8 ms·

Is there any wonder that moving to IPv6 is delayed? Assuming an average $25/address, that's $400m/class A. Some of the organizations that own a class A: Appl

by rswail 6y ago

Is there any wonder that moving to IPv6 is delayed? Assuming an average $25/address, that's $400m/class A. Some of the organizations that own a class A: Apple, Ford, Softbank, Daimler, Prudential, Comcast. Softbank needs the money and the others wouldn't sniff at $300m+. Not sure how long the 100m odd addresses would last though.

manquer 6y ago
Apple and Comcast probably can use them. Also Comcast may have to invest more in NAT grade infra if they sell significant chunk of their Class A allocation so it may not make sense for them.
tzs 6y ago
Comcast started rolling out IPv6 across their network a long time ago. They hit 100% support on their broadband network in 2014. They run a lot of the own stuff on IPv6, such as on-demand video on their set-top boxes. I think any home modem/router that they support that did not support IPv6 hit EOL long ago and was replaced. There may be some class A owners who are trying to hold back IPv6 so they can make money off their class A, but Comcast isn't one of them. Apple has long had IPv6 support in their operating systems, so I don't think they have been trying to hold it back either. Ford, Softbank, Daimler, and Prudential? It is not immediately obvious to me if they could do anything to hold it back even if they wanted to.
- ArchOversight 6y ago
  Comcast ran out of RFC1918 addressing to manage all the devices on their network (think cable modems/set top boxes/routers/everything) and had a massively segmented network setup. They rolled out IPv6 because it flattened that network again, and instead of worrying about being in the right segment for reachability everything now gets a globally unique IP address. It massively simplified their internal management.
- DaiPlusPlus 6y ago
  > Apple has long had IPv6 support in their operating systems, so I don't think they have been trying to hold it back either. And Microsoft has supported IPv6 since Windows 2000: https://news.microsoft.com/2000/03/15/microsoft-announces-ipv6-technical-preview-for-windows-2000/ https://news.microsoft.com/2000/03/15/microsoft-announces-ip.... I understand that LTE networks are exclusively IPv6, ditto "5G". My frustration with IPv6 is that my home ISP (Wave Broadband's DOCSIS service) is still IPv4 and they still have no plans to roll-out IPv6 even though Wave's WaveG service has been IPv6 for as long as I can remember. I understand the cause of this is because Wave's network is incredibly hetereogenous because Wave's business-model was to expand by buying-up smaller regional cablecos but to retain their infrastructure (and talent) compared to Comcast's approach of asset-stripping (bad for employees, but good for customers...). I want IPv6 so I can stop having to faff around with NAT. I'd love to be able to RDP-in directly to my desktop at home without having to pay companies like TeamViewer and Splashtop. ---- Update: I found Wave's peering page's IPv6 status page and it remains unchanged since 2011 (and possibly since 2008). Wow. Compare: https://web.archive.org/web/20110102140910/http://as11404.net/ipv6.html https://web.archive.org/web/20110102140910/http://as11404.ne... Live: http://as11404.net/ipv6.html http://as11404.net/ipv6.html
  - criddell 6y ago
    > I want IPv6 so I can stop having to faff around with NAT. What makes you think your ISP will give you more than one IP even when they are no longer in short supply? If it's something that has value to you, it's something they will want to be paid for.
    - DaiPlusPlus 6y ago
      IPv6's has a fixed subnet size of 64-bits[1]. A router needs its own subnet, ergo everyone gets their own 64-bit address range. ISPs cannot give each customer a single IPv6 address because that would require each customer to have a single device connected directly to their router with no router in the CPE - basically like how dial-up used to work. There's no demand for a service like that. [1] See https://www.ripe.net/about-us/press-centre/understanding-ip-addressing https://www.ripe.net/about-us/press-centre/understanding-ip-... . I note that there is a lot of misinformation and badly-informed blog posts in the top search results for "ipv6 subnet size" like this one, ugh: https://www.crucial.com.au/blog/2011/04/15/ipv6-subnet-cheat-sheet-and-ipv6-cheat-sheet-reference/ https://www.crucial.com.au/blog/2011/04/15/ipv6-subnet-cheat... - which is wrong because IPv6 simply doesn't have 128-bit-sized subnet option (whereas you can have a 32-bit sized subnet in IPv4, which is how most residential ISPs/CPEs are configured).
- mike_d 6y ago
  Deploying IPv6 is nice, but it doesn't impact the value of IPv4 at all. Any internet service you want to be widely available will need to be dual-stack as long as we are alive.
  - wmf 6y ago
    If (eventually) each service needs an IPv4 address, that implies that there isn't a shortage and thus the value of addresses would be near zero.
  - Animats 6y ago
    Maybe not. Google says their traffic is now at 30% IPv6, and climbing at 5%/year.[1] Worldwide, Google is undercounting, because they are blocked in China. Google claims that China is 0.27% IPv6, but actually China is around 55%, according to APNIC. Policy goal in China is all IPv6 by 2025. [1] https://www.google.com/intl/en/ipv6/statistics.html https://www.google.com/intl/en/ipv6/statistics.html
