Greetings from the Aspect hackathon, where 22 scientists, mostly early career, are spending a week of intense development on this software package.
The hackathon completes up CIG’16, two weeks of diverse activities for the Computational Infrastructure for Geodynamics. Here at HQ we are grateful to all the speakers, panelists, poster presenters, organizers, tutorial instructors, planning committee, and attendees, for making the meeting such a success. I’m personally grateful to the outstanding CIG staff for their hard work on this complex collection of events. There was a lot of conversation and collaboration between new friends and colleagues and those who have known each other for decades. This was a great wrap up for CIG-II and a launch of CIG-III. I look forward to continuing to work with all of you as we go forward.
We are looking forward to welcoming everyone to CIG’16, the meeting of the Computational Infrastructure for Geodynamics community. The workshop will feature tutorials, talks, panels, a discussion of future directions, posters, lightening talks, a field trip, a field trip, and time for sharing your science informally with your colleagues.
The weather in Davis will be warm and dry, so pack your sunscreen and a hat. You can follow the meeting on Twitter at #CIG16UCD.
I am pleased to announce that CIG-III will be supported by the NSF’s Geoinformatics program, with support from Advanced Cyberinfrastructure.
Over the last decade our geodynamics community has grown and developed, pushed the scientific and computational frontiers of geophysics, and become a model for other scientific communities. We’ve seen new models of geologic processes and Earth’s systems from the core to the surface, and established and sustained partnerships between computational scientists and geophysicists. We’ve established best practices for scientific software development, enabled effective use of some of the world’s fastest computers, and supported the next generation of computational geophysicists to carry out their research vision. Our community has collaborated with other organizations, agencies, and international partners on workshops, tutorials, and software development. It’s a great time for computational geophysics, and I look forward to what the next 5 years holds.
The upcoming CIG ’16 meeting will be a kickoff event for the next 5 years of CIG. I look forward to seeing many of you in Davis in June!
The CIG community has been researching ways to attribute scientific software to its creators and ways to cite software, through an EAGER project, Software Attribution for Geodynamics Applications (SAGA).
On October 30, together with the UC Davis Innovating Communication in Scholarship project, we co-hosted a one-day workshop featuring speakers from industry, publishers, scientists, the CIG community, and more, to look at the issue.
The event was webcast and will be made available online shortly. In the meantime, if you’re interested in how this looked to the online community, UC Davis evolutionary biologist and open science advocate Jonathan Eisen has storified the workshop, using its twitter hashtag #SoftCiteUCD.
Here’s what happens when more than 20 scientist user-developers get together for 9 days of intensive coding at an isolated location:
- Approximately 300 commits merged into Aspect
- 6706 lines of code added
- More than 120 pull requests merged
- Approximately 45 issues opened, with many then closed again following some of the commits.
- Lots of learning, new collaborations, and new friendships
Most impressive was the amount of learning that went on. Anyone who teaches knows that one of the most effective ways to learn new material is to teach it to someone else, and that happened a lot over the 9 days of intense work. Most of the participants were graduate, students or postdocs, and they take that new knowledge back home with them. Thanks to Bodega Marine Laboratory for providing us a great working and living environment that was just isolated enough.
Which other CIG codes would benefit from a hackathon type approach?
This month, we are preparing for the 2nd Aspect Hackathon, which will bring more than 20 students, postdocs, faculty, and staff together for 10 days to intensively develop the Aspect AMR code for mantle and lithospheric dynamics. The hackathon serves a second purpose: while contributing to the code, the participants gain proficiency in scientific software best practices and in development strategies for Aspect in particular. This diverse group of mainly early-career scholars emerge more skilled numerical modelers, with deeper connections to computational science. Because of this broader impact on training, hackathons are likely to be increasingly popular for CIG in the coming years.
Training is just one of CIG’s broader impacts, created by you, the CIG community. CIG’s infrastructure and the codes we distribute help scientists sustain and increase the impact of their scientific software; software distributed through CIG is used for an impressive range of scientific research and training, including the cryosphere and of other planets. Partnerships with organizations including IRIS and CIDER, have opened up the use of geodynamical modeling to new communities. Sustained cross-disciplinary collaborations between geodynamicists, computational scientists, mathematicians, and information scientists are changing the culture of scientific software development for geodynamics. International collaboration opens up new opportunities for early career CIG scientists. Let us know: what is your favorite broader impact of CIG?
Oberkampf and Roy’s book on “Verification & Validation in Scientific Computing” discusses “five stages of maturity of predictive capability”, drawing on a 1986 NRC report on “Current Capabilities and Future Directions in Computational Fluid Dynamics”. That’s not so current anymore, but this figure from the NRC report is interesting.
Stages I and II (partly) reflect where the geodynamics community was 10 years ago, before the establishment of CIG, while Stages IV and V (partly) reflect where we are now. I see some differences from this engineering approach:
– For scientists, the scientific payoff starts in stage I, as we build knowledge.
– We are ‘subject to surprise’ at every stage.
– Working on natural systems means we usually lack supporting experimental comparisons.
– We’ve progressed from hero codes, written by one or a small number of people, to community-supported codes.
The figure also presents a linear progress towards maturity. In practice, there is a feedback loop: as scientists learn to use a code effectively, we discover the new physics that we want to add to that code, so there should be an arrow looping partially back to stage 1. I suppose that means we’ll never fully mature.