----- title: Hello, PubPub abstract: An introduction to our grand experiment. author: travis_rich name: Travis Rich ----- # A new perspective on publishing Software has evolved into a dynamic process rather than a static product. For example, your phone does not ship with a finished piece of software, but rather a package that receives updates and evolves over time to get better and better. With PubPub, we attempt the same for publishing. Why can't a publication be an evolution rather than a termination, a product of a community as well as a household, one characterized by diversity of expression rather than parochialism defined by print? PubPub is our tool for initiating that change. PubPub is a platform for creating, distributing, and reviewing publications. It modernizes the process by granting first-class status to datasets, interactive visualizations, videos, and rich comments. Instead of snapshots and static images, they become part of an interactive living document, usable as well as readable. But more important, PubPub recognizes three facets of modern research: it is collaborative, it is ongoing, and it increasingly crosses traditional disciplinary boundaries. In order to address these changes, PubPub is deliberately author-driven. It supports publishing and versioning that is immediate and reviews that are post-publication, double-open, and public. It is a place for evolving ideas among communities large and small. [[{"pluginType":"image","source":{"filetype":"image/gif","originalFilename":"gifgif_viyji0.gif","thumbnail":",h_50,w_50/v1453075620/u1w3qsdpix2n6tkc2igp.gif","url":"","_id":"57048d1e843a54b13622cbd0","label":"gifgifviyji0gif"},"align":"right","size":"large","caption":"GIFGIF. One of the many data-heavy, interactive, and animated research projects that led to PubPub."}]] PubPub challenges the traditional model where publications are the stepping stones to tenure; it chooses a metric based on collaborations rather than citations, amplified by community acceptance rather than adulation. The final solutions to this dichotomy are unknown, but PubPub provides a platform for experimentation that allows metrics and observations to guide us forward. Stabilizing our open perspective and experimental tendencies are three core tenets: ## 1. Publishing should be author-driven Authors are in control of the content, styling, and publication of their work. They can receive input from editors, journals, or peers, but the final decision rests in their hands. They should not be held captive by the editorial choices of the journal. Publishing isn’t about journals and their branding - it's about science. As [Nicholas Negroponte]( often proclaims, those who use the medium should be the ones creating the medium. Like artists mining their own pigments or musicians making their own instruments, PubPub is designed to give authors powerful control over the medium on which they publish. ## 2. Distributed trust is real The “peer” in peer review is currently a bottleneck as we rely on individuals to identify worthy peers. However, as [Andy Lippman]( identifies: AirBnB, Uber, and other social systems have shown us that there is trust and scalability in the network. You wouldn’t trust a stranger to sleep on your couch - but you would trust a stranger with an AirBnB profile. Beyond this bottleneck, come the challenges associated with the current system, eloquently outlined by Freire [[{"pluginType":"cite","reference":{"title":"Peering at Open Peer Review","url":"","year":"2015","_id":"56af6d459d09fa3800090214","label":"freire"}}]]: [[{"pluginType":"quote","quote":"The system has been criticised for being too slow (Ware 2008), conservative (Eisenhart 2002), inconsistent (Smith 2006; Hojat, Gonnella, and Caelleigh 2003), nepotist (Sandström and Hällsten 2008), biased against women (Wold and Wennerås 1997), affiliation (Peters and Ceci 1982), nationality (Ernst and Kienbacher 1991) and language (Ross et al. 2006). These complaints have fostered interesting academic debates (e.g. Meadows 1998; Weller 2001), but thus far the literature offers little practical advice on how to tackle peer review problems.","attribution":"Danilo Freire","size":"85","align":"full"}]] We believe that the solution to an inefficient peer review system lies in a distributed peer network, transforming peer-review into peer-to-peer review. While the incentives, motivations, and experience of this are up for design, we put strong faith in the bottom-up approach. Integrating publishing into a system that allows us to reward data sharing, insightful comments, high-impact publishing, and everything else measurable enables diverse metrics on which to infer reputation and “peer-worthiness.” ## 3. Curation is critical but not exclusive Journals were created at a time when distribution was expensive and content was rare. Neither are true anymore, yet journals remain largely unchanged. The critical role that journals do still play