An Exploratory Exercise on Journalistic Initiatives on Medium

By Maria Clara Aquino Bittencourt

This work presents an initial exploratory exercise of a research project which investigates the transformations and continuities of digital journalism through an in-depth study on journalistic initiatives on the Medium platform. This study identifies the use of Medium as a platform for independent journalists and publications and, at the same time, its ability to curate content highlights it as one of the practices that redefines the role of the journalist in the digital environment.


After 20 years of consolidation in digital journalism (Salaverria, 2015), the research field on this topic has grown through the development of digital communication technologies. This scenario is discussed by Oliveira and Henn (2014) as the interaction between new agents through a network in which a systemic crisis in the journalistic practice has arisen. The impacts on the autonomy and identity of journalism trigger the systemic characteristic of this crisis, which, according to the authors stem from transformations which “even though are dominated by a centripetal force which act in the maintenance of certain controls, would alter the identity physiognomy of journalism as a professional and academic field” (2014, p. 52). The problem revolves around a moment in which journalism maintains the role of protagonist in the public agenda and, at the same time, starts to dispute this position with other systems. Oliveira and Hen (2014) indicate how the meanings brought by journalism compete with multiple meanings circulating in social networks, produced by different subjects.

It is preferable to start with the idea of transition rather than take the issue of crisis as a basis. Neither the crisis nor the difficulties faced by journalists in their daily activities are denied, neither the issues of the business models are excluded (Anderson, Bell & Shirky, 2012; Costa, 2014; Saad Corrêa, 2011). It is understood that assuming a position supported by arguments that these are opportunities rather than threats – it is more productive behavior when having the objective to think about digital journalism as new theoretical and methodological paths for research are identified.

In a 2005 article that evaluates the idea of “Web 2.0” five years after the coining of this term, O’Reilly and Batelle (2009) recall that one of the main insights was the understanding of the web as a platform, which meant more than just the offering of applications throughout the Internet, but that such applications would improve through people’s usage, in a way in which it would be possible to learn from this usage and the contributions to perfect the technology offered.

The research question is driven by a sense of discomfort felt by those who deal with specific issues of this profession as well as by those who consider journalism a science and see in the crisis the opportunities to rethink the future in this area. Thus, the question is: how structured are the forms of production and the dynamics of circulation in the journalistic use of the Medium platform?

The selection of Medium for this study was made due to the understanding that the features available and the content management structure redefine old online publishing practices, grouping in a single platform the characteristics of different computer-mediated communication tools. Furthermore, by integrating elements of websites and social networks (Boyd & Elison, 2007; Recuero, 2009) they induce a diversity of agents involved in content production and circulation. Launched in 2012, Medium is a creation of Evan Williams and Biz Stone, the duo who created Twitter. Williams worked at Pyra, which in 1999 launched Blogger, a blog creation and management platform, which effectively made it possible that anyone could share content on the Internet. Brazilians produce around 3,500 texts in Portuguese per week, thus, Portuguese is the second language among the nine operating on Medium around the world (English, Portuguese, Spanish, French, German, Italian, Turkish, Russian, and Japanese). The text response and recommendation structure opens possibilities for participation and interaction that increase circulation, which also depends on the use of tags by users who publish content on the platform. The Medium user has a profile, through which he or she can create publications as digital magazines and recommend texts by other authors and publishers. There are already 35 million users per month around the world on the platform.

In a publication from August 2012 on Medium, Biz Stone recalls how Blogger helped democratize content publishing on the Internet, but acknowledges that these are new times and that is why Medium goes beyond the possibility of entering content on the web.

Thirteen years ago we helped democratize publishing with a web-native approach called blogging. That was a long time ago and everything is different now—social networks, mobile devices, you name it. We felt compelled to build a content network for the technology age we’re living in now, and we have a vision for what publishing should be. (Stone, 2012, para. 2)

At that time, when Medium was open to public, it was based on usage patterns, so over time, the tool has been modified from the uses and appropriations made by users. Evan Williams, also on Medium, frequently presents new features, showing how the platform goes to the next level. More than a space for exchange, in which the publishers share stories and find audiences, Medium has been transforming people’s behavior towards the content, changing practices that stimulate the production and the circulation of knowledge.

The experimental nature of different initiatives, which appropriate platforms such as Medium, for example, characterizes the current context of digital journalism. The way through which Medium operates—correlating opportunities of collaborative content production with circulation structures, including elements of convergence (Aquino Bittencourt, 2012; Jenkins, 2006) and spreadability (Jenkins, Green & Ford, 2013), gathering texts that, through recommendations, have increased viewing of such content—justifies choosing the platform as the focus of this analysis to investigate the issues and goals in this project.

