Things like "17th century", "1600s", or "1990s" are rarely exact dates, and almost always fuzzy. It really doesn't matter what the exact start and end day is. If you need exact dates then use exact dates.
A calendar change like this is a non-starter. A lot of disruption for no real purpose other than pleasing some pedantics.
Exactly. Historians often talk about things like "the long 18th century" running from 1688 (Britian's "Glorious Revolution") to 1815 (the defeat of Napoleon) because it makes sense culturally to have periods that don't exactly fit 100-year chunks.
I thought the article was going to argue against chunking ideas into centuries because it's an arbitrary, artificial construct superimposed on fluid human culture. I could get behind that, generally, while acknowledging that many academic pursuits need arbitrary bins other people understand for context. I did not expect to see arguments for stamping out the ambiguities in labellibg these arbitrary time chunks. Nerdy pub trivia aside, I don't see the utility of instantly recalling the absolute timeline of the American revotion in relation to the enlightenment. The 'why's— the relationships among the ideas— hold the answers. The 'when's just help with context. To my eye, the century count labels suit their purpose for colloquial usage and the precise years work fine for more specific things. Not everything has to be good at everything to be useful enough for something.
A calendar change like this is a non-starter. A lot of disruption for no real purpose other than pleasing some pedantics.