    - mprovost 6y ago
      So it's taken 25 years to get to ~30% adoption. And a few years ago APNIC reported [0] that the rate was going down - I haven't found an updated report. In 2005 the US government mandated that all their computers be v6 connected by 2008. [1] This year they updated the mandate to 80% by 2025. [2] I learned networking in college in 1996 and I remember v6 coming up but so far I've never had to use it and chances are that I'll be able to retire still getting away with only working with v4. [0] https://blog.apnic.net/2018/05/21/what-drives-ipv6-deployment/ https://blog.apnic.net/2018/05/21/what-drives-ipv6-deploymen... [1] https://georgewbush-whitehouse.archives.gov/omb/memoranda/fy2005/m05-22.pdf https://georgewbush-whitehouse.archives.gov/omb/memoranda/fy... [2] https://www.cio.gov/assets/resources/internet-protocol-version6-draft.pdf https://www.cio.gov/assets/resources/internet-protocol-versi...
      - zajio1am 6y ago
        Considering that World IPv6 Launch was 2012-06-06 and nothing really happened before, it makes more sense to say that it has taken 8 years to get to ~30% adoption.
DaiPlusPlus 6y ago
> Assuming an average $25/address It'll be $50/address in a year, $100/address in 2022; I'm sure of it. Methinks now would be a very good time to invest^wspeculate in IPv4 addresses - I reckon it will be the next Bitcoin bubble-type event and likely even make it to the silly segment of TV news, with lampoonably inaccurate explanations of what an IPv4 address is to the general public.
- Cullinet 6y ago
  Exactly. What was the reason why do much money was required at this specific time? British businesses might have been natural customers for this address space, since the UK IDD code is 00 44. I would gladly pay $50 per IP for what is almost the only virgin address space going, if I had been offered. Diligent address space title management including some genuinely tough identification requirements, managed by any top tier corporate actions specialist, could have multiplied the value several times immediately, and then there would have been increased inflation rates applicable to the range precisely because of this management.
- nixgeek 6y ago
  Its been appreciating slowly for the last 5 years, I’ve seen deals in the $20-25 range throughout that time, and a few larger contiguous blocks going for a little more but not $50/ea. You can monitor secondary market prices through RIPE fairly easily. What’s leading you to be sure it will increase 400% in the next 24 months?
  - DaiPlusPlus 6y ago
    > What’s leading you to be sure it will increase 400% in the next 24 months? Honestly? Just a hunch. But also looking at trends on IPv4Auctions.com (now auctions.ip4.global), a few years ago blocks would go for $7/address, now they're $25/address - and as demand for IPv4 addresses won't be going away[1] and supply will only decrease (barring rare gluts of supply like AMPRnet's when a major /21 or /16 is broken down into smaller blocks to auction off). I think $100/address would be an upper-bound though, because at that point it's likely cheaper to finally invest in IPv6 deployment, including issuing new CPE than it costs to retain IPv4. [1] If an ISP could have deployed IPv6 by now then they would have done 5+ years ago. Any ISPs that haven't already deployed IPv6 by now (or even 5 years ago) must be facing some massive internal hurdles, such as incompetent management who see it as a massive and unnecessary technical expense when CG-NAT is good-enough for the majority of their customers - or have a hetereogenous network that would be a nightmare to bring up-to-code on any reasonable timescale). The only way these ISPs will upgrade is when the cost of operating an IPv4 network becomes unbearable - hence why I want to see IPv4 address costs skyrocket :)
elihu 6y ago
Yeah, that's been my theory for awhile: that IPv4 addresses are effectively a currency, and that a move to IPv6 would devalue that currency. A disproportionate amount of that currency is held by companies and institutions based in the United States, and in many cases they're the same companies and institutions that one might otherwise would expect to be responsible for much of the work of transitioning to IPv6. Meanwhile, Asia has a relative scarcity of IPv4 addresses and so is (as far as I know) ahead of the curve because their incumbents have less to lose.
- laurent92 6y ago
  Sorry for the stupid question but each time I got allocated an IPv6 address, I was allocated a /124 subnetwork (16 IPs) and not a single IP. How is IPv6 not going to be in shortage if even a small random actor like me gets allocated many IPs for each server?
  - dragonwriter 6y ago
    The IPv6 address space has 2^128 addresses. If you assign 2^4 addresses to every actual node, that still supports 2^124 nodes. There's about 2^33 people on the planet, so that still supports about 2^91 nodes nodes per person. A shortage isn't likely.
mindslight 6y ago
You've got the incentives backwards. An organizational push for IPv6 would allow a company to free up their IP addresses and sell (/lease) them, actually realizing that valuation.
- wmf 6y ago
  So far it looks like companies are mostly selling IPs that they have never used so they get the money and they get to stick with IPv4.

