SciViD: The Publisher of Video Science

OUR VISIONScholarly publishing remains essential for disseminating knowledge, but it faces persistent problems of cost, access, transparency and format that limit who can participate and how research is communicated. SciViD is being developed as a complementary publishing model that combines video presentation, AI-assisted production and an inclusive editorial framework, with the aim of addressing some of these problems.CHALLENGES IN SCHOLARLY PUBLISHINGCost and access. Subscription journals restrict access through paywalls, and many open-access journals charge article processing charges (APCs). Both can disadvantage early-career researchers, students and institutions with limited funding.Opaque peer review. Peer review varies in quality, can be affected by bias, and is rarely visible to readers.Unequal participation. Scholarly publishing is concentrated in a small number of regions, and researchers who are not native English speakers face additional barriers to publication.Limits of static formats. Text and static figures convey dynamic processes, experimental technique and simulations poorly, which can hinder reproduction and teaching.HOW SCIVID RESPONDSFree posting. Posting a video preprint on SciViD is free. Professional video production through SciViD Studio is a separate paid service.Transparent status. Every posted item is labeled "Preliminary · Not peer reviewed" and carries a version number, so readers can see which checks it has and has not passed.Multilingual access. SciViD is developing multilingual subtitles and AI-assisted translation to widen access for researchers and readers who are not native English speakers.Visual presentation of methods. Video papers present methods, simulations and devices visually rather than in text alone. This format is intended to support reproduction and to make published work usable for teaching.AI-assisted production. AI tools for narration, animation, editing and translation are intended to reduce the cost and effort of producing a video paper. All AI use must be disclosed, and AI-generated visuals must be labeled in accordance with SciViD’s Integrity Guide.Pathways for early-career researchers. Emerging Scholars gives high school, undergraduate and graduate students and early-career researchers a supervised route to posting their research as a video preprint, with parent or guardian consent for authors under 18.

LIMITATIONS OF STATIC PUBLICATION FORMATSThe text-based research paper remains the principal record of scientific work, and its precision and archival stability are not in question. Text and static figures are, however, poorly suited to conveying certain kinds of information: practical technique, dynamic processes, and the course of an experiment or simulation over time. SciViD’s video papers are intended to address these limitations.1. Under-reported practical techniqueDetails that determine whether an experiment can be repeated, such as instrument settings, timing, sample handling and manual technique, are difficult to specify fully in prose. In the Reproducibility Project: Cancer Biology, none of the experiments selected for replication was described in enough detail to design a replication protocol without contacting the original authors. Incomplete methods reporting is one of several causes of replication failure, alongside publication bias, incomplete statistical reporting, and limited sharing of data and materials; visual documentation of technique addresses only the first of these.2. Dynamic processes reduced to snapshotsProcesses involving motion or transformation, such as fluid flow, self-assembly or the behavior of living cells, must be reconstructed by the reader from a sequence of still images. Research on instructional animation shows a moderate overall learning advantage of animation over static pictures, with larger effects for realistic, video-based material and for learning procedural skills. The benefit is not automatic: dynamic formats help mainly when they show change that static images cannot, and when pacing and signaling are well designed. These findings concern learning outcomes; whether they translate into higher replication success has not, to our knowledge, been tested directly.3. Barriers for readers outside the specialtyResearch papers are written for specialists, and their terminology and compressed style limit access for students, researchers from adjacent fields and non-native English speakers. Video papers that pair narration with visual explanation can lower some of these barriers, because people generally learn better from words and pictures together than from words alone.4. Video as a supplement rather than the recordMany journals already accept supplementary videos, and dedicated video journals exist. In conventional papers, however, video is usually supplementary material rather than the primary object of review and citation. SciViD treats the video preprint itself as the posted, versioned and citable record.THE PATH FORWARDText-based papers will remain the foundation of the scholarly record. Formats that show what text describes, including video papers, interactive simulations, and shared data and code, can complement them by making methods easier to see, evaluate and repeat.

