Hacker News Reader

Comment by swinglock | Hacker News Reader

IPv4, the original decentralized Internet currency. Just get on IPv6 already! These numbers wouldn't be worth enough to fight over if we just upgrade.

View on HN

It's not a currency, it's a resource. Mistaking resources for a currency results in bad outcomes, which is why e.g. "gold" was a bad idea as a world currency. We need gold to make stuff, it's a resource.

In fact v4 addresses are not even just an ordinary resource, they're very like Real Estate aka Real Property, or "land". There's a finite amount of it, we can't do anything about that - to some extent it's effectively "occupied" already in many cases, so an imagined "property right" doesn't trump that in practice, and it's very much not fungible.

> decentralized

The internet might be entirely decentralized, but IP address allocations are very centralized.

> AFRINIC is one of the world's five regional internet registries, or RIRs: AFRINIC; Asia-Pacific's APNIC; North America’s ARIN; Europe’s RIPE; and Latin America’s LACNIC. These are the organizations that manage, allocate, and track the use of internet number resources – IP addresses and autonomous system numbers.

Which one is the central authority?

The one assigning numbers to the five regional authorities. That would be IANA.

And even if there was no IANA, the 5 regional authorities would still very much be centralized.

The Number Resources Organization is the body which represents the regional internet registries. You could maybe consider that to be the central authority, but when you're talking IP addresses, IANA is the central authority.

5 regional authorities is still _centralized_

I agree, but I can understand wanting it narrowed down to a single org.

> just upgrade.

I wish it was that easy.

Doesn't basically all modern hardware and software already support IPv6 just fine? Aren't we just waiting for more ISPs to start rolling it out and sysadmins to remove their stupid "disable IPv6" policies?

It's not just the ISPs. It's the organizations themselves. All the on-premises infrastructure whether it's an SMB or international corporates. It's not just out of laziness. It's a huge change that requires an enormous amount of time, money and efforts with little added value to their business. No one in the world would pay more to a company who invested in IPv6 migration while the service is exactly the same. No customer will prefer one vendor over another just because they recently migrated to IPv6. The value proposition of IPv6 was about scarcity of resources. The problem is there and mitigation cost is higher than workarounds. And as people mentioned, an IPv4 address is as cheap as a a dinner for two.

A lot of programmers are still writing v4-only code somehow. They'll write things like "socket(AF_INET, ...)" or "struct sockaddr_in address;" or "gethostbyname(...)" completely seriously and not seem to realize that any one of those things is an immediate red flag that your code is wrong and broken.

OSs are fine, but the software and systems that we build on top of them are often v4-only.

TBH, if it is not a requirement of the software, why would devs spend time on developing and testing IPv6? YAGNI kicks in way that point. If it's FOSS, it depends on the community. But if it is commercial, it's up to the specs.

I'm not even asking for that, I'm just asking for people to use the "new" APIs (they're older than some of the programmers making these mistakes), which are agnostic to address family. Use them even vaguely correctly and your code won't care about what address family the addresses are.

Having bugs in untested functionality is understandable, but using long-deprecated known-broken APIs, or using the current ones in ways that are clearly and obviously wrong, isn't.

(Also you should be using, and therefore testing, v6 at this point. But that's a separate discussion.)

Reply on news.ycombinator.com