Plant Pathology 2.0
Can the World Wide Web (=web) help to significantly enhance the discovery and pedagogical ecosystem supporting the science of plant pathology and related disciplines? My intent herein is to answer this question and also to stimulate a community discourse about how we can address shared challenges by more effectively harnessing and preserving collective knowledge and experience via the web. Hereafter, I use the terms ‘web’ and ‘Internet’ synonymously, although, strictly speaking, the former represents a major application of the latter. To set the stage for this discourse, I start with a brief overview of the web and its impacts.
‘We tend to overestimate the short-term impact of a technology and underestimate its long-term impacts’—Roy Amara. The web seems no exception. Although the dot-com bust quickly dashed wild initial hypes, its impacts have been pervasive and transformative, catalysing ‘disruptive innovations’ in practically every sector of society (Nielsen, 2012; Tapscott and Williams, 2010). Advances in areas, such as the semantic web, smart devices, data transmission capacity and cloud computing, will further its influence. The web has quickly evolved from mainly allowing users to access a cache of information (aka ‘web 1.0’) to enabling mass collaboration on scales and in forms that we had rarely seen before its inception (‘web 2.0’). Many web users have also evolved from passive content consumers to active creators, fuelling the rapid expansion of the type and amount of products. This evolution has been driven by diverse cyber-communities, ranging in size from a few to more than one billion members (e.g. Facebook), in which people interact, share information or work together without being restricted by geographical boundaries and traditional governance structures.
Given that many cyber-communities primarily function as mega cyber-plazas (e.g. Facebook, YouTube, LinkedIn) or -bazaars (e.g. eBay, Amazon), one may question their relevance and application to plant pathology. As outlined below, cultural changes and tools that enabled the formation and growth of such communities can also contribute to innovating research, education and outreach. In particular, mass collaboration via the web, which has been called ‘crowdsourcing’, ‘open innovation’ or ‘groundswell,’ offers proven tools and models for innovation in science. Wikipedia, an online encyclopaedia that has been built and curated by volunteers around the world, illustrates the power of crowdsourcing. Since its birth in 2001, Wikipedia has quickly become the most popular reference source, currently offering 30 million articles in 287 languages. Despite criticisms of Wikipedia's quality, it is comparable with the Encyclopaedia Britannica, whose birth dates back to the 18th century (Giles, 2005). More importantly, Wikipedia users, with others working as collaborators or editors, can create, correct and expand articles with a speed and flexibility that are unapproachable with traditional references.
Other notable examples of crowdsourcing include social media and open source software, such as Linux, a computer operation system, and Firefox, a Linux-based web browser. Not surprisingly, entrepreneurs seeking new business opportunities and models have led technological innovations that enable and nurture crowdsourcing. Politicians have also been quick to exploit its power to build and mobilize support groups. There have been many individual and small group efforts to innovate research and education using the web (see below). However, comparable efforts by scientific and educational organizations have been rather limited and mostly siloed, with the focus on keeping their presence known in cyber-space and supporting consumptive uses of what they traditionally offer with the web as a new delivery mechanism.
Why should we move together towards Plant Pathology 2.0? First, because of the need to study and control diverse diseases threatening agricultural and environmental systems with limited resources, the importance of working closely together and pooling resources and knowledge is much greater in plant pathology than in most other disciplines. As Benjamin Franklin noted, ‘what good men may do separately is small compared with what they may do collectively’; therefore, crowdsourcing may result in much greater impacts than those achieved by disparate individual efforts. Second, as science grows on accumulated data, knowledge and experience, the significance of coordinated community efforts to preserve such foundational resources in a format that can be easily shared and updated cannot be overemphasized. Such efforts are essential, especially to effectively manage and reap benefits from ‘Big Data’ (Pentland, 2013). Third, rapidly evolving web technologies can help to meet multifaceted educational needs. Although the pool of educators and resources in plant pathology has been shrinking, the need to prepare future generations of plant pathologists has not diminished. Globalized agricultural trade and production systems facilitate the long-distance movement of pests and pathogens. Therefore, effective preparedness requires human capacity building on a global scale and in a proactive fashion, so that potential threats are dealt with early and at their origins. Otherwise, we may continue to ‘respond to the ripples after a splash and not prevent the splash’ (Brasier, 2005). Support for life-long learning also becomes vital as the need for the frequent retooling of knowledge and skills increases. The creation of shared online educational resources warrants a serious consideration to meet these needs. Fourth, for generations who grew up with the web as an integral part of communication and learning, better cyber-infrastructures for research and education are indispensible for their professional development.
