Hi, Javier. I don't really see it as a 'political' issue, but rather one of competence. Content distribution algorithms are designed to recommend a constant stream of new content to users based on their interest in previous content. For Netflix this is quite hard - all they have is 'Watched: Y/N', and 'Minutes of total running time completed', as practically no one actually checks the 'liked/didn't like' boxes.

They still manage a plausible job of digging stuff out (admittedly from a much smaller list of discrete elements).

Medium has huge amounts of data about what we liked, and how much, but fails to put that into practice.

I'm not overly concerned with monetary conversion - what I want is for the data that they use for that also to be used effectively to funnel subsequent articles to the right people. I don't believe that happens. And I have never come across anyone on this platform who does.

Either they deliberately keep it largely random (which is dishonest and in bad faith, and which one could deem a 'political' decision) or they are incapable of doing any better, which is a sorry state of affairs. I believe there are a fair number of software engineers in the San Francisco area where they are based...

The reward for writing is for one's work to be read and enjoyed. Medium delivers only a small fraction of what it could and should in that regard. And it is worth bearing in mind that this is equally to the detriment of readers, who do not receive the content they have indicated they would enjoy.