Project description

La Cuba que viene (The coming Cuba) is a special transmedia interactive product that offers explanations, contexts, comparisons and other resources to understand the constitutional reform in Cuba between July 2018 and February 2019.
At the same time, it is the news coverage of the entire process by
The reform process had several phases. A first discussion on the part of the national assembly. Then a process of national debate with more than 100 thousand public assemblies. Then another debate in the National Assembly and finally a referendum on February 24, in which the new text was approved.
This multimedia special is also divided into phases, which correspond to the different parts of the reform.
The journalistic material was available since September 2018 at In each phase there are key questions answered from interactive components, texts and videos.
It is a joint work between journalists, jurists, and political analysts, and for its implementation participated programmers and designers. It includes resources to compare several versions of the same text, and helps to understand a technical and complicated language for the common user who was the beneficiary or damaged with the new constitutional text.
Our objective was to inform the debate of the citizens that had to vote so that the democratic exercise was conscious. We pointed out alerts, we identified positive and negative elements and we made meticulous analysis of the topics; but with the intention that the users would reach their own conclusions.
In addition, we offered a useful product for key audiences such as academics, influencers, specialists, deputies and other media.
The total coverage includes 11 interactive components, 23 texts by 11 authors, 6 short videos for social media and two videos summaries of public presentations. All organized and accessible from the same page and from our mobile application.

What makes this project innovative?

The election of the editorial team to make the coverage of the reform opted for the interactive visualization of data with novel tools for the Cuban context. It was the most important political process of the year for Cuba and there was an excess of traditional information, with an enormous amount of analysis in opinion articles, which produced oversaturation in the users. As the Cuban political power has control of the most influential media, it imposed a totally biased coverage. We seek to reach an audience with unbiased arguments and an original approach, which would help them analyze a topic of which there was widespread ignorance. The result was the call for a joint work of journalists, jurists, and political analysts to explain technical issues in an understandable and attractive manner, even playful, for the audience. For this reason, we compile in a single place a series of texts, videos, analysis and interactive components with a formal unit that help to keep track of the whole reform. This material is available in offline version and in the mobile application of so that thousands of users without Internet all the time in Cuba could also consult it in different contexts.

What was the impact of your project? How did you measure it?

During the first phase of this coverage, we designed the products so that people could use the arguments offered in the multimedia special as a tool for the debate that took place in their neighborhoods and work centers. Also, in the following phases, resources were incorporated so that the deputies could prepare themselves better for the second debate in the National Assembly, as well as other academics and interested parties in general. Several of the issues questioned by our product had indirect answers from members of the drafting committee of the new constitution. Each of the components of the special multimedia was distributed by social media, so that users could react and give their opinions on the topics. In addition, two debates and public presentations (link) were held with the specialists. Those events were filmed and distributed as videos for those who could not go to the physical space. In the two public presentations, more than 100 people participated, who interacted with the specialists. As of April 2019, more than 5585 users had accessed the text in its online version and more than 3800 had consulted it through the semi-offline mobile application. Although due to the characteristics of the Cuban technological context, there is a whole audience that we can not count. This is called "Package of the Week", a kind of offline internet that works distributing content by USB drives and hard drives. We distribute several of the products through that weekly package and by email (to more than 3900 subscribers). The videos that make up this special and that we publish on Youtube and Facebook account for more than 130,000 views. In addition, several media have referenced our work, considering it a novel and useful initiative: These results, although modest, are not insignificant in our national conditions.

Source and methodology

For this special a specific database was built. For the first phase, the text of the new constitution project was the most important content to build the database. In addition, we had as reference the database of other projects such as, and The database is generated from the paragraphs of the text and classified according to the taxonomy of the Constitution: Preamble, Titles, Sections, Chapters, Articles and Final Section. For the first component (comparison): We searched which paragraphs of the project had more coincidence with those of the current Constitution based on the Dice coefficient. Then we did a manual review to verify that the compared elements were semantically similar and we discarded several comparisons that had different meanings. In addition, new relationships were found, because although the algorithm did not include them, their meaning was similar. Finally we relate each paragraph of the project with one or several paragraphs of the current Constitution and find the differences between those contents to know which were eliminated, added or stayed the same. Then we calculate among the Constitutions which paragraphs of the project were modifications (textual and semantic). The paragraphs of the project for which we did not find a match with the previous text were considered new contents and the elements that were textually equal in both constitutions were classified as such. The eliminated ones are those contents of the old constitution that were not used in the new text. This work was carried out progressively from relationships with tools and with specialists. From those relationships, the percentage of text corresponding to each category (Equal, New, Deleted or Modified) were calculated. Then we added notes with comments from specialists. For components 2 (rights), 3 (cards) and 4 (search engine): Each article of the Constitution project was manually classified according to the rights they contained, the references to other laws they made and the topics they addressed, and all that information was added to the database. In the case of cards, the current laws were identified and classified according to their possible transformation or not as a result of the entry into force of the new constitution. For the search engine, the manual classification allows the search to respond to semantic and non-textual criteria. The fifth component (word cloud): We mapped the main debates and positions on the new constitution project published in 15 unofficial Cuban media and blogs representative of different ideological tendencies. The main objective was to identify the issues related to the constitutional text and the position that the authors expressed their articles. Each reference was assigned to a topic category and was classified according to the assessment issued by the authors with respect to the subject they analyzed. The classifications were: Positive (without reservations), Positive predominance (according to some suggestions or minor criticisms), Neutral (does not make evaluative judgments or presented in a balanced way arguments for and against), Negative dominance (critics predominate to the content although its relevance is recognized), Negative (Disagreement) and Themes that do not appear (suggestions for inclusion of topics not included in the project). That information was represented in a cloud of words that allows you to reach each text published from the debate and filter it through the media outlet that published it. The sixth component (references of other constitutions): Exemplifies 46 contents of other constitutions that could be considered for the Cuban case. The second phase, "road to popular endorsement", was a moment of revision of the popular proposals and the presentation of a new project before the National Assembly. In this case, the comparison of the texts, now triple, is repeated between the current Constitution, the first project and the second project, using the same techniques. In addition to the comparisons, two new components were added. The first was built based on a thematic classification of the contents of the new constitution, which allowed identifying which of the issues raised by the population were taken into account or not. The second component of this phase compares the amount of rights that exist in the constitutions of other countries with the rights that appear in the new Cuban Constitution. In the latter, the source of data was obtained from the ranking of rights provided by The third phase, the end of the process, includes the final results of the referendum offered by the electoral authorities and the legislative and electoral route established by the new Constitution from 2019 onwards.

Technologies Used

Each component of the special of the constitution was developed in two stages, a stage of capture and analysis of the data and another one of presentation of the content on the web. The capture and analysis of data was programmed in JavaScript in the NodeJS environment. The main libraries used for the comparisons between the constitutions were the following: diff-match-patch: to obtain the word differences between the paragraphs of each text. stringSimilarity: to calculate the similarity between strings of texts using the coefficient of Dice. natural: to use the technique of stemming on the texts and optimize the evaluation of similarity using only the roots of the words. In the rest of the components the capture of the data consisted of extracting information from CSV files, and transforming them into JSON in a structure that was comprehensible to be used on the web. The presentation of the contents was built on the VueJS framework. The d3.js and chart.js libraries were used for the construction of the graphs. The solution is deployed with Github Pages at the address

Project members

José Jasán Nieves, Abraham Calás, Wendy Valladares and Eloy Viera



Additional links


Click Follow to keep up with the evolution of this project:
you will receive a notification anytime the project leader updates the project page.