9 ms·

It's been what, over twenty years now? If a technology is useful, you don't have to shame people into using it. At some point we will have to admit we made a

It's been what, over twenty years now? If a technology is useful, you don't have to shame people into using it. At some point we will have to admit we made a big mistake.

blueflow 2y ago
At some point IPv4 address space will be all allocated. Its just that many people don't feel the pain yet nor have the foresight to do any preparation. I'm all IPv6 and i got my first network segments that are IPv6 only. I got nothing to fear. And no NAT.
- this_user 2y ago
  Alarmists have been telling us for at least 15 years that the IPv4 space is about to be exhausted and how disastrous that will be. So far, nothing has happened.
  - blueflow 2y ago
    Hetzner charges extra for IPv4 allocation.
    - nasmorn 2y ago
      Which is the opposite of a catastrophe. It is just a tiny inconvenience
      - blueflow 2y ago
        Yet. Lets talk again in 2034.
    - steeleduncan 2y ago
      As does AWS - https://aws.amazon.com/blogs/aws/new-aws-public-ipv4-address-charge-public-ip-insights/ https://aws.amazon.com/blogs/aws/new-aws-public-ipv4-address...
  - Latty 2y ago
    Because now most new users are shoved behind CGNAT where they can't accept inbound connections. The internet as consumers know it becoming only a tool to browse websites and not an actual internet connection that can do say, peer-to-peer gaming, sucks.
  - nfeutry 2y ago
    The RIPE waiting list is at 400+ days. And getting bigger everyday.
  - skrause 2y ago
    > So far, nothing has happened. Everything the "alarmists" told us has already happened. IPv4 address space has already run out which now means that you often have to pay an ever increasing fee to use a public IPv4 address, if you get one at all (more and more people are sitting behing a CG-NAT). Public IPv6 blocks are completely free.
  - scbrg 2y ago
    You a climate change denier too, by any chance? The rhetoric is the same. Using the word "alarmist" to dismiss valid concerns, conveniently ignoring all the negative consequences that already have occurred and sprouting nonsense like "nothing has happened." What's happened is that everybody has been forced behind NAT, completely obliterating pretty much the whole point of Internet to begin with, and that addresses that could be essentially free are super expensive. But la la la nothing has happeeeeeeeened!
    - blueflow 2y ago
      You can overexaggerate a valid concern in an alarmistic manner. I'd argue the alarmistic, implausible reporting on climate change is the main cause of the scepticism.
      - jeffhuys 2y ago
        Being told 18 years ago in middle school that it’d be 50° outside 12 years ago has kept me skeptical during quite a few years. Overexaggerating only makes your point come across as weak, and in my case, made me think it was all bullshit for money. Just be honest.
        wredcoll 2y ago
        So 18 years ago "they" said it would be 50 degrees ... 12 years ago? What? And exactly who do you think is profiting?
      - scbrg 2y ago
        Yes. I'm sure it's the occasional exaggerated reporting in tabloid media that is at fault. The billions of dollars that have been funneled into extremely focused lobbying campaigns spreading FUD and disinformation for 50 years have probably only had marginal effect.
        blueflow 2y ago
        Guess why that didn't work, either! People are not that stupid.
  - Hizonner 2y ago
    It's not that it will be disastrous. It's that it already is disastrous right now. The functionality of the Internet is severely degraded by NAT, and has been for all of those "at least 15 years". The network can't deliver its most basic function, which is to let any node talk to any other node at will. As a result, the designs of huge numbers of higher-layer protocols have been warped and complicated, and the whole thing has contributed to the whole maelstrom of extreme commercial centralization pressures, too. Everybody who wasn't willing to hard cut over to IPv6 should have been forced off of the Internet 20 years ago.
    - wyager 2y ago
      The problem is that there are way fewer than 4 billion people who actually have anything worth connecting to their node for. 99% of internet users are strictly consumers, and a protocol change won't do anything about that. For them, there's little difference.
      - themoonisachees 2y ago
