Exercise idea for Interoperability

[MathewBiddle]

This exercise example is specific to interoperability of oceanographic data, if that is appropriate.

• Using the data server ERDDAP, along with the Climate and Forecast conventions, combine temperature data from multiple sources and make a profile or timeseries plot.

• Maybe building off this example jupyter notebook, but reducing some of the complexity with xarray and the various functions introduced.

[libcce]

Hi @MathewBiddle thanks and sorry for the slow reply! Yes, I think it could be an example in the lesson episode on interop. If you boiled it down, based on the outline for this lesson, where would you highlight it https://librarycarpentry.org/lc-fair-research/04-interoperable/index.html? I was going to guess but best to ask you. As you can probably see, that part of the lesson still needs work, so if you are inclined, it might also benefit from you sketching out some of the lesson to go around that example. If you are up to it!

[MathewBiddle]

Thanks for the feedback @libcce! As I was looking through that lesson, I felt like it was missing the 'so what' and 'where's the benefit of all this'. I guess I was thinking this would be a use-case as to why all of the elements in the lesson are important. This could fall into the objective Identify widely used metadata standards for research, including generic and discipline-focussed examples. But, maybe that topic intended to be covered in the introduction? https://librarycarpentry.org/lc-fair-research/01-introduction/index.html

In reality, from my experience, people need to see examples of the benefits of interoperability first, to capture their interest. Once you drive home how difficult it (typically) is to work with data not intended to be interoperable, it's easier for the learner to relate and recognize the benefits of the objectives laid out in the lesson.

Digging into the details of controlled vocabularies, machine vs human readability, and Linked Open Data might be cognitive overload. Additionally, at least in the marine data world, we haven't figured out how to appropriately identify and apply some of those. So, that might lead to confusion.

Being an ocean data manager, I could see a 'flavor' of this entire lesson be dedicated to FAIR Data and Software in the ocean sciences. Maybe some collaboration with https://carpentrieslab.github.io/python-aos-lesson/ would be possible?

[libcce]

Really good points @MathewBiddle! I'm at the AGU/Data Leadership, so I'd be interested in that ocean sciences flavored lesson :) Have you seen the oceanography example here https://librarycarpentry.org/Top-10-FAIR/ (also geoscience). Also, have you seen https://openscapes.github.io/series/? I like that they take a team approach based on the "Our path to better science" approach. Also, you mention, seeing examples first and the benefits, why this is important, then moving into the different aspects to get there might be helpful? I spent a good amount of time on the assessment module and wonder if that might be a better place to start https://librarycarpentry.org/lc-fair-research/07-assessment/index.html ? Another possibility... @selgebali has a great idea called FAIR Bytes which takes the approach of inviting a speaker to walk through important FAIR aspects of a topic, for instance, ocean sciences, and ultimately the talk/discussion gets filtered into a top/important things as a researcher you should know. That might also be of interest? If you ever want to chat, let me know. I'm definitely interested from the AGU perspective.