In this case, the problem is that DNS is a distributed database, and the idea that people are hosting websites with all of their DNS from a single provider that is at the same time acting as their registrar (whose only purpose in the DNS infrastructure is to mediate your ability to change your DNS servers and renew your account) is horrific: it means something horrendously wrong has occurred in this community.
Meanwhile, the comments here are just strange: people talking about "switching providers without downtime" when the whole point of how DNS works is that you can have arbitrarily many servers and thereby have multiple providers hosting your zones at once. To even have a webpage in the first place you had to setup DNS, and if you somehow skipped that step then you probably skipped tons of other important steps. :(
Reading this entire thread is thereby just depressing: this isn't some advanced corner case of A/B testing leading to improved knowledge of how to do pricing, this is web hosting 101. Yet, somehow, we have 308 upvotes and 238 comments that have been left about an outage of a single provider for the only component in the entire stack of a website where you almost have to go out of your way to not have fault tolerance.
Then, as opposed to trying to get this discussion out of the way as soon as possible, we are just being flooded with a combination of people claiming that this is important, and that those who disagree are being "snarky", combined with opportunistic bloggers submitting tutorials like the "GoDaddy Outage: How to Migrate to AWS Route 53" that was just posted.
Therefore, I will claim that it isn't drivebyacct2 that is indicative of a loss of HN quality: it is instead that somehow any of this was relevant in the first place, and that once posted it keeps spreading. I can understand people being interested when AWS or Heroku or even GitHub goes offline, but no one on Hacker News should care if GoDaddy DNS goes offline.