There are two directions in this investigation on Medium: the focus on the initiatives that use the tool aiming at reinventing journalism in terms of content production and circulation and the analysis of the changes and continuities in digital journalism which are expressed through the potentiality and appropriations of the tool by multiple initiatives and agents involved in the production and consumption of the content published. The first direction implies the mapping of journalistic initiatives on Medium, which propose transformations in the forms of production and in the dynamics of circulation of content. The second generates an analysis of these initiatives, in an attempt to establish observation criteria on elements related to the processes of production and circulation of content. The methodological procedures are described in detail next.


Writing, recommending, replying and leaving notes to writers are the main media opportunities offered by Medium. The interface is light, based on WYSIWYG (What You See Is What You Get), with few text formatting options, though it is easy to insert images, songs and videos through hypertext. Medium bought and this allows embedding content from an extensive list of other platforms such as YouTube, Flickr, SoundCloud, some news websites and agencies. Medium’s API (Application Programing Interface) was opened in October 2015, allowing for sharing stories from other platforms, such as WordPress, among other possibilities. Besides the open API, Medium also allows domain customization.

Unlike YouTube, where video is the main form of communication, or Instagram, where photographs (and, now, videos) are the main form, text is still the main form on Medium. A feature in Medium that reinforces the relevance of text in the platform is letters. They are offered only for publications, not for profiles and are, basically, a newsletter. They are sent via email to the followers of the publication. Medium itself sends a daily letter to users, the Medium Daily Digest, which gathers the most popular stories according to the tags, people and publications followed by the users. The letter is customized and is not the same for every user.

Collaboration is one of the features that shape how the platform works, and in the case of certain publications, it is what defines a model based on curation. Before a text is published on Medium, it appears on other websites via sharing; it is immediately viewed by the users on the platform where it circulates first. Unlike publishing content on a website or blog, for example, and maintaining and generating visibility to the page, which is, first, disconnected from the rest of the web, publishing on Medium is part of a web of connections based on individuals following each other, besides following publications. These connections create interaction regarding the stories published, in profiles and publications alike. The text recommendation, activated via clicking on a heart icon at the end of the stories, is the currency for trading internal visibility. The conversations about stories are the motivation for the publication of new ones. That way, the spreading (Jenkins, Green & Ford, 2013) of stories happens first on Medium, and can be expanded to other platforms, like Twitter and Facebook, but does not depend on those websites. Medium is built as a web. The connections between individual profiles and publications allow the narrowing of relations around the texts, which can be edited on the platform itself, through the interactions between writers and editors. The mentioning of profiles and publications ensure the strengthening of these connections and the expansion of the conversations by linking multiple authorships to the texts of the stories.

In this research, the focus is first on the publications, which have their own stories and are written by their exclusive editors and writers, or gather stories from writers around Medium—which can be seen as a curation process (Corrêa & Bertocchi, 2012; Rosenbaum, 2011) made by the editors in search of stories to compose their publications. There are cases in which publications act both ways, uniting their own content with curation.

The organization and searching of texts on Medium is made using tags. Each text can be marked with up to five tags and the search for content can also be done through them. Thus, the list of stories appearing to each Medium user is assembled according to the tags he or she chooses to see, the profiles and publications followed and the recommendations made by the ones being followed. When someone chooses to follow a profile or a publication on Medium, he or she will automatically receive all texts published by this profile or publication, either on the web or via application. The combination of an editorial (personal) and algorithmic (via tags) curation made by Medium does not limit the content viewing as on Facebook, for example. This means that a text published by a certain person or publication will be fully distributed to all followers. However, it is clear that the more users or publications one follows, the more difficult it is to see all the content published on the platform.

Each text published on Medium comes with information on the amount of reading time, just below its title, as shown in Figure 1:

Figure 1 Bittencourt

The TTR—Total Time Reading—is used by Medium as their main measurement system. The number of viewpoints is considered more relevant and is expressed via replies to the texts or by the publication of other stories. More than knowing how many people have read a story on Medium, what matters is knowing how long each person spent reading a text and how the sidebar was scrolled—irrelevant measures for short news produced to meet a demand for frequent updates, as in many digital journalistic vehicles. The reading percentage is a result of this calculation, and the more viewers a text has, the lower the reading percentage. This happens due to the traffic generated in the texts through the sharing in sites like Facebook, for example. The circulation in other platforms helps to discover how many times a text was viewed, but the number of views does not effectively match the same number of people that have read the whole text.