is one of curation. Identifying work worthy of a community's attention is not trivial and remains a task best done by humans. We simply haven't been able to automate what it means to be "important" yet. The role of the journal, devoid of the expenses of distribution and access, no longer needs to be exclusive. Towards the goal of great curation, PubPub suggests that the solution is not a better filter - but a set of many more specifically tuned filters. PubPub is built to allow anyone to create a journal. These journals have the autonomy to deem what is important to them and their community. There is no cost to create a journal and we facilitate operating at scales from high-school classrooms to international communities. # Openness and Incentives The goal of scientific publishing is the relentless adventure for truth. However, validating truth is no longer as simple as reproducing an oil-drop experiment. Science today involves generations of genetically engineered mice, million-dollar server farms, or experiments with enormous ethical considerations. The same process that got us atomic theory simply won’t get us a solution to climate change or a cure for cancer. The challenges of modern science are exacerbated by a publishing system that dismisses or fails to reward the reproduction of past results and negative findings. They are exacerbated by a system that incentivizes data hoarding and promotes a dangerous idolization of those who have been fortunate enough to be the winner in a winner-take-all game. They are exacerbated by system that clings to a closed copyright and subscription model that prohibits participation from the majority of the world. PubPub is built with the belief that rewarding the work of many is more productive than encouraging the celebrity of a few. As [Philipp Schmidt]( remarked, there will be no Newton or Einstein of climate change, it will take everyone. PubPub is built to serve as a platform that allows everyone to contribute and encourages science as product of a society. To facilitate science as a social movement - PubPub is designed to enable rich metrics that provide a more honest picture of what it means to contribute. Critical to the advance of science are those who provide peer-review; who design visualizations; who share datasets; who document negative results; who perform long-term, high-risk research; and who provide context to journalists. PubPub gives each user tools to aggregate, visualize, and demonstrate these contributions within the platform and to communities beyond. While many players have done amazing work to augment the existing system with more detailed and complete insights, as long as the publishing process lies with a small set of editors, it is not clear that these will be enough. Forcing authors to optimize for citation counts makes it nearly impossible to optimize for quality science. # Format and Evolution Entangled with the goal of richer representations of contribution is the desire to match the complexity and speed at which research actually happens. The idea of a single snapshot of your research, published months after it has been developed, simply doesn't fit the working mode of many researchers, scientists, and designers. [[{"pluginType":"iframe","url":"","width":"85%","height":"315px","align":"full","caption":"The Macro Connections group at the MIT Media Lab often uses interactive visualizations to communicate their research. Here, we show a visualization of the Pantheon dataset.","reference":{"title":"Pantheon 1.0, a manually verified dataset of globally famous biographies","journal":"Scientific data","volume":"3","year":"2016","publisher":"Nature Publishing Group","_id":"56af6d459d09fa3800090219","label":"yu2016pantheon"}}]] From datasets and visualizations to images and text, we strive to make everything living in PubPub accessible, forkable, and citeable - to be reimagined, redesigned, and rebuilt. Standing on the shoulders of a giants is great - but standing on the shoulders of millions of normal-sized people may be greater. # What's next Our first step in this journey is to launch PubPub as an open platform to the world that allows anyone to publish, comment, and curate. We also begin by facilitating and studying three incredibly open and supportive groups in their efforts to launch new journals. In collaboration with the MIT Media Lab, MIT Press, and MIT Libraries, we are launching the [Journal of Design and Science]( With Kevin Esvelt and his group of collaborators, we are launching []( - focusing on pre-experiment publication and discussion. With the Government of Mexico City, through the Laboratorio de Cuidad de Mexico, we are launching [GlobalCDMX](, a platform that experiments with peer-reviewed legislation and public policy. Even in our first steps we acknowledge the capability for journals to transform into something diverse and entirely different than what we have come to expect. Let’s see what we discover!