Interview conducted by Joseph Smith, NASA EOSDIS Science Writer
NASA's Earth Science Data and Information System (ESDIS) Project manages the Earth science data systems of NASA's Earth Observing System Data and Information System (EOSDIS), which provides science data to a multidisciplinary and worldwide user community. This means that ESDIS is responsible for processing, archiving, and distributing the full suite of EOSDIS Earth science data products; providing center-unique tools for functions such as searching and subsetting of Earth science data; collecting metrics and user satisfaction data to learn how to continually improve the services it provides; ensuring scientists and the public have access to data to advance Earth system science; and promoting the interdisciplinary use of EOSDIS data, services, and tools to a broad range of existing and potential user communities. It’s an important job and as Deputy Manager for the ESDIS Project’s Science Systems Development Office, Dr. Justin L. Rice is actively involved in every part of it.
Rice’s affiliation with NASA began in 2004, when, as a sophomore at Jackson State University in Jackson, Mississippi, he took his first plane ride to Baltimore, Maryland, to serve as a NASA summer intern. Now, nearly 18 years later, Rice plays a leadership role in the development of several projects, including Earthdata Cloud, Cumulus, Openscapes, Harmony, and Earthdata Pub, that are critical to the ESDSIS and EOSDIS missions. He also serves as a mentor to NASA summer interns and regularly counsels college students on how to make the most of their experiences and skills. In the following wide-ranging interview, Rice discusses how he and his colleagues work to keep up with the ever-changing technological advances, NASA’s commitment to Open Science and expanding the Earth science data community, mentoring in a virtual environment, the benefits of following your curiosity, and the importance of communication.
The ESDIS Project and EOSDIS are in a period of transition with regard to the implementation of new technologies—artificial intelligence, machine learning, cloud computing. What are the biggest technological challenges facing ESDIS and EOSDIS?
The biggest challenge of these new technologies is that they’re new. They’re moving targets that are ever-changing. Given this, the dilemma then becomes: how do we leverage new technology today to best future-proof our science data systems? Another question that comes to mind is: how do we ensure things don’t change too fast once we commit to a using a particular technology as a core part of our data system operations? These types of questions remind me of the optical disc format competition between HD DVD and Blu-ray back in the early 2000s. That is to say, sometimes it’s clear that technology is moving in a certain direction. However, it’s not as clear which candidate will win out, because you’re right in the middle of the transition.
So, what is NASA’s approach?
Our approach is to keep our finger on the pulse of what is happening in the Earth science community. To do this, we engage with early adopters to learn about their scientific use cases and the types of technologies, data formats, and software stacks they employ. It’s key that we stay plugged in. A good example of a group that we’re following closely is the Pangeo community (i.e., a group of researchers who encourage the use of cloud platforms for in-place analysis of big geoscience data). The other part of our approach is to conduct trade studies and actively participate in the development of new standards. The Open Geospatial Consortium (OGC) plays a major role in developing open geospatial standards. As an OGC member, we provide our use cases and requirements before new standards are developed. This way we’re more proactive in our stance, and we influence the direction of where certain standards go.
How can ESDIS find new communities for NASA Earth science data?
To find new user communities, we meet them where they are. We give poster and oral presentations in a number of different venues, such as the American Geophysical Union conference and the American Meteorological Society annual meeting. By actively participating in these conferences, we’re able to highlight all our latest development activities and seek out early adopters. We also have our User Working Groups (UWGs), where we bring in users from the research community and have discussions with them so we can understand what their current challenges are and hear how they are partnering with other researchers to solve them. We create and disseminate the American Customer Satisfaction Index survey to our existing users annually, so we can receive feedback from the community and learn how to improve future services. If we keep our existing users happy, we increase our chances of gaining more users by word-of-mouth. So, in several aspects, we’re always reaching out to our existing users as well as leveraging different conferences, different venues, different working groups to find new users; especially those who are interested in transitioning their science workflows to the cloud.
Is there a role for Open Science to play in finding new users?
Yes, there is a big push for Open Science and along with that comes funding—and funding brings along new users. Therefore, when we say we want science to be more open, we just don’t stop there. We put our dollars behind that. For example, one of the things I do is review and approve cloud credit requests for workshops that will advance Open Science and use NASA data. I’m always eager to review requests – especially ones from new communities, as these tend to attract new users.
So, as the lead in developing the Cloud Primer tutorials, what is the best or most effective approach for training users on new techniques and datasets?
I’ve found that interactive tutorials are the most effective approach. The more interactive things are, the better. One thing that really resonates with our users is the use of Jupyter Notebooks. We use Jupyter Notebooks as an interactive and streamlined way to create, explain, execute, and share code that accesses and uses cloud-based data and services. They lower the barriers of entry into the cloud by abstracting users away from having to learn all the details of cloud-based infrastructure and services. This, in turn, allows our users to spend more time on science and less time on computer science.
How do you know when a software or system change has been effective?
For operations, A/B testing is an established way of determining the effectiveness of a software or system update. Have the old version and the new version of the software run concurrently, randomly assign users to each version, and collect metrics to determine the impact of the change. When it comes to the cloud, we’re still in the development phase. So, we use a different approach. When possible, we define metrics of success ahead of time, develop a prototype, and collect metrics to see how well we’re meeting our targets. Right now, we’re in the middle of identifying key performance indicators, meaning, we’re defining criteria that will help us measure performance. For example, our cloud data transformation services make it easier for users to egress only the data they need from the cloud. Comparing the amount of data leaving the cloud to the amount of data being accessed may give us insight on the effectiveness of these services.