by rswail 6y ago

manquer 6y ago

Apple and Comcast probably can use them. Also Comcast may have to invest more in NAT grade infra if they sell significant chunk of their Class A allocation so it may not make sense for them.

tzs 6y ago

Comcast started rolling out IPv6 across their network a long time ago. They hit 100% support on their broadband network in 2014. They run a lot of the own stuff on IPv6, such as on-demand video on their set-top boxes. I think any home modem/router that they support that did not support IPv6 hit EOL long ago and was replaced. There may be some class A owners who are trying to hold back IPv6 so they can make money off their class A, but Comcast isn't one of them. Apple has long had IPv6 support in their operating systems, so I don't think they have been trying to hold it back either. Ford, Softbank, Daimler, and Prudential? It is not immediately obvious to me if they could do anything to hold it back even if they wanted to.

ArchOversight 6y ago
Comcast ran out of RFC1918 addressing to manage all the devices on their network (think cable modems/set top boxes/routers/everything) and had a massively segmented network setup. They rolled out IPv6 because it flattened that network again, and instead of worrying about being in the right segment for reachability everything now gets a globally unique IP address. It massively simplified their internal management.
DaiPlusPlus 6y ago
> Apple has long had IPv6 support in their operating systems, so I don't think they have been trying to hold it back either. And Microsoft has supported IPv6 since Windows 2000: https://news.microsoft.com/2000/03/15/microsoft-announces-ipv6-technical-preview-for-windows-2000/ https://news.microsoft.com/2000/03/15/microsoft-announces-ip.... I understand that LTE networks are exclusively IPv6, ditto "5G". My frustration with IPv6 is that my home ISP (Wave Broadband's DOCSIS service) is still IPv4 and they still have no plans to roll-out IPv6 even though Wave's WaveG service has been IPv6 for as long as I can remember. I understand the cause of this is because Wave's network is incredibly hetereogenous because Wave's business-model was to expand by buying-up smaller regional cablecos but to retain their infrastructure (and talent) compared to Comcast's approach of asset-stripping (bad for employees, but good for customers...). I want IPv6 so I can stop having to faff around with NAT. I'd love to be able to RDP-in directly to my desktop at home without having to pay companies like TeamViewer and Splashtop. ---- Update: I found Wave's peering page's IPv6 status page and it remains unchanged since 2011 (and possibly since 2008). Wow. Compare: https://web.archive.org/web/20110102140910/http://as11404.net/ipv6.html https://web.archive.org/web/20110102140910/http://as11404.ne... Live: http://as11404.net/ipv6.html http://as11404.net/ipv6.html
- criddell 6y ago
  > I want IPv6 so I can stop having to faff around with NAT. What makes you think your ISP will give you more than one IP even when they are no longer in short supply? If it's something that has value to you, it's something they will want to be paid for.
  - DaiPlusPlus 6y ago
