Going off on a tangent… I'm not sure whether it's improved or deteriorated IMO, myself. I'm pretty sure I have an opnion, but I'm honestly uncertain what it is.

Going off on a tangent… I'm not sure whether it's improved or deteriorated IMO, myself. I'm pretty sure I have an opnion, but I'm honestly uncertain what it is. It seems to change by the day. I wrote https://rant.gulbrandsen.priv.no/programming/suckiness-today https://rant.gulbrandsen.priv.no/programming/suckiness-today on a frustrated day long ago, but maybe I'd rather make the opposite point today: If modern code reuse leads to less knowledge of what the (third-paryt) code does, it also brings more function, which is arguably an increase in quality. Take code that uses images an an example: So the development team (and the documentation team, if any) may not know which image formats support alpha channels. But supporting more formats means there's a higher chance that any random image used by an end-user works. Is that overall good or bad? When I wrote that posting I was frustrated, today I might consider the overall extension of functionality to be the most important part. I think. Perhaps.