The editors on Medium estimate seven minutes as the ideal reading time. But long-form texts, which can vary between 30 and 40 minutes of reading, also reach a high reading percentage on the platform. An example is an article about a Brazilian train on Bang, published in May 2015, with 34 minutes of reading. The Portuguese version reached about 300 recommendations and seven replies, 25% of reading percentage and 20,000 views, big numbers for content this size on Medium.

Considerations on the Depth of Journalistic Content

In 2012, Tenore questioned what should be considered the definition of long-form. More than the content size, the author underscores the depth of references, justifying her argument with the fact that quality is attractive and would often guarantee that the reader will return to the site. She does not, however, rule out the ease with which long and quality texts may end up forgotten on the web, which can be related to the amount of other kinds of content that are shorter and focused on the speed instead.

According to Fischer (2013), long-form means: “1) a level of in-depth reporting that goes beyond the everyday standard of production and/or 2) narrative storytelling that’s presented in an appealing way, often with multimedia elements to enhance the piece”. In Brazil, Longhi and Winques (2015) started a discussion on the format, mentioning how paradoxical it might seem, in the online journalism environment, the increase in lengthy, deep text. The authors highlight: “In English, long-form is a term that has always been used to define the size and depth in which a subject is approached,” thus it is not exclusive to digital journalism (Longhi & Winques, 2015). The authors place long-form as an element of the great multimedia report, which refers to the idea that multimedia can be considered an element of the long-form format, although it is not a determinant.

In the face of the digital media consolidation, Tenore (2012) asks: how to embrace the constant demand for content without losing focus on quality? And how do you get readers to shift their attention to the amount of content and to give more value to quality content? If, on the one hand, journalists would need to think about content production with a focus on quality, another factor to consider would be how to captivate audiences around these materials. Tenore (2012) highlights the need for tools that guarantee a permanent power to the contents; and in addition to these tools, the behavior of producers and consumers of journalistic content in these spaces is a central element in the constitution of practices that value the deepening of the content, as well as the captivation of the public. Medium does not lack publications and profiles that provide long texts. However, the size of a text does not attest to its quality, nor does its depth guarantee it will be accessed, not even the reading time (TTR).

The time of writing, editing, and then presenting to the reader is one of the factors considered when talking about long-form journalism. This is a characteristic of what has been called “slow” journalism, and that Le Masurier (2015, 2016) attributes its consolidation to the discussions about a new way of thinking about journalism in the face of problems caused by pressure to produce news to meet the demands for continuous updating—something that the market itself has generated. “This journalism does not require a checklist of key characteristics to qualify as Slow. The term, like the Slow movement itself, is more a critical orientation to the effects of speed on the practice of journalism” (Le Masurier, 2015, p. 439). It is not something essentially new, as she states, since in previous centuries the practices of slow journalism were already a reality. The novelties in this context are the loss of content quality, the compromise of ethical standards and the decrease of the attention of the user, amid the abundance of contents. Without striving to define characteristics that make up what she calls slow journalism, Le Masurier (2016) says that practitioners of this format work on a critique of the limits and dangers that the speed of production and the production overload imposed by the contemporary environment of mainstream journalistic vehicles.

Hermann (2015) understands that production time adds value to the final content. Hermann (2015) speaks of slow journalism as a practice belonging to a type of ethnographic journalism. For her, journalists and media watchers understand slow journalism as a type of long-form journalism.

The difficulties of reading on the screen in the early years of the web and the diversity of activities performed on the computer are reasons pointed out by Rosenstiel (2013) for the failure of some initiatives based on long-form. Smartphones and tablets would be the most appropriate devices in their design, considering them as one-activity devices. It is understood that there is an audience for in-depth content on different types of device. On the other hand, a simple notification that appears on the screen may distract the attention of the reader. Nevertheless, the use of mobile devices can help in content production, which is the case of Medium allowing drafts of stories, thus increasing the creation of new texts regardless of where the individual is.

Medium brings together functionalities that allow the creation of initiatives focused on long-form, as identified in the collection of publications, but goes further, by incorporating functions that require individual and collective engagement, which happens through the network structure, but is not ensured to all publications and stories on the platform.

Considerations on the Reach of Journalistic Content

The ways through which online content circulate nowadays are a result of the evolution of a digital culture described in detail by Jenkins, Green and Ford (2013) from the concepts of convergence and spreadability. In the context of this research, both notions direct the investigation issues on Medium.

