Unidata Policy Committee Meeting Summary

10-11 October 2013
Boulder, Colorado

Action Items
Committee Members
Acronym List

Committee Members:

Steven Businger, University of Hawaii, Chair (outgoing)
Bill Gallus, Iowa State University, Chair (incoming)
Dave Dempsey, San Francisco State University
Rob Fovell, University of California, Los Angeles
Kevin Kloesel, University of Oklahoma (outgoing)
Kerstin Lehnert, Columbia University (outgoing) (absent)
Brian Mapes, University of Miami (incoming)
Lynn McMurdie, University of Washington
Michael Piaseki, City College of New York
Sepi Yalda, Millersville University (incoming)
George Young, Pennsylvania State University (incoming)

Users Committee Representative:
Kevin Tyle, University at Albany

UCAR/NCAR Representatives:
Greg Byrd, UCAR/Comet
Emily CoBabe-Ammann, Director, UCAR Community Programs
Vanda Grubišic NCAR/EOL
Steve Worley, NCAR/CISL

Sponsor Representative:
Bernard Grant, NSF (absent)

NOAA Representative:
Leroy Spayd (absent)

NASA Representative:
Christopher Lynnes (absent)

(Note: Government representatives were not allowed to attend due to government shutdown.)

UPC Staff Attending

John Caron
Julien Chastang
Ethan Davis
Doug Dirks
Ben Domenico
Ginger Emery
Ward Fisher
Yuan Ho
Michael James

Linda Miller
Terry Mitchell
Jennifer Oxelson
Mohan Ramamurthy
Russ Rew
Sheri Ruscetta
Mike Schmidt
Tom Yoksas
Meeting agenda

Thursday 10 October 2013

Administrative Items

Next meeting tentatively set for 20-21 May 2014. The location for the meeting is still to be determined; traditionally Spring meetings are hosted at a committee member's institution (including agency representatives). The University of Washington and SF State University were mentioned as possibilities.

Users Committee report - Kevin Tyle

Kevin provided highlights from the Fall Users Committee meeting, which engendered several side discussions.

During Kevin's description of new progressive display features in the IDV, a discussion of polar orbiting satellite data ensued:


Kevin noted that the new User Committee's student representative (Kim Hoogewind) is enthusiastic about Python, which led to additional comments:

Director's Report - Mohan

Mohan's slides.

Some additional comments during and after Mohan's presentation:

Budget Report - Terry Mitchell

Attrition and vacancies played a big part in the budget for the Unidata 2013 award. With labor costs and the associated overhead accounting for 87% of the total budget, the long stretch of vacant positions in FY12 & FY13, although a strain on the program, allowed us to build up a reserve, and helped defray the affects of an increase in both the benefit and indirect rates in two level funded years. Along with new funds for the second and third year of the OPULS project and the third year of funding on ACADIS, we were able to leverage our surplus of funds to continue two term positions and fill two vacant positions. And with the recently awarded EarthCube proposals, we are also able to add an additional position to the tech group. Over the course of the Unidata 2013 award, we've maintained a balanced budget in the core which has positioned us well to face a potentially challenging financial climate starting in FY14. Although we're in better financial shape than a lot of other UCP programs, we're still hoping for a modest CODB increase in the core for FY14 in anticipation of another increase in both the benefits and indirect rates for the new fiscal year. As always, we will strive to increase non core funding with more proposals and white papers to various funding opportunities and we'll continue to leverage our resources with new projects and funding as appropriate. The recent government shutdown on top of the sequestration makes the financial outlook for FY14 and beyond even more uncertain, but with the surplus created over the 2013 award period, we're in a good financial position to start tackling the goals set forth in the Unidata 2018 proposal.

Open Discussion

Dave D followed up on training. "Flipping the classroom" is a technique that asks students to study some materials first, then using classroom time for collaborative projects. See Khan academy video concepts. Unidata training is mostly a high-bandwidth transfer of information that could be done beforehand, allowing workshop time to be used in other ways. Online does not replace classroom time, but allows it to be used for more one-on-one or collaborative exercises. At Carleton college workshops, students are required to come to class having done the homework first.

There was general agreement that this is an interesting approach, but concern over whether professionals attending training workshops would do the pre-class homework.

Lunch
UCAR Community Programs vision for the future - Emily CoBabe-Ammann, UCP Director

Emily described recent Climate Services meetings in which UCAR was proposed as a "bridge between data and stakeholders," although it was not clear what exact form this might take. Participants in those meetings suggested Unidata as a model ("Unidata for climate data").

Some participants in the UCAR meetings advocated a "climate data center" with data storage/archival facilities. There was also impetus for a set of "community-driven best climate data practices."

Lynn M wondered what the criteria are for "good" climate data, and noted that no one has really defined "climate data," although at times it looks a lot like "weather data."

Business development - Mohan

Mohan described UCAR investigations into creating more commercial partnerships as a way of diversifying funding sources while maintaining core missions of supporting research and education. Some comments:

Committee name change discussion

There has been a perception that the name "policy committee" makes it difficult to attract nominees. (There was some confirmation of this from committee members.) After some discussion of the qualities of the committee's activities (providing advice, long-term strategy, "vision"), several new committee names were proposed:

Steven B suggested that the committee mull these over and discuss again during the Friday session.

Discussion of Staff Status Reports

Steven B asked for questions/comments on staff status reports.

Equipment awards
George Y wondered how the number of proposals fluctuates?
In 2013 there were 6 proposals, one year there were 18.
Mohan asked how we can encourage community members to write quality proposals?
George Y noted that Penn State has IT staff to do much of the proposal writing work.

