Netlogo icon
![netlogo icon netlogo icon](http://ccl.northwestern.edu/netlogo/3.1/docs/images/tutorial3/monitor-1.gif)
Not knowing what to do, I posed a question to the developer of the package (see Figure 3) and luckily got a response that the R ecosystem stores all the assigned objects in the RAM, but even with gigabytes of RAM, it struggles to write 96,822 patches over 8764 ticks on a spreadsheet. However, much to my surprise, I received error messages due to early terminations of failed HPC jobs. Here, R was used as a compiler to submit iterative NetLogo jobs on the HPC (High Performance Computing) cluster to improve the execution speed. The anonymous person, whose ID was JenB, kindly responded to me with a new set of codes, which helped me structure the codes more effectively.įigure 2 Raising a question about sending agents from one location to another in NetLogoĪnother example was about the errors I had encountered whilst I was running NetLogo with an R package “nlrx” (Salecker et al., 2019).
#Netlogo icon how to
Providing my NetLogo codes, I asked how to send an agent group from one location to the other. Figure 2 is a pragmatic example of posing questions to a wide range of developers on Stackoverflow – an online community of programmers to share and build codes. If there was a technical issue that can’t be solved, the problem should not be kept hidden, but rather be opened and solved together with experts online and offline. More practically, one can learn new ideas by helping each other. All of this can improve the quality of research.įigure 1 A screenshot of a Github page showing how open platforms can help other people to understand the outcomes step by step Agent-based models are mostly uploaded on (previously named OpenABM). Github, Gitlab), and social media to ask for advice. Even during the development, many developers share their work via online repositories (e.g. Also, people can comment if any errors or bugs are identified, or the model is not executing on their machine or may suggest alternative ways to tackle the same problem. This evidence enables scholars and technicians to visit the repository if they are interested in the source codes or outcomes. In doing so, my thesis shared the original data, the scripts with annotations that are downloadable and executable, and wiki pages to summarise the outcomes and interpretations (see Figure 1 for examples). I argue here that being transparent and honest about your model development strengthens the credibility of the research. I conclude my writing by accommodating personal experiences or other thoughts that might give more insights to the audience. During the writing, there are some screenshots taken from my PhD work (Shin, 2021). I wrote this short piece to openly discuss the benefits of conducting open research and suggest some points that ECRs should keep in mind. I somewhat empathised with their opinions, but at the same time, would insist that open research can gain more benefits than shame. A few answered that they were too embarrassed to share their codes online because their codes were not well coded enough. Having spoken to many early career researchers (ECRs) regarding the need for open science, specifically whether sharing codes is essential, the consensus was that it was not an essential component for their degree.
![netlogo icon netlogo icon](https://www.cdc.gov/pcd/issues/2016/images/15_0414_02.png)
Through my experience, there have been some challenges to learn other people’s work and replicate them to my project, but I found it more beneficial to share my problem and solutions for other people who may have encountered the same problem. Inspired by the talk, I redesigned my research beyond my word processer and hard disk to open repositories and social media. It was about the importance of reproducibility and replicability in science. In March 2017, in my first year of PhD, I attended a talk at the Microsoft Research Lab in Cambridge UK. Research Associate at the School of Geographical and Earth Sciences, University of Glasgow, UK