As pointed out by Heinrich (2011) on the use of the expression “web journalism,” there is no novelty in the idea permeating the understanding of this type of journalism. For her, it is the “‘concrete network’ of an evolving (global) journalism sphere” (Heinrich, 2011, p. 31). Castells’ (1999) notions of space of flows and network society are the basis for her argument. Understanding that network logic is more powerful than the individual powers in the network, she identifies an adequate environment for the implementation of networked structures—which refers to the model thought by Medium to consolidate the platform in the connections between its participants. The interactions enhanced by the digitalization of processes and a connectivity model marked by the network paradigm are questions used to think about journalism, in this case.

According to Heinrich (2011), the transformations resulting from the changes in the processes of investigation, production and distribution of news impact on the work of journalists as: a) The number of agents involved in these processes increase and it is no longer limited to journalists; b) The amount of information available online increases, as well as the use of tools for collecting and checking facts; c) The online environment allows instant feedback and active participation of users, redefining the communication flow, previously verticalized, and allowing the establishment of networks. These levels of impact are a result of the situation of the systemic crisis narrated by Oliveira and Henn (2014), due to the diversity of agents that can be part of the process of investigation, production and journalistic distribution.

In this context, Heinrich (2011) indicates the demand for a repositioning of journalistic organizations in the face of these impacts through a structural transformation at the moment in which the emergence of a global news sphere constitutes multidirectional information flows. “The evolution of journalism and the adaptation of new technologies thus have to be viewed as a constant flow in which the working procedures of journalists are developed and reshaped” (Heinrich, 2011. p. 16). Among several understandings, Heinrich (2011) makes an attempt to exhaust the meaning of the expression “web journalism.” First, recalling that the idea is associated with citizen journalism or participatory journalism, from which the interference of different agents in the journalistic processes can be traced, thus involving topics such as participation, partnership and conversation. Jarvis, for example, focuses on the journalist as a mediator of a collective work among these different agents (2007). The creation of many publications in Medium takes place in this way: some journalists, along with other professionals, manage the publication, being able or not to produce content, but also acting as curators of contents published by writers on the platform. Conversations are established between these individuals, as a collaborative editing process, which results in the linking of the writer’s profile text to the publication space on the platform.

Heinrich (2011), among the definitions she presents, points out that what brings together the understandings about web journalism is “the interplay between civic participation and journalistic action” (p. 58). Network construction is the glue between the different arguments that attempt to define web journalism. She points out, however, that among these references, there is no application to draw a conceptual model of professional journalism as well as its organization. They are considerations made randomly, without any pretension to consolidate a journalistic practice or genre.

Hypertextuality as a feature of digital journalism is undoubtedly an element that fosters the deepening and contextualization of online content. Wu, Hofman, Mason, and Watts (2011), while trying to map who says what to whom on Twitter, approach the link as an element that assists in keeping interest. In their research, they identify which blogger links, advertised in tweets, are clicked more than links published by the press, but also show how broader magazine articles are able to extend readers’ interests. On Medium, the practices around the possibilities of using links in texts are diverse, but the resource strengthens the network character of the platform by linking not only texts and internal publications, but also the profiles of users, as well as in other social networking sites, can link to each other, and be mentioned in the published texts. The tags that organize the content on Medium and govern the search on the platform also act as links that guide the registration of the stories and facilitate the search for the contents. The tags, when thought of as links (Aquino Bittencourt, 2007), extol the collective character of the platform, revealing how deep and far reaching they are for the journalism practice.

Heinrich (2011) cites a post by Karp (2008), which highlights the importance of hypertextuality for the contextualization of content, for the enrichment of a dataset, sources and information about what is being addressed. By equating the way journalism uses links to mechanisms like Google or Digg, it understands that the greater the combination of the best links, the better is the use that journalism makes of hypertextuality, “because those links are determined by networks, not individuals—and networks are the most powerful force on the web” (Karp, 2008).

This hypertextuality of contents is then extended to the actors in a network. Recuero (2012) explains how the connections between individuals go beyond social ties. The tools that provide the emergence of representations of these connections are fundamental to understanding how connections are established and maintained in networks. Operating in a social network logic means that it is the content that promotes and supports connections between Medium users. Choosing to follow a profile or a publication on the platform may even be the result of a previous friendship, but since it is the content that moves the interactions in the system, the formation and duration of these relations depends, on most cases, on the quality of the content.


A quantitative-qualitative survey was made to try to identify elements that trigger the perspective of transforming the format of the content in relation to TTR (Total Time Reading). The incipient character of the project at the time allowed only a few inferences to be proven. It was noticed that the click culture ceases to be the focus on what has been built through Medium by different publications that direct their efforts for the deepening of the texts and for the construction of visibility and reach through new strategies focused on the quality of content. It is questionable, however, if this click culture has not been replaced on the platform by a culture of  recommendation, since, unlike sites in which the access to the text already attests to the popularity of the content, on Medium it is the amount of recommendations that makes explicit how much a text is viewed inside. Still, it is like exchanging it for something similar, because the recommendation still does not guarantee that the story has been read in full—nor does the access. It is at this point that the TTR becomes a relevant way to measure the consumption of stories on Medium, by linking in its measurement performance the full reading certification of the story.

