It is an improved captcha version. In principle I used two test texts, one is the traditional one, distorted so that it can only be recognized by one person, but the second is not a test but a collaboration, a word that OCR book digitization services do not recognize and offer it for people who use a captcha to decipher them; for if the first is correct it can be assumed that the latter is also correct. This is not really a validation method, it only took advantage of one to collaborate with the creation of digital books, so later it did include other interactions such as following the movement of a pointer or the historical behavior of a user on the network.
* Only one "like" per meaning and day, the more "likes" the meaning will appear higher in the list
