Is there a reason why the researchers did this on live production BGP network rather than in a test environment, running their own BGP and FRR routers? Seems a bit haphazardly, as doing so first seems to have uncovered the issue.
Overall I agree with your message though, and of course things shouldn't stop working just because you use attributes reserved for development.
The researchers claimed that they had first tested it with a few different vendors' routers without any issues, but they had not included FRR in those tests [0]. There was a bit of discussion on the mailing list regarding just how many BGP implementations (and configurations) a responsible researcher is obligated to test with before exposing something new on the public Internet.
The correct answer is "All of them on your own network first".
Thou shalt not blow up another's network knowingly.
Blowing up another's network because you didn't test your theory in your own isolated lab space first counts as blowing up another's network knowingly. Because what the hell else did you think would happen?
Overall I agree with your message though, and of course things shouldn't stop working just because you use attributes reserved for development.