        This is patently false. They don't host services, sure, but the current landscape has all but shaped application design into assuming at least one party will be behind a NAT, which heavily centralizes traffic because it becomes impossible to make truly decentralized apps. Take a small look at the TURN and STUN, things that would have no business existing if all nodes could have an address. There are very valid reasons to want to send traffic to a node so called "out of the blue", and suggesting otherwise suggests you have never looked deeper than HTTP.
      - blueflow 2y ago
        Video conferencing and online games. A pretty big market.
  - prmoustache 2y ago
    ...and climate scientists have been ringing the alarm bell for more than 40 years and only now that it is kind of too late people timidly start to take them into account and do non significative changes in their life. And unless laws force them to, which will be hard to implement because individual countries do not want to hamper their own economic competitivity, the industries won't do anything more than the actual greenwashing. Chabges usually have to be implemented BEFORE the disasters happen. In a seismic zone, we should build earthquake resistant buildings now, not after the first major earthquake.
  - throw0101c 2y ago
    > Alarmists have been telling us for at least 15 years that the IPv4 space is about to be exhausted and how disastrous that will be. So far, nothing has happened. Tell that to millions of people stuck behind CG-NAT. Tell that to ISPs trying to serve communities on shoe-string budgets: > I work for a Native American tribe in the PNW. We scrambled to get the reservation reliable internet in the later part of 2019. We managed to cover most of the reservation with wi-max and wifi with a fiber back haul configuration. We are now slowly getting more stable and reliable fiber to the home(FttH) service installed to as many homes as we can, but it is slow process covering the mostly rural landscape doing all the work in house. > Our 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-2022-and-still-no-IPv6/m-p/854673/highlight/true#M35732 https://community.roku.com/t5/Features-settings-updates/It-s... * https://news.ycombinator.com/item?id=35047624 https://news.ycombinator.com/item?id=35047624 IPv4-only products and services is a problem right now.
- sambazi 2y ago
  following geoff huston, v4 address pressure/usage is actually declining. reasons range from cdn prevalence to cgnat and v6 "islands" but the old internet seems to be chugging along
- silverquiet 2y ago
  I vaguely remember "celebrating" IPv6 day I believe in 2011 when the last block was assigned from IANA (? - I can never remember which authority does what other than ARIN). Obviously IPv4 addresses can be recycled and NAT and SNI have greatly increased the capacity of a single IP. My company "only" has a pair of /27 blocks which we used to use quite carefully, but now acknowledge will practically never be a limit for our uses.
- SamuelAdams 2y ago
  IPv4 ranges will always be available for sale. Businesses will only start to migrate when the cost of holding those allocations outweighs the balance on their books. The lack of adoption by major players (GitHub, twitch, IMDb, etc) is a financial one, not a technical one.
- izacus 2y ago
  Well, it doesn't help that IPv6 is a technological equivalent of a christmas tree bill. Besides the address space expansion, it shoves down our throat a bunch of fundamental changes that noone asked for (and are not supported at all in most SOHO and prosumer router equipment). This is why it's much more likely that you'll continue seeing things like CGNAT instead of IPv6 - the IPv6 standard is just an utter engineering failure no matter how you look at it.
  - blueflow 2y ago
    NAT alone is definitely worse than IPv6.
    - izacus 2y ago
      [Citation needed] We've been using NAT and expanding it to decades which makes it an easy proof that it brings value. IPv6 is still barely adopted and barely available in two decades after having "ipv6 parties". What does that tell you about its value?
      - red_trumpet 2y ago
        40%-45% of Google users use IPv6[1]. I wouldn't call that "barely adopted". [1] https://www.google.com/intl/de/ipv6/statistics.html https://www.google.com/intl/de/ipv6/statistics.html
        sambazi 2y ago
        true, though i would guess that's mostly phones and it seems to be stagnating
        orangeboats 2y ago
        APNIC stats show that assigned IPv6 /32 prefixes are still growing YoY. Wouldn't call that stagnating. https://rex.apnic.net/statistics?allocationType=asn,ipv4,ipv6&from=1982-12-13&to=2024-04-15&interval=year&rir=apnic,afrinic,arin,lacnic,ripencc https://rex.apnic.net/statistics?allocationType=asn,ipv4,ipv...
