2016 ESIP Summer Meeting

Watch it live (and after the meeting) here -> http://commons.esipfed.org/node/9014

There is general agreement that managing your research data is important. There is good support for post docs or curation teams putting in the time to prepare the data for the repository. But, it’s not a common belief that everyone has a role to play in data management starting with the researcher. How can we change the value we place on data management. How can we embrace the importance of data management practices throughout the lifecycle. How can we convince our research community to love data management?  

The communication of scientific ideas and discoveries has not kept pace with the overall flow of information across the planet that we seek to better understand, and it is not moving fast enough to address the full breadth of the challenges facing our societies and our environment. On the other hand, we have entered an age where nearly all scientific thought and the execution of our experiments and trials can be encoded in software, opened for broad scrutiny, and accelerate our ability to make new discoveries. In the USGS, we are experimenting with a new method of organizing data, information, and knowledge toward a real time, iterative National Biodiversity Assessment. We borrowed heavily on the ideas and methods for traceability and transparency from the Global Change Information System developed by the US Global Change Research Program but are taking our Biogeographic Information System a few steps further to incorporate live data services and working scientific software. We are building on the idea of flexible Synthesis Compositions that assemble Analysis Packages to communicate scientific findings in multimodal ways from traditional GIS maps and reports to story maps and dynamic web presentations. We are testing ideas for how peer review and agency approvals work in an environment where data and software are the primary products of the scientific process and prose descriptions of scientific findings are parsed down to their simplest essence to explain to humans what is evident in the execution of software algorithms.

So, you’re an early career scientist, or maybe a data manager who’s tired of explaining WHY metadata is important, or a data curator who needs to know how to convert this brand-smackin’ new format into the more tried & true format required by your data repository – doesn’t matter who you are, you need some training about data management related issues!! How do you go about finding the training information you need quickly? Thanks to seed funding from the USGS Community for Data Integration, a Data Management Training working group under the ESIP Data Stewardship Committee is in the process of creating & implementing a moderated, crowd-sourced Clearinghouse. The descriptions within the Clearinghouse for the data management training materials will reside on the ESIP Commons and made widely discoverable through Internet search engines such as Google and Yahoo. This session will demonstrate how anyone with knowledge about data management training resources or events can input brief descriptive information about them, and also search for other resources or events. This brief demo is a preview for a longer breakout session in which we will provide more background, and invite you to give feedback on the interface and functionality of the fledgling resource.  

Big Data means more than just Volume; it can also show up as Big Variety. There are so many science datasets available, that searching for and finding the right one is becoming harder every day. One solution is to return search results with the most relevant ones at the top. Why so difficult? Well, dataset relevancy is a little different than the ordinary relevancy rankings one would use for web pages. Dataset versioning, temporal overlap, spatial overlap, download frequency are all potential means of presenting the datasets most likely to be useful to a user.

Each day, The Weather Company serves users in every country in the world, with forecast data for 2.2 Billion locations, to help them make decisions about their day, and provide government weather alerts for 34+ countries to provide actionable severe weather warnings to citizens.

Thousands of businesses, from aviation and insurance, to grocery stores and major retailers, rely on The Weather Company's services to plan and adjust based on how weather may affect their business and how to in turn, support their customers.

With a growing daily ingest of 100+ Terabytes of data like pollen, turbulence, radar, satellite imagery, traffic, personal weather stations, cars, and smartphone pressure data, The Weather Company helps citizens, businesses, and governments plan before, during, and after significant events to keep people and property safer and help companies understand the impact of weather on their business and take action.

What do a campus tour, a learning-robot, and a subtle murder plot have in common? They are all aspects of this lighthearted tale about a researcher's (nine-minute) journey to increasingly improve how data is communicated.

Details: Tripp Corbett, Christine White, and Sudhir Shrestha (and possibly other willing victims) will show – through an interactive multi-media tale - how to create a story map to share a message and its supporting research with a target community or the public. They will also attempt to stretch the boundaries on what it means to truly understand a dataset. For what we may lack in theatrics, we will make up in good mapping!

 

The Jupyter Notebook environment has become an excellent platform for presenting the utility of LP DAAC web services for a number of reasons. Jupyter notebooks are great for demonstrating real world science use cases; giving a step-by-step account of the data, processes, and services used in a scientific workflows. Notebooks can also bridge the gap between users and their knowledge of web services by providing recipes for web service execution. Finally, Jupyter notebooks not only provide the functionality to create beautiful and informative recipes with narratives that provide context, but also can be easily shared through a variety of mechanisms.