Can these metrics for success always be predetermined?
Not always. Sometimes, it’s an evolving thing. We may come up with a key performance indicator and later realize that it’s not the best way of determining success. As with a lot of things we’re doing, we’re agile, meaning that we try things, we assess how well they go, and we adjust. It’s the same thing when it comes to developing metrics for success.
To switch gears a bit to more human-centered concerns, how do you go about facilitating discussion among all the groups ESDIS brings in to develop new applications and processes, etc., as I would imagine there are competing interests at times?
I’ve found that it’s very important to have trust when facilitating discussion among stakeholders that may have competing interests. Each party has to trust that you are for them, that you’re willing to listen to and learn from them, and that you’ll advocate on their behalf. If you don’t have trust, it’s going to be hard to move things forward. As you mentioned, there are multiple stakeholders—there are the Distributed Active Archive Centers (DAACs), there’s ESDIS, there’s NASA Headquarters—and everyone wants to get their job done. Sometimes there are conflicts when working together to develop enterprise level solutions. Stakeholders may have different requirements, different schedules, different system configurations, different science use cases, different user communities, etc. So, understanding that, and making it clear—that while there may not be a one-size-fits-all solution, working together to find the best solution given the constraints—is key. Being able to negotiate is key as well. Trading a nice to have today for a need to have tomorrow also helps parties to find common ground.
So, in terms of establishing that trust and rapport with people, especially in your work as a mentor, I imagine that working in a virtual environment might make that more difficult. Are there any specific strategies or approaches you use to help you establish that trust and rapport?
Indeed, there several limitations to working in a virtual environment. However, I find that honesty, authenticity, active listening, and clear communication are all key to establishing trust and rapport. When possible, I try to schedule virtual one-on-one time to connect with co-workers on a more personal level and learn about their story, hobbies, career goals, concerns, etc. I take the same approach in my work as a mentor. However, in this role, I am more attuned to providing day-to-day guidance and willing to share career and life lessons. All in all, I’ve found that these small things matter most in helping to bridge the virtual divide.
Going back to the beginning of your career, how did you become interested in working for NASA?
My story is a little different. I didn’t know a lot about NASA growing up. However, I did know that I was interested in understanding how electronic devices worked. It started out as a very small thing. I would take apart radios and other small devices and wonder, how did someone know how to put this together and get it to do what it’s doing? From there, my career journey unfolded just like a movie. I had no idea that I would be on the path I am now. So, NASA wasn’t really the goal. I was just curious about how things worked. A lot of times people say the sky is the limit. Others note how that this can be somewhat limiting because space is beyond the sky. I think the limit is much closer to home. For me, it’s the mind and one’s own experiences. Defining something for yourself based on what you’ve seen or what you know may be limiting. I understand that everyone needs goals and direction, but if I said, “the best I can do is do what the people around me are doing,” I would not be where I am today. When I reflect on my career journey, I realize that following my curiosity opened me up to a world I never thought I would be in — a world beyond who I thought I could be or what I thought I could do — a world beyond my imagination.
So, when did computer engineering and becoming an intern with NASA become something you wanted to do?
I discovered the engineering disciplines by reading encyclopedias and perusing various resources at the local library. From there, I decided that computer engineering was a good fit for my natural curiosity in electronic devices. This led me to majoring in computer engineering at Jackson State University. During my sophomore year, my friend told me about an opportunity, an internship program in Maryland called the Summer Institute in Engineering and Computer Applications (SIECA) program. It was managed through Bowie State University. I applied and later received an email informing me that I had been selected. I wasn’t sure: Would I be working at Bowie State? Would I be working at NASA? Should I even go? I’d never been on a plane, and I didn’t know anyone in Maryland. There were a lot of unknowns, but I ultimately decided to go. I later learned that Bowie State had a grant funded through NASA to hire underrepresented students as interns. In fact, a lot of Goddard’s diversity hires started as interns in the SIECA program.
Nonetheless, Jackson State University is a historically black college and university—a school that is majority minority. At NASA, however, there were students and professionals from all over the country. It was a very interesting and sometimes challenging experience to be somewhere so different from my background. Despite this, I decided to the make the best of the opportunity. I still had a passion for engineering and still loved to learn. Besides, everyone at home already considered me to be an astronaut. There was no turning back. This internship eventually led to another internship, which led to a co-op, which led to me going back to NASA for several summers.
What advice do you have for those computer engineers currently in college who are looking at NASA as the place to start a career?
My advice is the same for computer engineers and other majors alike. I would say to make sure that you are taking care of things at school, meaning that your grade point average is as high as it can be. Make sure that you seek out opportunities—internships—that’s very important. Know why you want to work at NASA and how your interests intersect with NASA’s missions and goals.
I often tell students to make sure they document internship experiences on their resumes and, if they don’t have any, be sure to include any relevant personal or classroom projects. Mention all the technical, leadership, and management skills learned from those experiences and specify how they relate to NASA opportunities. Be sure to spend time refining communication skills and learning how to work well in a team environment. These capabilities are oftentimes overlooked, but they’re just as important. Lastly, don’t be afraid to follow your curiosity and try new things.