Proper database management in engineering - or how in their right mind uses spreadsheets (optimized for human readability) to store data? That was the worst thing I encountered until now. Zhao et. al (2017) managed to publish a dataset of tensile strength test as a word document (10 pages). I’m grateful that they published the dataset under a creative common license (CC-BY).
Data can be structured or unstructured. The majority of engineering datasets are structured, therefore a SQL database should be solution of choice to store datasets. They are easy to maintain and if needed we can dump the data to a portable format such a SQLite file, h5, csv and so on. Furthermore, I have problems to understand why so many engineers use painful software such as spreadsheets to do heavy analysis with (way too slow) tools they wrote in VBA years ago. If we use proper databases and suitable software, then such an analysis is much much faster and more importantly can be done with a few lines of code compared to hours of clicking through sub menus and waiting until it is finished. Moreover, a proper approach leads to figures that are worth to be published, because we can actually see what’s plotted and don’t have to play random pixel guessing.
More general what is wrong with many engineers? Frequently, I encounter large sets of equations written with a word processor or a well-known presentation tool. That seems like a huge waste of time and looks shitty compared to LaTeX.