Considering that the organization of the contents on Medium works through tags, it was defined as strategy the collecting of four tags to conduct the mapping of Brazilian and American initiatives: jornalismo, jornalismo digital, journalism and digital journalism. The search method for the four tags chosen was applied only to select publications. This is because it is understood that in order to achieve the research objectives of identifying and analyzing how digital journalism is transforming, it is more productive to analyze publications, which have teams to think about the production of content and the dynamics of circulation, than to focus the study on individual profiles maintained on the platform. However, the possibility of including individual profiles in data collection and analysis throughout the project is not ruled out.

At the beginning of the investigation, it was possible to notice that many publications had few followers, which caused the first collection cut: publications with less than 1,000 followers would not be included in the dataset. It was then defined that the publication description would be collected to try to understand its main purpose on Medium. The next selection sought to identify the publications in Portuguese and English. Then, a final set of six publications in Portuguese and 26 in English was identified. See the Appendix for the details.

Publications marked with the tags in Portuguese direct their content to talk about journalism, which is the word that most appears in the set of these descriptions. The same happens with the publications in English, which also due to the greater amount if compared to what was collected in Portuguese, broaden the approach regarding vehicles, news and the media. It is in the descriptions in English that technical terms of journalism  are highlighted, as, for example, content, covering, newsroom, publish, story, storytelling, among others. One can see in these descriptions an objective that goes beyond the practice of reporting facts and events. In some cases, these publications propose to discuss journalism and/or to experiment with journalistic practices, formats and languages and talk about these experiments.

In order to deepen the study of these publications, a set of four criteria was established, to evaluate how the publications incorporate the elements of deepening and reach in the stories they publish:

  1. Frequency of publication and size of the stories: it evaluates the periodicity of publication of stories and the average time of reading of each publication.
  1. Nature of content: identifies whether the content is proper to the publication or result of curation by editors in stories of writers who are Medium users.
  1. Technical characteristics of digital journalism: evaluates how publications use hypertextuality and multimedia to deepen the content.
  1. Circulation strategies: identifies circulation mechanisms, indicating the tools used to spread the contents and increase the reach of stories.


The selected publications, both in Portuguese and in English, do not maintain a daily frequency of publication of stories, but there are exceptions, like Agência Democratize and Jornalistas Livres, which might publish more than one story a day. What is common among most publications is that they have flaws in this timeline of content publishing, in some cases in the early days of the platform, and later on in other cases. A Nébula, for example, has some stories published many days in a row, but never more than one story a day. It is more common to see a period of non-publication than daily stories. In some cases, these periods are of some days, or even months. Launched in 2016, InsideClimateNews has only four stories published until November of that year.

Brio Stories is a publication that publishes long, in-depth articles and spends many months without publishing anything new. For a new Medium user, a publication that stays for long periods of time without being updated can be interpreted as abandoned, and therefore miss the chance to win new followers and recommendations to the stories already made available. Among the collected publications, Brio Stories is the one that has the most extensive stories, dividing many of them in chapters and converging with what Le Masurier (2015, 2016) says about slow journalism, in relation to the times of investigation, writing and editing. Revista Poleiro offers texts that surpass 10 minutes, but the majority is part of what was called here short stories, which are below 10 minutes of reading. The frequency of publication and the size of the stories are directly related to the type of content, that is, in the cases of A Nébula and The Nib, for example, the reading time is always short, around one to two minutes, because the content is always a comic strip. It is common to find stories that exceed 10 minutes of reading, showing how longer contents, deep or not, are common on the platform. This significant occurrence of stories that exceed 10 minutes converges with the popularity of longer stories. According to Medium editors, the average reading time on the platform is seven minutes. Even though long-form texts can vary between 30 and 40 minutes of reading, they also achieve a high percentage of reading.

But what makes someone recommend a story on Medium? This is a frequent question and thought here based on criteria related to the type of content and circulation strategies adopted by the publications. In addition to the personal interest that those who recommend have in the subject, one can think that there is also the reputation that a profile and a publication aim to build on Medium to be taken into consideration. In the previous example, about Bang’s 35-minute story, which generated about 300 recommendations, it is important to note that the author of the story is the editor of Medium in Brazil, and has about 4,000 followers of his profile, which contributes to the number of recommendations increasing. And it is here that one realizes that it is within Medium itself that a process of construction of circulation strategies adopted to maximize the reach of a story begins. Initially, this reach will depend on who is publishing, even though the story is linked to a publication with a large number of followers. When the story is published by a user’s profile and is linked to a publication, the reach can be even greater because the story is seen by the profile and the publication’s followers. Thus, the circulation of stories within Medium depends on the number of followers of the publications and the profiles that are linked to them. This means that in addition to a battle for recommendation, there is a new fight for the gaining of followers.

