-
Notifications
You must be signed in to change notification settings - Fork 0
N1_Open science and Reproducibility: in Civil Engineering
This two week's discussion was mainly about learning the concept of open science and reproducibility and their impact in different field of research. We have had the preliminary idea about open science, the skills we need to or want to learn in the upcoming weeks, the restrictions and protocols we need to be aware of (i.e. FAIR, CARE) and follow while publishing any of our data or while using any of the published data. I have started to think about the potential of open science, to change how we do our research or how we value others publications in any sector. In the following sections, I will try to summarize my learning this week and my thinking with respect to my research area.
In general sense, we can define open science as some scientific research data/analysis method/results to be publicly accessible and transparent to all. However, to be more specific, "open science should achieve four goals:
- Transparency in experimental methodology, observation, and collection of data.
- Public availability and reusability of scientific data.
- Public accessibility and transparency of scientific communication.
- Using web-based tools to facilitate scientific collaboration." [1]
On the other hand, reproducibility can be defined as the capability to be reproduced or reusable and upgradable.
It seems too easy to have the idea of open science and reproducibility and to make it implementable. However, it is as difficult to implement this concept in research community where future research prospects is a concern, where the additional work to make the data public seems like "not so significant" in comparison to the work that can be done meanwhile. Therefore, we not only need to learn the tools and skills in the making of open science but also the consciousness about the approach needed to disseminate the concept of open science among the broader research community.
Open science is not just about the transparency it can bring about in the research community or making the analysis and results acceptable. The way I think about the prospects are as follows:
- Learning the contemporary tools and skills needed and having the capability to validate the tools being used
- Some research opportunity might be lost, however, open science brings about a lot more opportunities with excellent research ideas.
- Since the research community can use the expertise of others, the growth in a particular field can be huge.
The necessary tools may seem technical, however, it is also non-technical:
- Programming knowledge (e.g. C++, Python, Matlab, R, RStudio etc.)
- Collecting, storing and managing data properly
- Specific platform/software knowledge needed to handle data or to publish data/anaysis/results
- Capability to handle big data or to perform high computation (either by online platform or taking the advantage of HPC)
- Having the knowledge of the protocols needed to publish
- Better understanding to bring about the consciousness among the research community about open science
- Spreading the knowledge to the peers for a broader data science community
According to discussion FAIR means "A set of principles to ensure that data are shared in a way that enables & enhances reuse by humans and machines" [do not know how to cite this definition]
- Findable
- Accessible
- Interoperable
- Reusable
During these weeks I have found some of the useful websites related to civil engineering which I can utilize in future. Hence, I have listed them here:
- Data.gov [e.g. (Bridge Conditions, NYS Department of Transportation)]
- Data.world
Currently, in my research field, (i.e. structural engineering) openness is not prominent yet. There are a few open-source data or analysis available, however, lack of consciousness about publishing might restrict the reusability of those data/analysis. Most of the research work data/tools are reserved for the "closed" circle.
My research focus is particularly in the area of structural health monitoring and wireless sensor network. To elaborate, wireless sensors are installed in the structures (e.g. buildings, bridges etc.) to collect specific data, to analysis the data, and to monitor the health of the structure (i.e. whether the structure is in good condition or not). Therefore, I need to handle a variety of data and need to visualize them in an efficient way. I am currently involved in two research projects. One of them is DOT (Department of Transportation) project and the other one is an NSF project. In the later one, we are working on the battery management of a wireless sensor network. Talking to my advisor, I am planning to work on this project to publish the data and results and keep it up-to-date.