Lot of half-baked ideas floating around in the early years of the commercial internet, but the Java thing held up better.
Then you prefetch to fix that problem, and now some slow calls you hadn’t gotten to the bottom of suddenly aren’t.
/s of course
I'm sure some braindead software out there attempts to retrieve namespace URIs, but it would surely be a drop in the bucket compared to traffic for schemas/DTDs (which are intended to be retrieved).