There are publications that use only their own unique profile, without identifying the author of the story and reaching considerable numbers of recommendations, others not so much, such as Behind the Headlines, which has few recommendations of their stories. The curation of stories that a publication makes can be considered an accelerator of recommendations, since, as said before, the visibility of content is enhanced when the text is seen not only by the followers of the publication, but also by the followers of the profile in which the story is made available. It is noticed that in publications in Portuguese and English, the largest number of recommendations are linked to a profile of a writer and not to a publication.

Social media such as Twitter and Facebook have a role that complements this social networking dynamic that Medium provides through this recommendation scheme. Facebook is the social network site most used by publications, which through fan pages spread the stories, expanding the reach beyond Medium. Twitter comes second, often along with a fan page, but in many cases, both Portuguese and English, Twitter accounts are outdated.

This publication’s choice of broadcasting its own content or curating writers’ texts through Medium also interferes with the format of the stories, which can be more or less deepened by exploring not only the features offered by the platform editing system, but also the online medium, such as hypertextuality and multimedia. Similar practices are identified regarding the use of links and formats other than text, such as images, audios and videos to compose the stories. What should be highlighted here is that just as the reach of the stories ends up depending greatly on the amount of followers of the writers, the deepening depends on how the writers linked to the publication structure their texts. Publications that carry only their own content have more control over the composition of the stories and the formats used. Publications composed of stories published on Medium have less editing power in this sense, and in many cases none. In publications such as The Nib and A Nébula, whose stories are always a strip, the deepening is limited, as is the exploration of links and other resources of the digital medium, considering that only one image composes the story, besides the title. Although Brio Stories is composed of very long content, in some stories it invests in literary journalism, and ends up leaving aside possibilities of hypertext and even of images to add up to the stories. In this same publication, stories guided by an investigative journalistic practice, already present a little more diversity in the languages that compose the contents, which they call reports. However, their storyline reveals a goal that seems to focus on story consistency through the written/typed word rather than images, videos, or links to other content.

Hypertextuality is a resource that can be explored by publications on Medium not only through links within stories, but also through tags, which can be used to organize sections of a publication, as well as sections for organizing websites. When folksonomy (Vander Wal, 2006) became a more common practice on the web, it soon became apparent that it was useful in organizing the large amount of information circulating. Flickr and were some of the pioneers in using tags as a collective management mechanism, which in the mid-2000s was called social tagging. At that time, these uses were investigated understanding the constitution of what was called hypertext 2.0 (Aquino Bittencourt, 2007). The formation of a new process of information organization and search based on hypertext was seen, but it subverted earlier forms of folksonomy and converged with collective and collaborative ideals derived from the notion discussed on the web on web 2.0. This indexing system that allowed the insertion of tags describing content of stored documents was based on a free organization, made by users in a manner unrelated to programmers. It was understood at that time that folksonomy opened space for a new type of link, the tag, created by any individual web user, who could thus manage the organization and search for data online as so far was not possible.

In the case of Medium, tags play this organizing role that facilitates the search for content. The deepening of the stories ends up being a more editorial choice than organizational, as it happens when the tags are used to organize the stories of the publication. It is hypertext made by writers and readers on the platform, allowing a content management free from the action of filters based on algorithms. Among the publications selected in this analysis, few use tags to organize their content internally. For those who access the publication, the tags titling internal sections make it easier to search for stories. For those who write on Medium and are interested in linking their content to a particular publication can tag a story with the corresponding tag, increasing the chances of having the story selected by the curation of the publication.

The publications selected in this mapping are more about the content they approach, than about how they build their stories and circulate them in and out of Medium. Among the ones, which are far from a discussion of digital journalism effectively, Latterly and The Cube are interesting cases. The Cube talks about games, maintains a strict editorial policy, with rules that value the quality of the content, very grounded in opinions and interviews of people related to the world of games. It does not discuss journalism, but employs journalistic techniques for checking content, researching, and editing. Latterly is a publication that addresses issues of social justice in more than 30 countries, with the collaboration of people in several countries and writing for well-known vehicles such as The New York Times and The Atlantic, for example.

