet users can follow to aggravate the work of monitoring programs.
Being human beings, we can grasp the meaning of ironical expressions and understand these better than the best technological devices currently available. As for phrases like “This mobile phone is so great it has a battery that doesn't even last for five minutes,” any reader would understand that the above written statement is irony and that the word “great” stands for its opposite here.
Writing in metaphors and similes and so forth encumbers the task to be accomplished by surveillance programs. For example, you can use words like “hashtag” in contexts which only people aware of the situation you are referring to would understand, in this case referring to ElSisi.
The other option is writing Arabic in English transliteration. This method is known as “Francoarabic” writing. Actually, the technology is able to understand “Francoarabic”, Microsoft Maren  and Yamli  are proofs of this, albeit the accuracy of these technological achievements is not exquisite, and gets even worse when encountering different dialects, spelling mistakes and abbreviations.
To make spelling mistakes intentionally  is a means to distract the surveillance programs and text analyzers. It will render their work more difficult, but won't ruin it. Joining words to each other is another option to mislead these programs.
Division of your message into several tweets makes it impossible to grasp one section without the other.
With regards to the usage of photographs and videos, examining photographs and extracting texts from these is certainly more difficult than analyzing tweets and Facebook updates. In other words, it takes more effort for the computer to interpret some jokes on Egypt's Sarcasm Society  than posts published on Facebook.
Any other innovative means show how human capacities rise above the machine. As an example, you can ask your readers to read a sentence from right to left.
There are different variations of these tricks, one can read more and more about such ideas as confusion and diffusion , and about how can they be matched to build more powerful and complex encryption systems, or reading the tools that are used to deceive  intrusion detection systems of the Internet. As for the thought of dividing letters into numerous tweets, it derives from the idea of partition of Internet packages to encrypt intrusion detection systems. 
After all, technology is developing and it is able to keep up with these tricks. However, you have to know that any attempt to solve these dilemmas from the technological part is confronted with high expenses and with less accuracy and speed. The roof has to be dropped in messages that can be monitored and analyzed – just as the garden fences do not stop anyone from climbing in, they make the task more difficult and protect from many burglars by reducing their chances to break in.