10 years ago I was all gung-ho about IPv6, but it's annoying at every level.
https://radar.offseq.com/threat/ipv4-mapped-ipv6-addresses-t...
If you want to handle two protocols, it is not unreasonable to use two sockets.
Having to deal with the separate socklen_t is mildly annoying, but you can just make a little struct that holds both.
This is a privileged view of someone whose ISP has enough money (or was around early enough) to get enough IPv4 addresses to assign one to every customer's WAN interface. Not everyone is so lucky.
A lot of folks get non-publicly-routable 100.64.0.0/10[1] on their WAN interface with no way to do hole punching because they're behind CG-NAT.
The lack of self awareness is appalling.
And the billion people in India? The billion in China? The billion on the continent of Africa? And even in the US:
> Our [American Indian] tribal network started out IPv6, but soon learned we had to somehow support IPv4 only traffic. It took almost 11 months in order to get a small amount of IPv4 addresses allocated for this use. In fact there were only enough addresses to cover maybe 1% of population. So we were forced to create a very expensive proxy/translation server in order to support this traffic.
> We learned a very expensive lesson. 71% of the IPv4 traffic we were supporting was from ROKU devices. 9% coming from DishNetwork & DirectTV satellite tuners, 11% from HomeSecurity cameras and systems, and remaining 9% we replaced extremely outdated Point of Sale(POS) equipment. So we cut ROKU some slack three years ago by spending a little over $300k just to support their devices.
* https://community.roku.com/t5/Features-settings-updates/It-s...
* Discussion: https://news.ycombinator.com/item?id=35047624
It's okay for the folks that got in early on the IPv4 address gold rush to tell them "fuck you, we got ours"?
PSINet/Cogent got 38/8 in 1994: did they invent it? Ford got 19/8 in 1995: how about them?
How many places and people/companies didn't have the ability to go to a RIR in the 1990s or 2000s and get an allocation because their local infrastructure (power, telecom) wasn't developed at the time? So because they got computers, fibre, smartphones later they're SOL?
curl URL Format curl -6 "[2001:db8::1]:8080/"
wget URL Format wget "[2001:db8::1]"
ssh Standard Login ssh user@2001:db8::1
ssh Specified Port ssh -p 2222 user@2001:db8::1
scp Remote Path scp file.txt [2001:db8::1]:/tmp/
rsync Remote Path rsync -av file.txt [2001:db8::1]:/tmp/
nc (netcat) Positional nc -6 2001:db8::1 80
telnet Positional telnet 2001:db8::1 80.
nmap Target nmap -6 2001:db8::1So any place that can only take addresses, not some sort of URL, should/can (of course, the program author can do whatever he/she wants!) accept addresses without delineation.
90% of HN replies are debating trivialities to attempt to skirt the point. The point is that ipv6 usability is worse. CLI usability is a good example.
There are many benefits to ipv6, but let’s at least admit the faults so they can be addressed and encourage further adoption.
It’s clumsier than ipv4. It’s unnecessary since NAT was invented. In practice IPv6 requires dual stack, which means twice as many firewalls, names and routes to manage — so 4x the debugging because you have 2 dimensions that can either be working or failing. Addresses are too long to remember, too clumsy to write, and after 30 years still don’t have consistent representation (how many colons and brackets?).
Look, IPv6 has some benefits, but until the usability is fixed, it will be another 30 years before it’s close to 95% adoption.