What brings these publications closer are the practices they carry out on Medium that intensify the discussions in the field of digital journalism on formats and dynamics. The exploration of the criteria of analysis shows a very similar appropriation of the composition of the stories. In some cases, the life of a publication on Medium generates learning about the uses and intensification of potentialities, such as hypertextuality and multimedia, as well as the recurrence of the newsletter as an instrument of extension of reach. There is an attempt to standardize the publications and the collaboration policies indicate the intention of the editors to adapt the stories to a set of editorial principles. Interactivity is still scarce, except for cases of stories that are highly recommended, and with the visibility they reach on Medium, generate an internal movement on the platform.

It is relevant to consider that Medium brings together practices and functionalities of other platforms and publishing mechanisms, such as blogs and social networking sites. Medium is not a journalistic or content production company. With its current configuration, it is much more a company that offers technology for publishing content, and thus ends up offering journalism a space that is independent of social networking sites such as Twitter and Facebook, but at the same time generates new schemes of internal dependence using the model based on recommendations.


It is important to point out limitations of this study. Medium often makes layout changes, deletes features, resumes others, always with the goal of perfecting the user experience. These changes, however simple they seem, which go unnoticed by many users, can be limitations for the research. It was also noticed that some writers listed in some publications had no followers, which interferes directly in the visibility of the stories they publish. The same was verified in publications, some without followers and without stories, others with a very low number of followers, although with stories published. There are cases where writers who appear as contributors to a publication have never published any stories.

It is possible to notice that, among the analyzed publications, the only ones that maintain a daily frequency, with up to more than one story per day are publications of essentially journalistic bias, with very short stories, reproducing, in a sense, the practices of continuous updating. In this way, they disagree with the precepts about long-form journalism and slow journalism, and there is a need for an in-depth content analysis of this practice.

Medium has been used as a platform for publishing journalistic content of its own, by profiles and publications, but considering what has been verified in this study, curation is one of the practices that redefine the journalist’s role in the digital environment. In other platforms, the activity has already been carried out. One relevant example is Twitter Moments, in which a team of journalists do journalism by curating tweets. Medium allows this, valuing the journalistic activity as a guide to publications that even without producing their own content, can compose journalistic initiatives based on the content made available on the platform.

Many journalistic publications did not appear in this study because they do not use any of the hashtags defined for the search (#jornalismo; #jornalismodigital; #journalism; #digitaljournalism). In this sense, a new series of publications could be explored in future research projects. Another area to explore could be analyzing the profiles of writers of journalistic publications and evaluate how the number of followers of these profiles interferes with the reach of the stories linked to the publications.

For the journalism industry this project has relevance when it points to the movement of independent initiatives. However, this project also aims to observe how big publishers deal with new business models for publishing content on the internet.


Maria Clara Aquino Bittencourt is a professor and researcher at Post-Graduate Program of Communication Sciences at University of Vale do Rio dos Sinos/UNISINOS. Post-Doctor at the same program. PhD and Master of Communication and Information at Post-Graduate Program in Communication and Information at University Federal do Rio Grande do Sul/UFRGS. Graduated in Social Communication, Journalism qualification at University Católica de Pelotas/UCPEL. Researches about digital journalism on the following topics: media convergence, information spreading, movements and social mobilization, hypertext and folksonomy. Member of Laboratório de Investigação do Ciberacontecimento (LIC) and Grupo de Estudos em Jornalismo (GPJor). Co-founder of Aaron ( 



Anderson, C. W., Bell, E., & Shirky, C. (2012). Post-industrial journalism: Adapting to the present. Tow Center for Digital Journalism. Retrieved from

Aquino Bittencourt, M. C. (2007). Hipertexto 2.0, folksonomia e memória coletiva: Um estudo das tags na organização da web. In E-compós Agosto de 2007 – 2/18. Retrieved from

Aquino Bittencourt, M. C. (2012). Convergência entre televisão e web: proposta de categorização analítica (Tese de Doutorado). Programa de Pós-Graduação em Comunicação e Informação. Universidade Federal do Rio Grande do Sul.

Boyd, D. M., & Ellison, N. B. (2007). Social network sites: Definition, history, and scholarship. Journal of Computer-Mediated Communication, 13, 210-230. doi:10.1111/j.1083-6101.2007.00393.x

Castells, M. (1999). A sociedade em rede. A era da informação, economia, sociedade e cultura. São Paulo, Brazil: Paz e Terra.

Costa, C. T. (2014). Um modelo de negócio para o jornalismo digital. In: Columbia Journalism Review, a Revista de Jornalismo ESPM nº 9 (abril, maio e junho de 2014), 51-115.