Text-based papers remain the foundation of the scholarly record. SciViD is being developed as a complementary model that keeps the conventions of scholarly publishing while using video to show what text can only describe.1. Scholarly standards with video presentationPosting checklist. Every submission is screened against a posting checklist covering format, originality, references, permissions and consent. Posted items are not peer reviewed and are labeled accordingly.Persistent identifiers. With the authors' consent, SciViD plans to offer a Zenodo DOI for any posted item; authors may also deposit their work themselves and link the DOI.Scholarly structure. Video papers will follow familiar scholarly conventions: abstract, introduction, methods, results and discussion.Visual evidence. Video papers can show dynamic processes, narrated walkthroughs of experimental setups, and simulation output.The format is particularly suited to fields in which motion and transformation are central, such as microfluidics, nanotechnology, soft matter and experimental biology.2. AccessibilityMultilingual access. SciViD is developing multilingual subtitles and AI-assisted translation for readers who are not native English speakers.Accessible video. Captions, transcripts and audio description are needed for video papers to be accessible to readers with hearing or visual impairments; SciViD intends to provide them.Low-bandwidth access. Because video requires more bandwidth than text, SciViD aims to offer lower-resolution and downloadable versions for readers with limited connectivity.3. Education and student authorshipTeaching use. Video papers can be used in flipped classrooms and project-based courses.Student portfolios. Through Emerging Scholars, students can build a public portfolio of posted work in research and scientific communication.Independence of screening. Teachers and mentors may advise students while they prepare their work but have no role in screening their submissions.4. Methodological transparencyDirect observation. In experimental video papers, viewers can see recorded footage of instrument calibration, sample handling and troubleshooting rather than inferring these steps from written description. Recorded footage is distinguished from conceptual animation in accordance with SciViD’s Integrity Guide.Complement to written protocols. Such footage can complement written protocols, particularly for procedures that depend on manual technique.5. Machine-readable recordsTranscripts and metadata. SciViD plans to publish each video paper with a transcript and structured metadata, so that its content can be searched, indexed and processed by text-based tools.Interactive extensions. Authors can link an interactive lab to their posted item, and SciViD hosts such labs free of charge.6. Publishing costsFree posting. Posting is free. Professionally produced video papers are available separately through SciViD Studio.Institutional partnerships. SciViD welcomes partnerships with schools and universities that wish to use video papers in teaching and research training.7. Scholarly and public visibilityHosting and versions. Each posted item is hosted on SciViD with its version history; copies may also be shared on public video platforms to reach a wider audience.Discoverability. With DOIs and structured metadata in place, published video papers are intended to be findable through search engines and academic discovery tools.Multiple uses. A single video paper can serve as a citable research output, a teaching resource, and supporting material for funding applications and conference presentations.SciViD aims to combine established scholarly standards with visual communication, making experimental science easier to follow and authorship more accessible to early-career researchers.

Dear colleagues, educators and students,For nearly two decades I have studied catalytic micro- and nanomotors, rolled-up nanomembranes and microfluidic systems, including postdoctoral research at the Max Planck Institute for Intelligent Systems in Stuttgart, a visiting appointment at Harvard University, and more than eight years as a tenured professor at Fudan University. As an author, a reviewer for more than 50 journals and a guest editor, I have repeatedly encountered the limits of text and static figures. The motion of a nanomotor or the timing of a microfluidic protocol could be obvious in the laboratory yet hard to convey in print, and students and colleagues sometimes misread procedures that looked clear on paper.I have also seen talented students, particularly those outside well-funded institutions or writing in a second language, struggle to publish their first results.SciViD is my response. Its aim is to keep the core standards of scholarly communication, including honest labeling, persistent identifiers and transparent ethics, while using video to show what text can only describe. It is intended to complement traditional journals, not to replace them, and to serve both research and teaching.We are building SciViD around three commitments:Integrity: clear posting standards, honest labeling and transparent rules for the use of AI.Inclusion: a supervised pathway for student authors and support for researchers who are not native English speakers.Impact: video papers that researchers can cite, teachers can use in class and students can learn from.SciViD is at an early stage and is now accepting submissions. I invite you to help build it as an author, editor or partner.Prof. Alexander A. SolovevFounder and Editor-in-Chief, SciViD