Travel
Dave D recalled previous was talk about not being able to fund face-to-face meetings.
Mohan said that Unidata has budgeted for face-to-face meetings in current proposal, and has a strong desire to continue them.
Bill G added that Teleconferences are especially bad for a long meeting.

AWIPS II
Bill G asked for an honest perspective: in two years will we in universities be using this in preference to GEMPAK?
Michael J said that things tend to go wrong in development, but a WFO must run for 30 days without failing or test must be redone. Will universities find this as useful as GEMPAK? GEMPAK will be an add-on to AWIPS II, but gui programs will be replaced by CAVE. Yes, in the future, but there are lots of unknowns.
Mohan stated that there is almost no chance of NWS walking away from AWIPS II, despite numerous delays.
Lynn M asked if universities should join beta test program now?
Dave D was troubled by the lack of innovation in the visualization. "A huge effort to recreate what we have." Also concerned about centralized server model.

Community Services
George Y asked what level of technical/program expertise do we expect from visitors to the Unidata web site. It is difficult to figure out what to look for and why. There is a big learning curve for an outreach web site. He suggested looking at the Marine Corps site as an example of a site that educated the new visitor well.
ACTION: Improve high-level webpage description of what Unidata provides including application function and linkages between applications.

IDV/RAMADDA
Sepi Y asked if there are any in-depth metrics to show where used across geosciences.
Yuan H replied that information we have is based on support questions coming to us; across earth science/solid earth, atmospheric.
Dave D asked if uncertainty in the future of Java had implications for IDV?
Julien C said that Java has tremendous momentum, and that Java3d on apple machines will update soon.
Dave D asked about a bridge to mobile devices
John C pointed out that most problems with Java have to do with running in the browser; it's very strong on server. Path for mobile devices may be "back to the browser" with HTML5 etc. which means Java doesn't need to be on the device. This would be a huge change and we should not be on the bleeding edge.
Michael P asked about the status of RAMADDA at Unidata?
Mohan said RAMADDA is actively developed, Unidata will be spending more resources to enhance integration with IDV.
Yuan H added that he is doing work to visualize Level III radar data directly in RAMADDA. The general development is directed at wider community.
Kevin T mentioned the Usercom action item to bring new RAMADDA sites into the list of federated sites.
Michael P asked whether Unidata could close RAMADDA documentation gaps in the open source environment?

International Activities
Kevin K asked if there are opportunities in other countries such as Asia
Tom Y pointed out that he participated in workshop in Columbia; exposed participants from around Latin America to LDM/IDV etc.
Michael J is planning to participate in a forecasting workshop in India next year - if funding is provided
Vanda G described the WMO's project to improve forecasting around Lake Victoria. Starting 2017; could potentially be a large field campaign

LDM
Multicast/virtual circuits University of Virgina project was funded

Discussion of Unidata's 2018 NSF proposal

The government shutdown means we don't have much new information on the progress of the proposal.

George Y observed that when Unidata started it was about looking at the recent past, now, and guessing at next ten minutes. Our whole system is designed around this type of analysis. Now we're shifting climateward — time series analysis and statistic analysis with long time records, not just in-the-moment. I don't need synoptic data, I need samples deep in time and (maybe) small in space. We're at a tipping point about what meteorology is, moving toward "decision-support meteorology". Need to be able to drill through time from past to future.
John C said that Unidata needs to know from you what are the data sets. We're putting up TDS on 6 Tb of NCDC /CFSR data.
George Y will come up with a small number of use-cases and send to John
Russ R pointed out that NetCDF4 chunking can help solve this problem

Break for Dinner

Friday, 11 October 2013

AWIPS II Demo - Michael James

Michael gave a thorough demonstration of the current AWIPS II version. Afterwards, the question of whether Unidata should continue devoting resources to AWIPS II arose. How many students go to work for the weather service? NWS will train students to use AWIPS II anyway. Perhaps it is more appropriate to think of AWIPS II as a replacement for GEMPAK? Can IDV provide the same set of products without the need for all the AWIPS II infrastructure? Some additional questions and comments:

Wrap up committee name change discussion

The committee voted to change the name to "Strategic Advisory Committee."

The Simple Science of Flight - Steven Businger

Steven gave a short seminar on the science of flight. The slides are here.

Wrap-up Discussion

There was a short reprise of the discussion about "flipping the classroom."

Dave D pointed out that current workshops require no preparation beforehand. Unidata should flip the classroom to take advantage of existing online resources. At least experiment with "flip classroom" strategies? Carlton College uses this technique; investigate how they provide incentives/expectations to prepare?
Ethan D pointed out that we have a wide range of experience coming in to Unidata training.
Dave D suggested using advance practice to level the range of experience.

ACTION 2: Dave Dempsey will prepare a whitepaper explaining concepts and giving resources/models pertaining to "flipping the classroom" and training.

Kevin T will ask Roberta Johnson in his department about this and will include a discussion in the Usercom agenda for spring.
Brian M wished for additional documentation in a "quick start" form, to inject people into the material faster.
Sepi Y asked what the learning outcomes would be? Are they different from what we get now?
Dave D felt that the objectives would be the same, but we could increase effectiveness.

Passing of the Gavel

Steven Businger officially passed the chairmanship of the committee to Bill Gallus.

Adjourn

Linda Miller and Doug Dirks
Community Services - Unidata
University Corporation for Atmospheric Research
P.O. Box 3000
Boulder, CO 80307-3000
303 497-8646 fax: 303 497-8690