The arXiv is the most popular preprint repository in the world. Since its inception in 1991, the arXiv has allowed researchers to freely share publication-ready articles prior to formal peer review. The growth and the popularity of the arXiv emerged as a result of new technologies that made document creation and dissemination easy, and cultural practices that incentivized collaboration and data sharing. The arXiv represents a unique place in the history of research communication and the Web itself, however it has arguably changed very little since its creation.  Here we look at the strengths and weaknesses of arXiv in an effort to identify what possible improvements can be made based on new technologies not previously available. Based on this, we argue that a modern arXiv might in fact not look at all like the arXiv of today.  

The current crisis of veracity in biomedical research is enabled by the lack of publicly accessible information on whether the reported scientific claims are valid. One approach to solve this problem is to replicate previous studies by specialized reproducibility centers. However, this approach is costly or unaffordable and raises a number of yet to be resolved concerns that question its effectiveness and validity. We propose to use an approach that yields a simple numerical measure of veracity, the R-factor, by summarizing the outcomes of already published studies that have attempted to test a claim. The R-factor of an investigator, a journal, or an institution would be the average of the R-factors of the claims they reported. We illustrate this approach using three studies recently tested by a replication initiative, compare the results, and discuss how using the R-factor can help improve the veracity of scientific research.

I was accepted into the cell biology program at Virginia Tech under conditional terms due to a mediocre undergraduate GPA. This was the deal: maintain good grades and I’d get to continue, or slip up and I was out. As an undergraduate, I spent a lot of time surfing and very little time  cramming for tests -- what can I say? I wasn’t exactly a traditional grad student applicant.Despite my shortcomings on paper, I was ambitious. Before grad school, I contacted a researcher from Harvard who’d proposed through mathematical models that we could kill cancer cells with cancer cells \cite{Deisboeck_2008}.  I told him I wanted to test his proposal experimentally. When he wrote back and I brought the proposal to my potential PI, I quickly realized that incoming grad students don’t actually do this. You’re supposed to go through rotations first, and then select a lab, pick a project that falls within the scope of your PI’s research, and so on. This wasn’t exactly my style.The deeper I got into my PhD, the more I realized the game you have to play in order to be successful: publish in certain journals, publish with the best coauthors you can manage, publish as much as you can. I played the game and published as much as possible within the scope of cell biology and cancer, but also papers within the scope of the scientific communication process itself -- papers on funding, peer review in high-impact journals, and peer review at the NIH. I wrote about cancer but I also wrote about all the problems I was seeing around me in the process itself.I never thought about actually doing anything about these systemic problems until I read The Trouble with Medical Journals \cite{Smith_2006}. The key tenet  -- that peer review misses most major errors -- is the idea that sent me down the path of building a publishing company to take the whole publishing process and flip it in favor of openness. Instead of filtering results and then publishing, I wanted scientists first to publish and then to filter -- to publish and then winnow, so to speak. That’s why the Winnower was born.From Scientist to EntrepreneurI didn’t know anything about starting a business but I knew I needed some money to do it.  I wrote up some ideas for a new publication, entered a business contest on campus, and lost. It was harder than I thought.  But then I sent that proposal to some people I knew from undergrad and through a lot of luck managed to get 50k from a private investor.The Winnower launched in May 2014 and over the course or two years we shifted away from publishing traditional papers to publishing so-called grey literature -- informal documents that traditional publishers ignore.  We published scholarly reddit AMAs, foldscope images, responses to NIH RFIs, journal clubs, and some of the coolest essays I’ve ever read. We formalized blogs and journal clubs so that they could act as reviewers themselves. People liked what we were doing, as judged by the growth of publications and readership. Why shouldn’t reddit AMA’s have DOIs and be given real scientific consideration? I gave talks around the world, raised more money, and met other academics doing similar things with their own companies. I felt lucky and privileged to be doing what I loved, despite the fact I was making less now running a company than I was as a grad student. The End.Okay, the story doesn’t have an ending yet because the story is still ongoing.  Very recently The Winnower was acquired by Authorea, another early company working on the same problem but from a different direction. Authorea, which is also founded by former academics, is fixing how researchers write, collaborate, and share online.  Together we’re working to become the place where researchers can write and publish whatever they want collaboratively and online. It’s an ambitious goal but so too was cancer research.I can’t say if we’ll achieve our goals and I know the road ahead is still daunting but I think the problems we’re working to solve are as hard as some of the most complex problems in science. What is certainly true is that we must work collaboratively to solve them.  I hope this essay inspires more academics to follow their own “crazy” ideas and I hope you’ll stand with our mission to build a more transparent system of research communication.  Let’s get it right.