- rottc0dd 2y ago
  https://en.wikipedia.org/wiki/List_of_countries_by_IPv4_address_allocation https://en.wikipedia.org/wiki/List_of_countries_by_IPv4_addr... I think it is to do with country wise allocation. I think India is seeing wide adaption of IPv6.
- ghusto 2y ago
  This go-to argument for why we must all start using IPv6 doesn't fly, at least for me. We ran out of IPv4 some time ago, and yet, here we still are.
  - orangeboats 2y ago
    It depends on what you're doing. I am using IPv6 in my home network because I simply couldn't host anything on my CGNATed IPv4 connectivity. Working around this CGNAT either costed money and/or required people to install additional client on their machine to holepunch through the CGNAT, which wasn't ideal since all I wanted was to host a simple blog. Guess why people are being forced to move to CGNAT? Precisely because we have run out of addresses. The dismissal attitude of HN towards IPv6 ("we ran out of IPv4 and everything still works!") will always baffle me. Because things are _failing_ to work.
  - throw0101c 2y ago
    > We ran out of IPv4 some time ago, and yet, here we still are. Yup. Spending money CG-NAT gear: > I work for a Native American tribe in the PNW. We scrambled to get the reservation reliable internet in the later part of 2019. We managed to cover most of the reservation with wi-max and wifi with a fiber back haul configuration. We are now slowly getting more stable and reliable fiber to the home(FttH) service installed to as many homes as we can, but it is slow process covering the mostly rural landscape doing all the work in house. > Our 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-2022-and-still-no-IPv6/m-p/854673/highlight/true#M35732 https://community.roku.com/t5/Features-settings-updates/It-s... * https://news.ycombinator.com/item?id=35047624 https://news.ycombinator.com/item?id=35047624 IPv4-only products and services is a problem right now.
- pif 2y ago
  > I'm all IPv6 and ... Good for you! I hope my thorough ignoring the issue will not hurt your feelings.
  - orangeboats 2y ago
    It won't hurt the feelings of the IPv6 people, since IPv6 clients can connect to IPv4 services just fine through NAT64. But IPv4 people won't be able to connect to IPv6 services no matter how, so don't go surprised_pikachu.jpg whenever someone decides they don't want to pay for the IPv4 tax: https://www.bottlecaps.de/index/ https://www.bottlecaps.de/index/ https://loopsofzen.uk/ https://loopsofzen.uk/
RiverCrochet 2y ago
Yes, NAT was a mistake.
- sambazi 2y ago
  that or it was an inevitable and powerful scaling technology that enables a double-digit billion number of devices to communicate using only 3B addresses
  - tenebrisalietum 2y ago
    So is IPv6.
    - sambazi 2y ago
      unquestionably
  - orangeboats 2y ago
    Scaling and NAT shouldn't be put together.
    - sambazi 2y ago
      yet here we are
      - throw0101c 2y ago
        Yes, here we are, spending many, many thousands of dollars unnecessary: > I work for a Native American tribe in the PNW. We scrambled to get the reservation reliable internet in the later part of 2019. We managed to cover most of the reservation with wi-max and wifi with a fiber back haul configuration. We are now slowly getting more stable and reliable fiber to the home(FttH) service installed to as many homes as we can, but it is slow process covering the mostly rural landscape doing all the work in house. > Our 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-2022-and-still-no-IPv6/m-p/854673/highlight/true#M35732 https://community.roku.com/t5/Features-settings-updates/It-s... * https://news.ycombinator.com/item?id=35047624 https://news.ycombinator.com/item?id=35047624
the_real_cher 2y ago
I dont know. I think IPV6 is fighting momentum more than it's fighting usefulness.
- rcxdude 2y ago
  This. It's a massive chicken-and-egg problem, and it's far more of a problem for those without the resources to influence things than the big influential organisations. What it means is that IPv6 doesn't get adopted, and where/when it is adopted, it's stymied by the fact that other parts of the network aren't adopting it, as well as generally lower usage meaning it far more frequently is subtly broken without being fixed. (residential ISPs I think are the biggest culprit: that means there's no customers on ipv6 so there's no reason for websites to adopt it, and it's still suicide to go ipv6-only. But of course with no ipv6-only websites there's no pressure for ISPs to support it, and with a lot of websites lacking ipv6 newer ISPs without the address space can't go IPv6-only and get any customers, so hacks like CGNAT are still mandatory. And so the kinda-working status quo makes moving to the properly-working system much harder).
