In less than two decades, the online video platform—colloquially known as "the tube"—has evolved from a simple repository of user-generated clips into the world’s largest and most influential moving-image archive. Unlike the selective, capital-intensive nature of Hollywood or broadcast television, this digital ecosystem operates on a seemingly infinite scale, hosting everything from abandoned vlogs and corporate web series to historical news footage and algorithmic experiments. To speak of a "tube filmography" is to attempt a near-impossible taxonomy: it is a filmography without a central author, without a fixed canon, and without a traditional distribution gatekeeper. Yet, patterns emerge from this chaos. By examining the structure of a tube filmography—the totality of a creator’s or channel’s video output—and the recurring anatomy of popular videos, one can decode the platform’s unique logic: a hybrid of cinema, television, data science, and participatory culture.

The traditional filmography lists an artist’s works chronologically, suggesting a linear, intentional career. A tube filmography, by contrast, is often nonlinear, recombinant, and shaped by feedback loops. For an individual creator—say, a beauty vlogger or a political commentator—their filmography is not merely a catalog of uploads but a living dataset. Each video’s title, thumbnail, description, tags, and closed captions function as metadata that interacts with the platform’s recommendation algorithm. Over time, a successful channel develops a discernible "filmography logic": early experiments give way to niche refinement, then to format standardization (e.g., "reaction videos," "unboxings," "deep dives"), and occasionally to stylistic branching. This evolution mirrors the serialized nature of television but with the accelerated feedback of digital metrics: a creator can know within hours which video in their filmography resonates, and pivot accordingly.

Moreover, the tube filmography blurs the line between professional and amateur. A major studio’s official movie trailer sits alongside a fan’s shot-for-shot remake, which itself sits alongside a critical video essay deconstructing both. The platform’s architecture—playlists, "up next" recommendations, and collaborative features—effectively curates a meta-filmography across channels. Thus, the unit of analysis is not the individual auteur but the algorithmic corpus: the set of videos that the platform treats as semantically related through views, shares, and co-watch patterns.

The tube filmography challenges traditional film and television studies. It replaces the concept of the "work" with the "video object" — mutable, updateable (edits, reuploads), and algorithmically recontextualized. Popular videos are not judged by mise-en-scène or narrative closure but by engagement metrics and community commentary. Moreover, the platform’s permanent nature (videos rarely disappear) creates a unique archive of vernacular culture: dead memes, obsolete webcam aesthetics, and the rise and fall of micro-celebrities. At the same time, the pressure to produce popular videos has led to homogenization: the same thumbnails, the same pacing, the same "YouTube face." The algorithm’s preference for high-retention, controversial, or emotionally charged content shapes not only what is popular but what is possible to film.