This remains an old concern of mine, and something I have never been able to wrap my head around. For instance, I have never understood why using your own sentences that you have crafted without citing yourself counts as plagiarism.
That being said- I am currently working on a paper and describing the data section, and the dataset is common enough that I see myself using it in the future for another paper. The efficient thing to do here would be to use the same description for both papers. This would, however, be considered as plagiarism. How can one reasonably be accused of plagiarism here? The data section is ancillary to both papers, is not based on a 'novel' idea, but is purely descriptive. To describe the data again using different words just for the sake of not plagiarizing from myself seems to be waste of time. Now imagine this happening to thousands of researchers over many years. It seems to me that productive hours that are wasted for such a strange, and honestly, unnecessary formality. Am I missing something here?