2020 ESIP Winter Meeting

The Council of Data Facilities (CDF) is committed to working with relevant agencies, professional associations, initiatives, and other complementary efforts to enable transformational science, innovative education, and informed public policy through increased coordination, collaboration, and innovation in the acquisition, curation, preservation, and dissemination of geoscience data, tools, models, and services. Existing and emerging geoscience data facilities – through the Council – are committed to serving as an effective foundation for EarthCube. The General Assembly meeting is open to the official representatives from all member data facilities, additional member organization personnel as desired by the members, as well as observers. How to

Agenda:
400-415 Welcome/introductions/sign-in - Danie415-430 High level Summary of OKN workshop - TBA
430-435 Updates on shared infrastructure - Kerstin, Danie

• White paper/report Sent to Eva, please send to your program managers

435-445 Update on COPDESS-Kerstin, Shelley
445-515 Update and next steps on P419-Doug, Adam
515-530 Progress on EC supplements for CCHDO and MagIC related to P418/P419 (GeoCODES)-Steve
530-550 Update from tech team EarthCube Office-Kenton McHenry
550-600 Summer topics - Danie

• Suggested Charter changes (to be voted on at july 2020)
• Announce  CDF exec elections in july 2020 - 2 co-chair and 3 at large positions

The Analytic Center Framework (ACF) is a concept to support scientific investigations with a harmonized collection of data from a wide range of sources and vantage points, tools and computational resources. Four recent NASA AIST competitive awards are focused on either ACFs or components which could feed into AQ ACF's. Previous projects have developed tools and improved the accessibility and usability of data for Air Quality analysis, and have tried to address issues related to inconsistent metadata, uncertainty quantification, interoperability among tools and computing resources and visualization to aid scientific investigation or applications. The format for this meeting will be a series of brief presentati.ons by invited speakers followed by a discussion. This generally follows the panel model How to Prepare for this Session: A link to a set of pre-read materials will be provided.

View Recording: https://youtu.be/fy4eoOfSbpo.

Takeaways

• Is there enough interest to start an Air Quality cluster? Yes!
• Technologists and scientists should both be involved in the cluster to ensure usability through stakeholder engagement
  
  

"Brains! Brains! Give us your brains!""
- Friendly neighbourhood machine minds
The collective knowledge in the ESIP community is immense and invaluable. During this session, we'd like to make sure that this knowledge drives the semantic technology (ontologies) being developed to move data with machine-readable knowledge in Earth and planetary science.
What we'll do:

In the first half hour of this session, we'll a) sketch out how and why we build ontologies and b) show you how to request that your knowledge gets added to ontologies (with nanocrediting).
We'll then have a 30-minute crowdsourcing jam session, during which participants can share their geoscience knowledge on the SWEET issue tracker. With a simple post, you can shape how the semantic layer will behave, making sure it does your field justice! Request content and share knowledge here: https://github.com/ESIPFed/sweet/issues
In the last, 30 minutes we'll take one request and demonstrate how we go about ""ontologising"" it in ENVO and how we link that to SWEET to create interoperable ontologies across the Earth and life sciences.

Come join us and help us shape the future of Geo-semantics!

Stuff you'll need:
A GitHub account available at https://github.com/
An ORCID (for nanocrediting your contributions) available at https://orcid.org How to Prepare for this Session:

Presentations:

View Recording: https://youtu.be/tr0coi5ZQvM

Takeaways

• Working toward a future (5-10 year goal) of making an open Earth & Space Science Foundry (from SWEET) similar to the OBO (Open Biological and Biomedical Ontology) Foundry. “Humans write queries”. Class definitions need to be machine-readable for interoperability, but must remain human-readable for authoring queries, ontology reuse, etc.
• Please feel free to add phenomena of interest to the SWEET https://github.com/ESIPFed/sweet/issues/ or ENVO https://github.com/EnvironmentOntology/envo/issues/ issue trackers. 
• At AGU they added a convention for changes to ontologies. Class level annotation convention. Can get now get textual defs from DBpedia for SWEET terms. See https://github.com/ESIPFed/sweet/wiki/SWEET-Class-Annotation-Convention
  
  

Learn what came out of two days of sprinting on how to advance APIs for analytics on coverages, arrays, and gridded data.