  - pif 2y ago
    > It's a massive chicken-and-egg problem This was known since the incipit, still we moved on <facepalm/>. > And so the kinda-working status quo makes moving to the properly-working system much harder. It's called evolution: there is no "properly-working system" unless it works at every stage. If it is at disadvantage in an intermediate stage, it will never be selected.
    - the_real_cher 2y ago
      I don't know how accurate this is. I don't think there's any evidence that things have to be evolved in the tech world. paradigm shifts happen all the time.
      - pif 2y ago
        New paradigms happen from time to time. Paradigm shifts are rare, though. Look, all the good of OOP has not obliterated C. All the excellence of functional programming has not even made a dent in OOP adoption. Hell, COBOL is still running fine! Paradigm adoption is a matter of evolution: the fittest survives. When IPv6 will offer an immediate advantage to its users, people will switch to it. Until then, it is confined to those companies who are betting on it, other companies who have some spare money and prefer to spend it rather than have it taxed, and the stereotypical nerds who haven't anything more interesting to ponder about.
lesuorac 2y ago
> If a technology is useful, you don't have to shame people into using it. At some point we will have to admit we made a big mistake. I think you're looking at this the wrong way. "If it ain't broke, don't fix it!" is probably the correct way. Amazon is surely aware of IPv6 but if everybody who is on IPv6 can be tunneled through IPv4 to your storefront why bother changing the site? The new technology doesn't have to _just_ be better than the current one to cause adoption. It needs to be so much better that its worth adopting. Although judging from the fact that tiktok is on the list; it seems like new-er websites do not consider IPv6 to be better than IPv4.
throw0101c 2y ago
> It's been what, over twenty years now? If there wasn't a flag day for IP (1983-01-01) I'm sure there would have been people still insisting on using NCP. It took several decades' worth of experience to go from NCP to IP, and now that we've had experience with IPv4, it's worth moving towards IPv6. > If a technology is useful, you don't have to shame people into using it. It is useful for some people. It is not useful for others: like newly-started ISPs who can't get / afford IPv4 allocations for their millions (tens of, or hundreds of, in many APAC countries) of users and are left with paying for CG-NAT to reach many parts of the Internet. (And then their users have to double-NAT IPv4 to run their local networks, which breaks all sorts of things). The fact that the early adopters / 'legacy' folks in some parts of the world lucked out on allocations does not mean IPv4, with its limited 32-bit address space, is "useful".
1vuio0pswjnm7 2y ago
If the parent comment can stay on top, then I will be amazed. Best use of IPv6 I have seen is in cjdns. Where are all the other examples of peer-to-peer connectivity using IPv6. Personally I do not want peer-to-peer connection with websites doing surveillance, data collection, tracking and advertising, or other commercial entities. I do want connections with other internet users. But that cuts out so-called "tech" companies, intermediaries that have nothing to offer except middleman services. Hence there is a conflict of interest in that the benefits I could see for IPv6 (e.g., cut out the middleman) and the benefits of IPv6 to the advocates trying to shame people into adopting it (e.g., people who work for middlemen) are mutually exclusive.
hwbunny 2y ago
Yeah and instead of shaming peepz, provide useful guides. At the same time if a thing is overly complex, then it's a smell, that's best avoided.
- finnthehuman 2y ago
  > if a thing is overly complex So is ipv4, but most people have the benefit of learning it piecewise over a decade+. When I do networking root cause investigations involving v4, I still get glazed over looks from professionals that don’t specialize in net, but in that case they blame their own knowledge. If it’s v6 it’s the protocol’s fault they don’t know it.