Fischer, M. C. (2013, December). Longform: More than just a lot of words. American Journalism Review. Retrieved from

Heinrich, A. (2011). Network journalism. London, UK: Routledge.

Hermann, (2015). The temporal tipping point: Regimentation, representation and reorientation in ethnographic journalism. Journalism Practice, 10(4), 492-506. Retrieved from

Jarvis, J. (2009) What would Google do? New York, NY: HarperCollins.

Jenkins, H. (2006). Convergence culture: Where old and new media collide. New York, NY: NYU Press.

Jenkins, H., Ford, S., & Green, J. (2013). Spreadable media: Creating value and meaning in a networked culture. New York, NY: New York University.

Karp, S. (2008). How networked link journalism can give journalists collectively the power of Google and Digg. Publishing 2.0. Retrieved from:

Le Masurier, M. (2015). What is slow journalism? Journalism Practice, 9(2), 138-152. doi: 10.1080/17512786.2014.916471

Le Masurier, M. (2016). Slow journalism: An introduction to a new research paradigm. Journalism Practice, 10(4), 439-447. doi:

Longhi, R., & Winques, K. (2015). O lugar do longform no jornalismo online. Qualidade versus quantidade e algumas considerações sobre o consumo. Trabalho apresentado no 24º Encontro Anual da Associação dos Programas de Pós-Graduação em Comunicação (COMPÓS). Brasília (DF), 09 e 12 de junho de 2015. Retrieved from 0b444ac7347b_2852.pdf

Oliveira, F. M., & Henn, R. C. (2014). Movimentos em rede e ocupação do espaço público: Limites e possibilidades ante a crise do jornalismo. Contemporanea: comunicação e cultura, 12(1). 39-54. Retrieved from

O’Reilly, T., & Batelle, J. (2009). Web squared: Web 2.0 five years on. Web 2.0 Summit. San Francisco, CA. Retrieved from

Recuero, R. (2012). A rede é a mensagem: Efeitos da Difusão de Informações nos Sites de Rede Social. E. Vizer. (Org.), Lo que Mcluhan no previó. Buenos Aires, Argentina: Editorial La Crujía, 1. 205-223. Retrieved from

Rosenbaum, S. (2011). Curation nation: How to win in a world where consumers are creators. New York, NY: McGraw-Hill Education.

Rosentiel, T. (Presenter). (2013, May 28). The Future of Journalism (TedX Talks, YouTube). Available from

Saad Corrêa, E. (2011). Estratégias de conteúdos para meios digitais. In C. Quadros, K. Caetano, & A. Laranjeira (Eds.), Jornalismo e Convergência: ensino e práticas profissionais. Covilhã, Portugal: LabCom Books.

Saad Corrêa, E., & Bertocchi, D. (2012). A cena cibercultural do jornalismo contemporâneo: Web semântica, algoritmos, aplicativos e curadoria. MATRIZes, 5(2), 123-144. Retrieved from

Salaverria, R. (2015). 20 años de ciberperiodismo en Iberoamérica: Tendencias y conclusiones. ISCTE – Escola de Sociologia e Políticas Públicas. Instituto Universitário de Lisboa. Retrieved from

Stone, B. (2012). Building Medium. Medium. Retrieved from

Tenore, M. J. (2012). What do we mean by ‘longform journalism’ & how can we get it ‘to go’? Poynter. Retrieved from

Vander Wal, T. (2006) Explaining and showing broad and narrow folksonomies. Retrieved from

Wu, S., Mason, W. A., Hofman, J. M., & Watts, D. J.  (2011, March-April). Who says what to whom on Twitter. Proceedings of the 20th international conference on World Wide Web, Hyderabad, India. Retreived from


Appendix: Publications collected for analysis 

Portuguese publications by keyword:


Agência Democratize

Jornalistas Livres

Brio Stories

Revista Poleiro

Entusiastas de Mídias Sociais

A Nébula

#journalismo digital

Entusiastas de Mídias Sociais


Jornalistas Livres

Brio Stories

#digital journalism

Entusiastas de Mídias Sociais


English publications by keyword:

#jornalismo – None available.

#journalismo digital – None available.


The Cube

Digital Trends Index

We Are Hearken

The Nib




Whiter News?


First Draft News

3 to read

The Team

The Guardian Mobile Innovation Lab

Texas Tribune

Google News Lab

InsideClimate News

Thoughts on Journalism

Mother Jones

Firsthand Stories

One Bendle

Prime Mind

Thoughts on Media

Journalism Innovation

Behind the Headlines

The Local News Web

The Compass Standard

#digital journalism

The Compass Standard