Join the conversation on the Gitter Channel or check out issues on Github.

Presentations:

View Recording: https://youtu.be/blWnKTlrgKY

Takeaways

• Overview of OGC API Sprint: What is distinction between Features and Coverages? Answer can be self-referential. Looking to gather feedback at https://github.com/opengeospatial/ogc_api_coverages
• Future API will be view of data. It will support feature view or coverage view as layers, functional render of underlying data.
• Idea is to provide single, holistic API that lets you daisy-chain any level of complexity by combining modular sub-APIs into workflow ‘processes’.
  
  

Session leaders will describe tools being developed to assess the learning resources in the ESIP"s Data Management Training Clearinghouse (DMTC) from the perspectives of both instructors and students. The feedback collected through these tools will aid in identifying and choosing resources appropriate for their needs. First efforts have been focused on using DataONE's EEVA tool to identify and adapt questions. Feedback will be requested from participants to help guide the content, look and feel of the tool. How to Prepare for this Session: Visiting ESIP's Data Management Training Clearinghouse (https://dmtclearinghouse.esipfed.org) would be helpful but not required for productive participation in the session.

Presentations:

View Recording: https://youtu.be/uc4tbjyePpI

Takeaways

The ESIP community and others in machine learning regularly lament the lack of labeled datasets, needed for certain classes of training algorithms. Generating accurate, useful labels is a hard problem, with no general automated solution in sight. Thus, labeling generally involves human effort, which is challenging because the volume of data needed for training can be very large.

Tools exist to help in labeling data. This session will demonstrate three labeling tools and associated processes:

• Image Labeler, a fast, scalable cloud-based tool to facilitate the rapid development of Earth science event databases, to aid in automated ML-based image classification, Rahul Ramachandran
• Labelimg, an open source graphical image annotation tool, https://github.com/tzutalin/labelImg, Ziheng Sun
• Bokeh, a Python based plotting and annotation tool set for building arbitrary labeling workflows, https://bokeh.org/, Jim Bednar

Time permitting, the session will conclude with a short discussion of thoughts and tradeoffs about the tools.

This session is followed by a hands-on workshop for using Labelimg and Bokeh. Please see the session abstract for "Hands on Labeling Workshop" for information on preparing for that workshop if you are interested in participating.

Presentations
https://doi.org/10.6084/m9.figshare.11629110.v1
https://doi.org/10.6084/m9.figshare.11591739.v1

View Recording: https://youtu.be/3ufBOoD3M1E

Takeaways

• Machine learning based classification applications require high-quality labelled data sets for both model training and evaluation. There are many existing tools for labeling images (including earth science data), but labeling tasks are very labor and time intensive.
• If the pre-built labeling tools don’t work for your problem, Anaconda provides a general-purpose labeler-building toolkit based on Bokeh for Python users; see https://examples.pyviz.org/ml_annotators/ml_annotators.html
• There is opportunity in combining partly automated, partly human labeling, to automate the easy cases while leaving the final call to a person. Currently not much tool support or good practices, hard to integrate.The art of avoiding extra work!

Intended as a follow on to the "Do You Have a Labeling Problem?" session and to get your feet wet, this working session is for people to experiment with two of the tools presented in that session, Labelimg and Bokeh. Presenters will provide some sample data for participants to work with. Attendees can also bring some of their own data to work with in the time remaining after the planned activities.

It would be best for workshop participants to preinstall Labelimg before coming to the session.   Regarding Bokeh, Anaconda is providing 25 accounts for workshop participants. (Thank you, Jim and Anaconda!).  Installing Bokeh is also an option.  Links for getting these tools are:

• Labelimg via https://github.com/tzutalin/labelImg#installation
• Bokeh as part of the HoloViz suite via http://holoviz.org/installation.html

Presentations

View Recording: https://youtu.be/y8NqTLgT8Ao

Takeaways