it still frustrates me to no end that the only reason why residential networking is complicated is IPv4
and IPv6 is almost three decades old
instead of all this bullshit with port forwarding, NAT punching, and other shit, you should be able to just know your own public IP address and send it to someone else to connect directly to you, but you can't on IPv4. instead you have to deal with bullshit
and just to explicitly point it out: IPv6 was standardised in 1998. it has been literally 27 years since IPv6 was fully described and implementable and people still aren't using it everywhere
shitheads like Microsoft offer only-IPv4 endpoints for sites like GitHub, but even then, people can use IPv4-over-IPv6 to communicate with them. everyone should have an IPv6 address and IPv4 should be a relic of the past but for some reason that's not the case despite some countries only being able to have tens of IPv4 addresses total
@clarfonthey as sad as it is I think it might be as simple as "my IPv4 address is 10 digits and I can remember it, and I cannot remember my IPv6 address for anything, specifically because it is a larger addressable space" (edit: I can't count either apparently)
@apisashla @clarfonthey I'm not convinced that that's a meaningful factor, to be honest - entirely too many people have dynamic IPs for "remembering your IP" to be viable at any real scale, for starters.
Not to mention that there's a much more obvious reason why IPv6 is lagging: because right now IPv4 addresses are a valuable asset that networking companies have a lot of, and the scarcity + needing them is what makes them valuable, so there's a direct monetary loss associated with broader IPv6 deployment.
@operand @joepie91 @clarfonthey tbqh "ipv6 is complicated and I'm scared" is something I relate to even knowing it's entirely a skill issue on my part
@apisashla @operand @joepie91 tbh after learning how to manage ipv6 myself I can safely say this entirely a learned skill issue
ipv6 configuration is ridiculously easy and a lot of the issues come from people not documenting it and making it seem so much harder to use when it's not
@clarfonthey @apisashla @joepie91 fair. my most recent experience with ipv6 has mainly been trying to configure ipv6 for my home network with openwrt. i can imagine the ipv6 options are a bit bodged on in openwrt and a lot of things are just not in the place i expect from ipv4 but that doesn't make ipv6 bad obviously
i did get most things working the way i wanted them, only designating DHCP subranges for different SSIDs so i could restrict routing between them didnt work out (but isnt really necessary for the home)
@joepie91 @clarfonthey yeah this is probably the more accurate overall reason
@joepie91 @apisashla @clarfonthey Shrug it off all you want, but it's cited regularly as an argument against IPv6 — usually in the context of one's RFC 1918 addresses.
That said, I'm not saying the reason you offer is wrong; there are a multitude of factors making IPv6 less preferred for many.
(I will note that for some parties, there is a "direct monetary loss associated with" continuing to depend upon IPv4, too — the adoption graphs don't go "up and to the right" just due to good vibes. 😀)
@jima @joepie91 @apisashla it's cited regularly but it genuinely feels like a bad-faith argument cited to avoid discussing the more likely reasons which you and others mentioned: reducing access to the public web, selling addresses as a scarce resource, not wanting to spend money to upgrade, etc.
@clarfonthey @joepie91 @apisashla While I won't disagree with you that SOME OF IT is very likely bad faith arguments masking something else, I implore you to understand that a lot of people are lazy and don't want to learn new things. 🥲
@jima @clarfonthey @apisashla This does not match my experiences. Rather, I've found that when it *seems* like people "don't want to learn new things", there's usually a different underlying reason that's not being spoken aloud for one reason or another.
@joepie91 @clarfonthey @apisashla That may be, but in my 23 (OK, as of next month 😉) years of IPv6 advocacy, I have seen a lot of excuses.
And rarely does someone out and say "I don't want to learn new things" (although it does happen! particularly from those approaching retirement); usually it's couched as some vague dismissal that IPv6 will ever happen.
@jima @clarfonthey @apisashla Sure, but the same applies for vague dismissals of things - there usually *is* a reason, it's just not necessarily being shared with you.
@jima @joepie91 @apisashla I mean, sure, but that's not the kind of reason that gets used to avoid infrastructure upgrades from companies and governments
they don't care if you don't wanna learn new things; they will force you to learn if they want it, but they don't for other reasons
@clarfonthey @joepie91 @apisashla Agreed, that's USUALLY more of a personal-level resistance to IPv6, although I will caution that people who are personally opposed to IPv6 often get promoted to positions where they have the ability to steer the organization's posture with regard to IPv6 — which is to say, *away from* IPv6. (This is not a hypothetical statement. 🙄)
@clarfonthey @joepie91 @apisashla Examples of typical enterprise excuses are more like:
- none of our equipment supports it! (usually wrong)
- XYZ software doesn't support it! (because nobody is telling XYZ Software that they need the support)
- we can't find staff that know it! (self-fulfilling prophecy)
- we have enough IPv4 addresses! (did you bring enough for the rest of the internet?)
- none of our customers want/need it! (usually nobody is keeping track of these kinds of requests)
(...)
@clarfonthey @joepie91 @apisashla
- we had it enabled before and everything broke! (hard to say, but it may have been the scapegoat for other root causes)
Probably others, and I didn't even hit up ipv6excuses.com 😉
@joepie91 @apisashla @clarfonthey IPv4 limitations making hosting stuff at home way more difficult than it should be is also great news for companies that want you to pay for their cloud services. I don't know how much that matters but I feel like it's definitely in there somewhere.
Though from my brief forays IPv6 is definitely also very complex in some aspects. If we got rid of NAT people would probably be getting confused about their link-local addresses not being routable instead.