It's been what, over twenty years now? If a technology is useful, you don't have to shame people into using it. At some point we will have to admit we made a big mistake.

blueflow 2y ago

At some point IPv4 address space will be all allocated. Its just that many people don't feel the pain yet nor have the foresight to do any preparation. I'm all IPv6 and i got my first network segments that are IPv6 only. I got nothing to fear. And no NAT.

this_user 2y ago
Alarmists have been telling us for at least 15 years that the IPv4 space is about to be exhausted and how disastrous that will be. So far, nothing has happened.
- blueflow 2y ago
  Hetzner charges extra for IPv4 allocation.
  - nasmorn 2y ago
    Which is the opposite of a catastrophe. It is just a tiny inconvenience
    - blueflow 2y ago
      Yet. Lets talk again in 2034.
  - steeleduncan 2y ago
    As does AWS - https://aws.amazon.com/blogs/aws/new-aws-public-ipv4-address-charge-public-ip-insights/ https://aws.amazon.com/blogs/aws/new-aws-public-ipv4-address...
- Latty 2y ago
  Because now most new users are shoved behind CGNAT where they can't accept inbound connections. The internet as consumers know it becoming only a tool to browse websites and not an actual internet connection that can do say, peer-to-peer gaming, sucks.
- nfeutry 2y ago
  The RIPE waiting list is at 400+ days. And getting bigger everyday.
- skrause 2y ago
  > So far, nothing has happened. Everything the "alarmists" told us has already happened. IPv4 address space has already run out which now means that you often have to pay an ever increasing fee to use a public IPv4 address, if you get one at all (more and more people are sitting behing a CG-NAT). Public IPv6 blocks are completely free.
- scbrg 2y ago
  You a climate change denier too, by any chance? The rhetoric is the same. Using the word "alarmist" to dismiss valid concerns, conveniently ignoring all the negative consequences that already have occurred and sprouting nonsense like "nothing has happened." What's happened is that everybody has been forced behind NAT, completely obliterating pretty much the whole point of Internet to begin with, and that addresses that could be essentially free are super expensive. But la la la nothing has happeeeeeeeened!
  - blueflow 2y ago
    You can overexaggerate a valid concern in an alarmistic manner. I'd argue the alarmistic, implausible reporting on climate change is the main cause of the scepticism.
    - jeffhuys 2y ago
      Being told 18 years ago in middle school that it’d be 50° outside 12 years ago has kept me skeptical during quite a few years. Overexaggerating only makes your point come across as weak, and in my case, made me think it was all bullshit for money. Just be honest.
      - wredcoll 2y ago
        So 18 years ago "they" said it would be 50 degrees ... 12 years ago? What? And exactly who do you think is profiting?
    - scbrg 2y ago
      Yes. I'm sure it's the occasional exaggerated reporting in tabloid media that is at fault. The billions of dollars that have been funneled into extremely focused lobbying campaigns spreading FUD and disinformation for 50 years have probably only had marginal effect.
      - blueflow 2y ago
        Guess why that didn't work, either! People are not that stupid.
- Hizonner 2y ago
  It's not that it will be disastrous. It's that it already is disastrous right now. The functionality of the Internet is severely degraded by NAT, and has been for all of those "at least 15 years". The network can't deliver its most basic function, which is to let any node talk to any other node at will. As a result, the designs of huge numbers of higher-layer protocols have been warped and complicated, and the whole thing has contributed to the whole maelstrom of extreme commercial centralization pressures, too. Everybody who wasn't willing to hard cut over to IPv6 should have been forced off of the Internet 20 years ago.