Research is really f**king important.  This statement is almost self-evident by the fact that you're reading this online.  From research has come the web, life-saving vaccines, pasteurization, and countless other advancements. In other words, you can look at cat gifs all day because of research, you're alive because of research, and you can safely add milk to your coffee or tea without contracting some disease, because of research. But how research is done today is being stymied by how it is being communicated.  Most research is locked behind expensive paywalls \cite{Bj_rk_2010}, is not communicated to the public or scientific community until months or years after the experiments are done \cite{trickydoi}, is biased in how it is reported - only "positive" results are typically published \cite{Ahmed_2012}, does not supply the underlying data to major studies \cite{Alsheikh_Ali_2011}, and has been found to be irreproducible at alarming rates \cite{Begley_2012}.Why is science communication so broken?Many would blame the fault of old profit-hungry publishers, like Elsevier, and in many respects, that blame is deserved. However, here's a different hypothesis: what is holding us back from a real shift in the research communication industry is not Elsevier, it's Microsoft Word. Yes, Word, the same application that introduced us to Clippy is the real impediment to effective communication in research.Today, researchers are judged by their publications, both in terms of quantity and prestige.  Accordingly, researchers write up their documents and send them to the most prestigious journals they think they can publish in.  The journals, owned by large multinational corporations, charge researchers to publish their work and then again charge institutions to subscribe to the content. Such subscriptions can run into the many millions of dollars per year per institution \cite{Lawson_2015} with individual access costing $30-60 per article.The system and process for publishing and disseminating research is inimical to scientific advancement and accordingly Open Access and Open Science movements have made big steps towards improving how research is disseminated. Recently, Germany, Peru, and Taiwan have boycotted subscriptions to Elsevier \cite{Schiermeier_2016} and an ongoing boycott to publish or review for certain publishers has accumulated the signatures of 16,493 researchers and counting.  New developments such as Sci-hub, have helped to make research accessible, albeit illegally.  While regarded as a victory by many, the Sci-hub approach is not the solution that researchers are hoping for as it is built on an illegal system of exchanging copyrighted content and bypassing publisher paywalls \cite{Priego}.  The most interesting technologist view of the matter is that the real culprit for keeping science closed isn't actually the oligopoly of publishers \cite{Larivi_re_2015}-- after all, they're for-profit companies trying to run businesses and they're entitled to do any legal thing that helps them deliver value to shareholders. We suggest that a concrete solution for true open access is already out there and it's 100% legal.What is the best solution to truly and legally open access to research?The solution is publishing preprints -- the last version of a paper that belongs to an author before it is submitted to a journal for peer review. Unlike other industries (e.g. literature, music, film, etc.), in research, the preprint version copyright is legally held by the author, even after publication of the work in a journal.Pre-prints are rapidly gaining adoption in the scientific community, with a couple of preprint servers (e.g. arXiv which is run by Cornell University and is primarily for physics papers, and bioRxiv which is similarly for biology papers) receiving thousands of preprints per month.Some of the multinationals are responding with threats against authors not to publish (or post) preprints. However they are being met with fierce opposition from the scientific community, and the tide seems to be turning. Multinationals are now under immense pressure not just from authors in the scientific community, but increasingly from the sources of public and private funding for the actual research. Some organizations are even mandating preprints as a condition of funding. But what is holding back preprints and in general a better way for Authors to have more control of their research?We think the inability for scientists to independently produce and disseminate their work is a major impediment and at the heart of that of that problem is how scientists write. How can Microsoft Word harm scientific communication?Whereas other industries, like the music industry, have been radically transformed and accelerated by providing creators with powerful tools like Youtube, there is no parallel in research.  Researchers are reliant upon publishers to get their ideas out and because of this, they are forced into an antiquated system that has remained largely stagnant since it's inception over 350 years ago.Whereas a minority of researchers in math-heavy disciplines write using typesetting formats like LaTeX, the large majority of researchers (~82%) write their documents in Microsoft Word \cite{brischoux2009don}. Word is easy to use for basic editing but is essentially incompatible with online publishing. Word was created for the personal computer: offline, single-author use. Also, it was not built with scientific research in mind - as such, it lacks support for complex objects like tables and math, data, and code. All in all, Word is extraordinarily feature-poor compared to what we can accomplish today with an online collaborative platform. Because publishers have traditionally accepted manuscripts formatted in Word, and because they consistently fail to truly innovate from a technological standpoint, millions of researchers find themselves using Word. In turn, the research they publish is non-discoverable on the web, data-less, non-actionable, not reusable and, most likely, behind a paywall.  What does the scientific communication ecosystem of the future look like?What is needed is a web-first solution. Research articles should be available on distinct web pages, Wikipedia style. Real data should live underneath the tables and figures. Research needs to finally be machine readable (instead of just tagged with keywords) so that it may be found and processed by search engines and machines. Modern research also deserves to have rich media enhancement -- visualizations, videos, and other forms of rich data in the document itself.All told, researchers need to be able to disseminate their ideas in a web first world, while playing the "Journal game" as long as it exists. Our particular dream (www.authorea.com) is to construct a democratic platform for scientific research -- a vast organizational space for scientists to read and contribute cutting edge science. There is a new class of startups out there doing similar things with the research cycle, and we feel like there is a real and urgent demand for solutions right now in research.