So, I had both upvoted that guy's comment and flagged this entire thread (as in, the top-level post). To me, and I think to many other people, what is actually the core problem with HN is that it is being flooded with a ton of what is pretty much "spam": posts that are either repetitive (either to content from years ago, or even to content from last week) or content that are informing people of things that should be prerequisite knowledge.
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.
There is so much that "should" be prerequisite knowledge, that people simply can't know it all.
I admit that I know very little about DNS, to the point where I don't understand large parts of your post. I have a website, and I vaguely remember doing stuff to get the domain name to resolve, years ago. I know roughly that DNS translates domain names to IP addresses, but not much more. I'm okay with not knowing much more than that. I'm confident I can learn if necessary, but so far I've been learning the prerequisite knowledge of other fields.
If I was using godaddy, this thread would be helpful to me.
Unfortunately lack of hosting-related knowledge is indicative through and through here even to the actual startups out of YC. Is it too much to ask for someone to know the latest buzz about Rails but can't spell DNS? Good luck if you choose this as your crusade to educate.
You don't find it interesting that one of the largest DNS service providers has had a major outage rendering browsing many websites virtually impossible for several business hours? Not every site I am interested in using is maintained by readers of HN who should (perhaps) know better.
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.