  - wyager 2y ago
    The problem is that there are way fewer than 4 billion people who actually have anything worth connecting to their node for. 99% of internet users are strictly consumers, and a protocol change won't do anything about that. For them, there's little difference.
    - themoonisachees 2y ago
      This is patently false. They don't host services, sure, but the current landscape has all but shaped application design into assuming at least one party will be behind a NAT, which heavily centralizes traffic because it becomes impossible to make truly decentralized apps. Take a small look at the TURN and STUN, things that would have no business existing if all nodes could have an address. There are very valid reasons to want to send traffic to a node so called "out of the blue", and suggesting otherwise suggests you have never looked deeper than HTTP.
    - blueflow 2y ago
      Video conferencing and online games. A pretty big market.
- prmoustache 2y ago
  ...and climate scientists have been ringing the alarm bell for more than 40 years and only now that it is kind of too late people timidly start to take them into account and do non significative changes in their life. And unless laws force them to, which will be hard to implement because individual countries do not want to hamper their own economic competitivity, the industries won't do anything more than the actual greenwashing. Chabges usually have to be implemented BEFORE the disasters happen. In a seismic zone, we should build earthquake resistant buildings now, not after the first major earthquake.
- throw0101c 2y ago
  > Alarmists have been telling us for at least 15 years that the IPv4 space is about to be exhausted and how disastrous that will be. So far, nothing has happened. Tell that to millions of people stuck behind CG-NAT. Tell that to ISPs trying to serve communities on shoe-string budgets: > I work for a Native American tribe in the PNW. We scrambled to get the reservation reliable internet in the later part of 2019. We managed to cover most of the reservation with wi-max and wifi with a fiber back haul configuration. We are now slowly getting more stable and reliable fiber to the home(FttH) service installed to as many homes as we can, but it is slow process covering the mostly rural landscape doing all the work in house. > Our 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-2022-and-still-no-IPv6/m-p/854673/highlight/true#M35732 https://community.roku.com/t5/Features-settings-updates/It-s... * https://news.ycombinator.com/item?id=35047624 https://news.ycombinator.com/item?id=35047624 IPv4-only products and services is a problem right now.
sambazi 2y ago
following geoff huston, v4 address pressure/usage is actually declining. reasons range from cdn prevalence to cgnat and v6 "islands" but the old internet seems to be chugging along
silverquiet 2y ago
I vaguely remember "celebrating" IPv6 day I believe in 2011 when the last block was assigned from IANA (? - I can never remember which authority does what other than ARIN). Obviously IPv4 addresses can be recycled and NAT and SNI have greatly increased the capacity of a single IP. My company "only" has a pair of /27 blocks which we used to use quite carefully, but now acknowledge will practically never be a limit for our uses.
SamuelAdams 2y ago
IPv4 ranges will always be available for sale. Businesses will only start to migrate when the cost of holding those allocations outweighs the balance on their books. The lack of adoption by major players (GitHub, twitch, IMDb, etc) is a financial one, not a technical one.
izacus 2y ago