Finally, as major knowledge generators and disseminators, we should more actively support informed societal decision making on critical issues, such as global food security, environmental stewardship and climate change, by disseminating relevant information and leading discussions through the web. This effort becomes increasingly important in the light of the rapid proliferation of disinformation. In particular, for hot-button issues that incite highly polarized opinions, the web has become the main battleground for the control of public opinion. Given the limited mechanisms to guide the judgement of trustworthiness, the spread of disinformation via the web, which can be fast and pervasive, confuses the public and provides a fertile ground for demagoguery. Other problems, such as cyber-bullying and -crime, and the ‘digital divide’ that further widens existing social and economic disparity within and between countries, also call for our attention. As the web evolves to more intelligently serve users (i.e. cyber-concierge) through mining digital fingerprints left by their activities, the fear that the web will usher in ‘Big Brother’ cannot be discounted (Arthur, 2013). Similar to some technologies that drove disruptive innovations in the past (e.g. nuclear power), the web is a double-edged sword: some of the characteristics that make it a powerful agent for innovation can also cause harm when misused (Pentland, 2013). Our collective effort to increase the beneficial uses of the web will help to keep it on a constructive course of evolution.
Notable Impacts on Education, Research and Outreach
Public colleges and universities have played pivotal roles in the growth and prosperity of many nations around the world by empowering ordinary citizens to contribute to economic and societal development. Accordingly, support for these institutions has long been considered a critical societal investment. However, fiscal constraints, as well as doubts about the cost to benefit ratio, have weakened this premise in many countries. Recent rapid expansion of massive open online courses (MOOCs), spearheaded by for-profit (e.g. Udacity and Coursera) and non-profit (e.g. edX) organizations, in partnership with top-tier universities around the world, has intensified debate about the future of higher education. Although their operational models vary, all of these organizations basically aim to offer ‘a world-class education that has so far been available to a selected few’ (from the Coursera website) to anyone anywhere anytime. Millions around the world have already taken MOOCs.
Is the MOOC movement a harbinger of disruptive innovation for higher education? As the devil is often in the detail, its seemingly noble goal has raised abundant criticisms (Scholz, 2013). However, rather than summarily discounting online education as another technological fad that will quickly fade away, we should experiment with it based on a critical evaluation of its pros and cons. Even in the worst case, ‘experience is what you get when you didn't get what you wanted'—Randy Pausch. Tools and models for online teaching, evaluation and interaction among students and teachers will continue to improve based on such experiences. Online education may not replace experiential learning on campus, but it already serves to augment learning in classrooms (Singer and Bonvillian, 2013) and offers new models for effective teaching and learning (Wildavsky et al., 2011). We literally get paid to drive innovations, but when it comes to innovating how we educate future generations of innovators, we tend to be quite conservative (Wildavsky et al., 2011). I do not want to dwell on what underlies this contradiction, but defensive criticism or inaction for the sake of maintaining the status quo will only reinforce the view that we are dinosaurs. As social media forced traditional media to change their business models for survival, inaction will probably force us to adopt new models. The non-profit Khan Academy offers online lectures, practices, and evaluation tools to help mostly secondary school students and teachers, and symbolizes how the impact of even a single man's good work can be massively amplified through the web.
Changes in research caused by the web are numerous and profound. Once upon a time, submitting sequences via email used to be the only way to search GenBank, which typically took hours or even days to receive the results. How often do you wander around library stacks to find interesting books and papers? My hair was much darker when I did it last time. Remember when postcards were mailed to request reprints? Thanks to increasingly smarter search tools and open access policies adopted by most publishers, a comprehensive list of relevant literature, which is often freely available for download, is only a few keystrokes away. Science blogs, including tweets, offer dynamic and flexible global forums for the rapid exchange of new ideas and developments. The rapid increase in genomics data and large-scale field data presents vast opportunities to advance our understanding of intricate biological processes at levels ranging from individual molecules to ecosystems. However, without online informatics platforms and tools that support data curation, analysis and dissemination, we would mostly remain data rich but knowledge poor.
Although small in scale, the following plant pathology-related examples demonstrate how the web can assist us in pooling resources and knowledge to solve shared problems whilst enriching community infrastructures. Because disease diagnosticians deal with diverse problems in many different plants, they often rely on peer advice for diagnosis. Typically, they sought help through an email listserv, which did not permit the archiving of past communications in a format that could be effectively retrieved to diagnose similar problems later or be used by new members. In 2009, disease diagnosticians migrated to a web platform entitled, ‘Diagnostics Listserve Online’. With its database and basic social media functions, all queries, including photographs, and subsequent responses are catalogued in an easily searchable form. The ipmPIPE, originally built in response to the soybean rust (SBR) invasion of North America, has facilitated cooperation among researchers, field specialists and regulators to effectively manage SBR and other pathogens and pests (Isard et al., 2006). Field observations of SBR channelled into this platform become immediately available to researchers who add value through modelling and analysis, with the resulting management guidelines being transmitted to growers and industry agents. Given the vast microbial diversity and functions that have yet to be discovered and studied, it is critical to archive representative cultures of characterized species and associated data in a format that is readily accessible as references for future discoveries. The Phytophthora Database and Cyber-infrastructure for Fusarium were constructed to support the curation and sharing of such resources (Park et al., 2011, 2013).
