All posts (Page 82 of 122)
ZgotmplZ style=" object-fit: cover; object-position: center; height: 250px; width: 250px; margin-right: 15px"
/>
<p>
<em>Dr. Noam Ross [@noamross on Twitter] is a disease ecologist at <a href="https://www.ecohealthalliance.org/">EcoHealth Alliance</a> in NYC, as well as an editor for rOpenSci. Topics of discussion included Noam’s history with R and rOpenSci, working in a team-driven research environment, and inspirations for pushing research processes and rOpenSci projects in exciting new directions.</em>
</p>
</figure>
</div>
KO: What is your name, job title, and how long have you been using R?
...🔗
The general struggle
Something that will make life easier in the long-run can be the most difficult thing to do today. For coders, prioritising the long term may involve an overhaul of current practice and the learning of a new skill. This can be painful for a number of reasons:
- We have to admit to ourselves that we’ve been doing things inefficiently (i.e. wasting time). This makes us feel stupid and fosters a sense of missed opportunity: we could’ve done something cool with the time we’d have saved (e.g. vacation).
- We’re fond of our existing methods, probably because we’re used to them and they’ve served us pretty well thus far.
- The learning we have to do may seem beneath us: “I’m an expert in R, I’ve even written my own package. Surely ‘Version Control for Beginners’ wasn’t intended for the likes of me.” This kind of thought permits us to dismiss good ideas.
- We’re tired. Overhauling workflows and learning new skills takes energy and concentration. Today, we’ll go through the motions just like we did yesterday. And hey, no one complained about the work we did yesterday.
- We feel like the task is actually beyond us. This is almost never true (so long as we’re good at asking for help).
- In the context of a work environment, when we’re concentrated on learning, our apparent output drops (often to zero) until the learning period ends. A bad manager/supervisor who doesn’t appreciate the worth of taking the time to learn how to do things better may not forgive this short-term drop in apparent output.
Something that will make life easier in the long-run can be the most difficult thing to do today.
...
If you read my reflection #1 on rOpenSci Onboarding, then you know I see value in the Onboarding process. A LOT of value even. This post is about where that value lies.
This question has important corollaries which I will explore here based on my experience as a reviewer of bowerbird:
- How is a package peer reviewer’s time best spent?
- When is the best time in a software package’s life cycle to undertake peer review?
🔗
Doing a good job
As I’ve read the growing number of reflective posts by other Onboarding reviewers, I’m struck similarities in our experiences. Mara Averick, Verena Haunschmid, and Charles Gray all place emphasis on reviewing from a “user’s” perspective. For Mara and Charles, this perspective was used to gain traction after initial nervousness about their ability to do a good job.
...Hydrology is a concept to unify statistics, data analysis and numerical models in order to understand and analyze the endless circulation of water between the earth and its atmosphere.
That’s a lot alike Data Science, isn’t it? Hydrologic Processes evolve in space and time, are extremely complex and we may never comprehend them. For this reason Hydrologists use models where their inputs and outputs are measurable variables: climatic and hydrologic data, land uses, vegetation coverage, soil type etc.
...🔗
Olfactory Coding
Detecting volatile chemicals and encoding these into neuronal activity is a vital task for all animals that is performed by their olfactory sensory systems. While these olfactory systems vary vastly between species regarding their numerical complexity, they are amazingly similar in their general structure. The periphery of olfactory systems consists of different classes of olfactory sensory neurons (OSN). In mammals, OSNs are located in the nose, in insects, OSNs are located on the antenna. OSN classes are tuned to defined but overlapping sets of odors. Thus a single odor usually elicits differential responses across an ensemble of OSNs. This ensemble code is able to encode thousands of odors, even in comparably simple olfactory systems.
...
Working together to push science forward
Happy rOpenSci users can be found at
