Anne B. Reinertsen*
I wrote the article about fuzzytechie languaging with a wish to transcend a potential deadlock in digitalization and digital pedagogies, uncovering the lived experience of bias and procedural inequity. However, inequity, inequality and discrimination buried in institutional policies and procedures, technologies, or rules and regulations, are difficult to uncover in any direct manner. They are often unconscious and/or hidden in what we take for granted. Furthermore, the collection of information or data regarding minority progress in schools with respect to grades or other aspects of performance, or even micro-aggressions, is customarily done quantitatively, hence with methods that might obscure or even shield privilege embodied in majority-constructed policies. Digitalization and the algorithms we build from might also amplify such conditions and cement a deadlock even further. I indirectly therefore ask, how can we think adequately about the relation between knowledge, learning and ethics in educational systems and societies that are governed by algorithmic digital systems and objects endowed with agency?Further, how can we think adequately about the relation between ontology and language in educational systems and societies that are governed by such algorithmic systems and objects?
David Alexander Ells*, Christopher Mechefske and Yongjun Lai
DOI: 10.37421/2090-4886.2023.12.239
Wireless Sensor Networks (WSNs) can be used for machine condition monitoring to improve performance and safety. However, they present challenges with respect to energy and data management. This paper presents a novel low-power WSN and compares the performance of operating modes and data processing methods for vibration-based machine condition monitoring. The necessary software was developed to perform time and frequency analysis, and a data reduction method was proposed to reduce the data packet size. The performance of the WSN end node was then tested, and its energy consumption was compared for different operating modes. Testing showed that the end node was capable of performing basic vibration analysis. However, contrary to expectations and other reports, results showed that processing data locally to reduce the packet size consumed more energy than transmitting the raw vibration data. While the data packet was effectively reduced by 98.6 percent from 4096 bytes to 56 bytes, results showed that processing data locally consumed 8.8 to 21.4 percent more energy than transmitting the raw data.