Well, it doesn't help that IPv6 is a technological equivalent of a christmas tree bill. Besides the address space expansion, it shoves down our throat a bunch of fundamental changes that noone asked for (and are not supported at all in most SOHO and prosumer router equipment). This is why it's much more likely that you'll continue seeing things like CGNAT instead of IPv6 - the IPv6 standard is just an utter engineering failure no matter how you look at it.
- blueflow 2y ago
  NAT alone is definitely worse than IPv6.
  - izacus 2y ago
    [Citation needed] We've been using NAT and expanding it to decades which makes it an easy proof that it brings value. IPv6 is still barely adopted and barely available in two decades after having "ipv6 parties". What does that tell you about its value?
    - red_trumpet 2y ago
      40%-45% of Google users use IPv6[1]. I wouldn't call that "barely adopted". [1] https://www.google.com/intl/de/ipv6/statistics.html https://www.google.com/intl/de/ipv6/statistics.html
      - sambazi 2y ago
        true, though i would guess that's mostly phones and it seems to be stagnating
        orangeboats 2y ago
        APNIC stats show that assigned IPv6 /32 prefixes are still growing YoY. Wouldn't call that stagnating. https://rex.apnic.net/statistics?allocationType=asn,ipv4,ipv6&from=1982-12-13&to=2024-04-15&interval=year&rir=apnic,afrinic,arin,lacnic,ripencc https://rex.apnic.net/statistics?allocationType=asn,ipv4,ipv...
rottc0dd 2y ago
https://en.wikipedia.org/wiki/List_of_countries_by_IPv4_address_allocation https://en.wikipedia.org/wiki/List_of_countries_by_IPv4_addr... I think it is to do with country wise allocation. I think India is seeing wide adaption of IPv6.
ghusto 2y ago
This go-to argument for why we must all start using IPv6 doesn't fly, at least for me. We ran out of IPv4 some time ago, and yet, here we still are.
- orangeboats 2y ago
  It depends on what you're doing. I am using IPv6 in my home network because I simply couldn't host anything on my CGNATed IPv4 connectivity. Working around this CGNAT either costed money and/or required people to install additional client on their machine to holepunch through the CGNAT, which wasn't ideal since all I wanted was to host a simple blog. Guess why people are being forced to move to CGNAT? Precisely because we have run out of addresses. The dismissal attitude of HN towards IPv6 ("we ran out of IPv4 and everything still works!") will always baffle me. Because things are _failing_ to work.
- throw0101c 2y ago
  > We ran out of IPv4 some time ago, and yet, here we still are. Yup. Spending money CG-NAT gear: > I work for a Native American tribe in the PNW. We scrambled to get the reservation reliable internet in the later part of 2019. We managed to cover most of the reservation with wi-max and wifi with a fiber back haul configuration. We are now slowly getting more stable and reliable fiber to the home(FttH) service installed to as many homes as we can, but it is slow process covering the mostly rural landscape doing all the work in house. > Our 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-2022-and-still-no-IPv6/m-p/854673/highlight/true#M35732 https://community.roku.com/t5/Features-settings-updates/It-s... * https://news.ycombinator.com/item?id=35047624 https://news.ycombinator.com/item?id=35047624 IPv4-only products and services is a problem right now.
pif 2y ago
> I'm all IPv6 and ... Good for you! I hope my thorough ignoring the issue will not hurt your feelings.
- orangeboats 2y ago
  It won't hurt the feelings of the IPv6 people, since IPv6 clients can connect to IPv4 services just fine through NAT64. But IPv4 people won't be able to connect to IPv6 services no matter how, so don't go surprised_pikachu.jpg whenever someone decides they don't want to pay for the IPv4 tax: https://www.bottlecaps.de/index/ https://www.bottlecaps.de/index/ https://loopsofzen.uk/ https://loopsofzen.uk/