Many creative initiatives have engaged the public in scientific discovery, problem solving and education (=citizen science). Some have recruited volunteers to help mine a huge amount of data that would be impractical to process by researchers alone. Foldit enables participants to help predict protein structures or design new structures by simply playing multiuser online games (Cooper et al., 2010). Schrope (2014) reviewed more examples of ‘game with a purpose’ (Ahn and Dabbish, 2008) and how they contributed to solving complex research problems. Galaxy Zoo engages volunteers in classifying the galaxies using telescopic images. Through a phenology project, called BudBurst, a national network of volunteers collect data on the timing of leafing, flowering and fruiting of selected plant species to support studies on how climate changes affect plants. Monarch Watch not only archives data collected by citizens of all ages, but also engages participants in taking actions to help protect the monarch butterfly. Mushroom Observer documents observations about mushrooms around the world and supports mushroom identification. Zooniverse, a web portal, provides links to more examples. Platforms such as IdeaConnection and InnoCentive help businesses and non-profit organizations address specific research and development problems by linking them to potential problem solvers around the world. #SciFund and Petridish allow the public to directly support mostly small-scale research projects (aka ‘crowdfunding’). The biotech industry has also begun to use equity-based crowdfunding. Of course, citizen science is not a recent development. Since 1900, volunteer birdwatchers have taken annual censuses of birds in the western hemisphere via the Christmas Bird Count project. However, the web has drastically increased the type and scale of projects.
So What Can We Do?
First and foremost, we should encourage cultural change. Crowdsourcing will be quite difficult without a prevailing culture that promotes and acknowledges relevant activities. The path of the web is littered with numerous worthy projects that withered, because they failed to mobilize their target communities. Patience and perseverance are needed, because practices form a habit, which then becomes a culture when the majority shares the same habit. Second, the engagement and leadership of relevant scientific societies around the world is vital, in part because projects driven by either individuals or single organizations tend to discourage broader participation. An elephant in the room is how to mobilize established researchers who can, and should, contribute most, but probably are least motivated to participate in web 2.0 activities. As this group leads scientific societies, their contribution will probably be more forthcoming when these societies drive the effort. Third, the web is only a means to an end and should not lead but follow us. Success in crowdsourcing depends heavily on whether we choose worthy ends to pursue rather than the technologies used. Lastly, to build momentum, chosen projects should quickly provide palpable benefits to participating individuals and societies.
As a result of space limitations, I will discuss only one potential project: building a virtual library of experimental protocols, including guides for data collection and analysis. This project offers the practice of crowdsourcing and addresses critical community needs, but does not require a significant new investment, because many such resources already exist in individual laboratories and societies. WikiProtocols and Diagnostician's Cookbook began building a virtual protocol library. Although their progress has been slow, they serve as a model. Protocols have evolved from collective laboratory or field experiences to assist newcomers in acquiring basic technical competency, as well as to help others within the community. However, many protocols typically offer a condensed form of experiences and do not provide the history and rationale underlying their development and evolution. Commonly, multiple protocols exist for a simple experiment, which reflects procedural subtleties and preferences among independent laboratories. Such divergence is not necessarily counterproductive, but can confuse newcomers and potentially fragment community research by making it difficult to compare and integrate resulting data. This is why standard protocols were proposed early to facilitate the comparison of microarray data (Brazma et al., 2001). Increasing dependence on commercial kits has also cultivated the false perception that the preservation and sharing of individual experiences via written protocols are not critical. A series of protocol books and journals, including video guides (e.g. Journal of Visualized Experiment), aims to address these problems. However, they typically focus on protocols many researchers use to sell publications and are not in a format that allows users to interact with authors to improve them.
In this proposed library, the authors of each wikified protocol, including key background information and references, manage it as editors. Users may post a question or suggestion, which automatically sends an email to prompt the authors to respond. The trail of questions and answers is captured for others to see. There are multiple mechanisms to reward the authors (e.g. recognizing each protocol as a publication with its use and citation to document its impacts). We should also encourage the authors of community journals to use and cite relevant protocols in their publications, which increases the visibility of the protocol library and streamlines the description of a particular protocol used in different publications. It will also benefit those who are seeking to reproduce methods described in publications, because, too often, they need to follow a long string of papers to construct the complete protocol.
Other potential projects include: (i) an online hub for pathogen diagnosis and identification by archiving validated sequence data, identification keys and diagnostic protocols for pathogens, and sources of reference cultures; (ii) an online GIS platform that tracks the discovery of major pathogens of global concern in temporal, geospatial and environmental contexts as a baseline for monitoring the movement of such pathogens; (iii) a MOOC-like community platform that provides foundational educational materials and professional development resources; and (iv) citizen science platforms that help empower and engage groups, such as Master Gardeners, growers associations and commercial growers, in monitoring emerging diseases/pests. Although these community infrastructures alone are not sufficient to address all current and future challenges facing the global plant pathology community, establishing a robust mechanism for global crowdsourcing through such initiatives and resulting experience will help us to meet future challenges more effectively and grow together.
I would like to thank my colleagues Pete Romaine, Jill Demers, Gary Moorman and Scott Isard for their insightful suggestions. Support from the US Department of Agriculture-Agriculture and Food Research Initiative (USDA-AFRI) (2005-35605-15393, 2008-55605-18773 and 2010-51181-21069), which has helped me to conduct a number of web-based projects, is acknowledged.