RiverCrochet 2y ago

Yes, NAT was a mistake.

sambazi 2y ago
that or it was an inevitable and powerful scaling technology that enables a double-digit billion number of devices to communicate using only 3B addresses
- tenebrisalietum 2y ago
  So is IPv6.
  - sambazi 2y ago
    unquestionably
- orangeboats 2y ago
  Scaling and NAT shouldn't be put together.
  - sambazi 2y ago
    yet here we are
    - throw0101c 2y ago
      Yes, here we are, spending many, many thousands of dollars unnecessary: > I work for a Native American tribe in the PNW. We scrambled to get the reservation reliable internet in the later part of 2019. We managed to cover most of the reservation with wi-max and wifi with a fiber back haul configuration. We are now slowly getting more stable and reliable fiber to the home(FttH) service installed to as many homes as we can, but it is slow process covering the mostly rural landscape doing all the work in house. > Our 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-2022-and-still-no-IPv6/m-p/854673/highlight/true#M35732 https://community.roku.com/t5/Features-settings-updates/It-s... * https://news.ycombinator.com/item?id=35047624 https://news.ycombinator.com/item?id=35047624

the_real_cher 2y ago

I dont know. I think IPV6 is fighting momentum more than it's fighting usefulness.

rcxdude 2y ago
This. It's a massive chicken-and-egg problem, and it's far more of a problem for those without the resources to influence things than the big influential organisations. What it means is that IPv6 doesn't get adopted, and where/when it is adopted, it's stymied by the fact that other parts of the network aren't adopting it, as well as generally lower usage meaning it far more frequently is subtly broken without being fixed. (residential ISPs I think are the biggest culprit: that means there's no customers on ipv6 so there's no reason for websites to adopt it, and it's still suicide to go ipv6-only. But of course with no ipv6-only websites there's no pressure for ISPs to support it, and with a lot of websites lacking ipv6 newer ISPs without the address space can't go IPv6-only and get any customers, so hacks like CGNAT are still mandatory. And so the kinda-working status quo makes moving to the properly-working system much harder).
- pif 2y ago
  > It's a massive chicken-and-egg problem This was known since the incipit, still we moved on <facepalm/>. > And so the kinda-working status quo makes moving to the properly-working system much harder. It's called evolution: there is no "properly-working system" unless it works at every stage. If it is at disadvantage in an intermediate stage, it will never be selected.
  - the_real_cher 2y ago
    I don't know how accurate this is. I don't think there's any evidence that things have to be evolved in the tech world. paradigm shifts happen all the time.
    - pif 2y ago
      New paradigms happen from time to time. Paradigm shifts are rare, though. Look, all the good of OOP has not obliterated C. All the excellence of functional programming has not even made a dent in OOP adoption. Hell, COBOL is still running fine! Paradigm adoption is a matter of evolution: the fittest survives. When IPv6 will offer an immediate advantage to its users, people will switch to it. Until then, it is confined to those companies who are betting on it, other companies who have some spare money and prefer to spend it rather than have it taxed, and the stereotypical nerds who haven't anything more interesting to ponder about.

lesuorac 2y ago

> If a technology is useful, you don't have to shame people into using it. At some point we will have to admit we made a big mistake. I think you're looking at this the wrong way. "If it ain't broke, don't fix it!" is probably the correct way. Amazon is surely aware of IPv6 but if everybody who is on IPv6 can be tunneled through IPv4 to your storefront why bother changing the site? The new technology doesn't have to _just_ be better than the current one to cause adoption. It needs to be so much better that its worth adopting. Although judging from the fact that tiktok is on the list; it seems like new-er websites do not consider IPv6 to be better than IPv4.

throw0101c 2y ago

> It's been what, over twenty years now? If there wasn't a flag day for IP (1983-01-01) I'm sure there would have been people still insisting on using NCP. It took several decades' worth of experience to go from NCP to IP, and now that we've had experience with IPv4, it's worth moving towards IPv6. > If a technology is useful, you don't have to shame people into using it. It is useful for some people. It is not useful for others: like newly-started ISPs who can't get / afford IPv4 allocations for their millions (tens of, or hundreds of, in many APAC countries) of users and are left with paying for CG-NAT to reach many parts of the Internet. (And then their users have to double-NAT IPv4 to run their local networks, which breaks all sorts of things). The fact that the early adopters / 'legacy' folks in some parts of the world lucked out on allocations does not mean IPv4, with its limited 32-bit address space, is "useful".

1vuio0pswjnm7 2y ago

If the parent comment can stay on top, then I will be amazed. Best use of IPv6 I have seen is in cjdns. Where are all the other examples of peer-to-peer connectivity using IPv6. Personally I do not want peer-to-peer connection with websites doing surveillance, data collection, tracking and advertising, or other commercial entities. I do want connections with other internet users. But that cuts out so-called "tech" companies, intermediaries that have nothing to offer except middleman services. Hence there is a conflict of interest in that the benefits I could see for IPv6 (e.g., cut out the middleman) and the benefits of IPv6 to the advocates trying to shame people into adopting it (e.g., people who work for middlemen) are mutually exclusive.

hwbunny 2y ago

Yeah and instead of shaming peepz, provide useful guides. At the same time if a thing is overly complex, then it's a smell, that's best avoided.

finnthehuman 2y ago
> if a thing is overly complex So is ipv4, but most people have the benefit of learning it piecewise over a decade+. When I do networking root cause investigations involving v4, I still get glazed over looks from professionals that don’t specialize in net, but in that case they blame their own knowledge. If it’s v6 it’s the protocol’s fault they don’t know it.