    IPv6's has a fixed subnet size of 64-bits[1]. A router needs its own subnet, ergo everyone gets their own 64-bit address range. ISPs cannot give each customer a single IPv6 address because that would require each customer to have a single device connected directly to their router with no router in the CPE - basically like how dial-up used to work. There's no demand for a service like that. [1] See https://www.ripe.net/about-us/press-centre/understanding-ip-addressing https://www.ripe.net/about-us/press-centre/understanding-ip-... . I note that there is a lot of misinformation and badly-informed blog posts in the top search results for "ipv6 subnet size" like this one, ugh: https://www.crucial.com.au/blog/2011/04/15/ipv6-subnet-cheat-sheet-and-ipv6-cheat-sheet-reference/ https://www.crucial.com.au/blog/2011/04/15/ipv6-subnet-cheat... - which is wrong because IPv6 simply doesn't have 128-bit-sized subnet option (whereas you can have a 32-bit sized subnet in IPv4, which is how most residential ISPs/CPEs are configured).
mike_d 6y ago
Deploying IPv6 is nice, but it doesn't impact the value of IPv4 at all. Any internet service you want to be widely available will need to be dual-stack as long as we are alive.
- wmf 6y ago
  If (eventually) each service needs an IPv4 address, that implies that there isn't a shortage and thus the value of addresses would be near zero.
- Animats 6y ago
  Maybe not. Google says their traffic is now at 30% IPv6, and climbing at 5%/year.[1] Worldwide, Google is undercounting, because they are blocked in China. Google claims that China is 0.27% IPv6, but actually China is around 55%, according to APNIC. Policy goal in China is all IPv6 by 2025. [1] https://www.google.com/intl/en/ipv6/statistics.html https://www.google.com/intl/en/ipv6/statistics.html
  - mprovost 6y ago
    So it's taken 25 years to get to ~30% adoption. And a few years ago APNIC reported [0] that the rate was going down - I haven't found an updated report. In 2005 the US government mandated that all their computers be v6 connected by 2008. [1] This year they updated the mandate to 80% by 2025. [2] I learned networking in college in 1996 and I remember v6 coming up but so far I've never had to use it and chances are that I'll be able to retire still getting away with only working with v4. [0] https://blog.apnic.net/2018/05/21/what-drives-ipv6-deployment/ https://blog.apnic.net/2018/05/21/what-drives-ipv6-deploymen... [1] https://georgewbush-whitehouse.archives.gov/omb/memoranda/fy2005/m05-22.pdf https://georgewbush-whitehouse.archives.gov/omb/memoranda/fy... [2] https://www.cio.gov/assets/resources/internet-protocol-version6-draft.pdf https://www.cio.gov/assets/resources/internet-protocol-versi...
    - zajio1am 6y ago
      Considering that World IPv6 Launch was 2012-06-06 and nothing really happened before, it makes more sense to say that it has taken 8 years to get to ~30% adoption.

DaiPlusPlus 6y ago

> Assuming an average $25/address It'll be $50/address in a year, $100/address in 2022; I'm sure of it. Methinks now would be a very good time to invest^wspeculate in IPv4 addresses - I reckon it will be the next Bitcoin bubble-type event and likely even make it to the silly segment of TV news, with lampoonably inaccurate explanations of what an IPv4 address is to the general public.

Cullinet 6y ago
Exactly. What was the reason why do much money was required at this specific time? British businesses might have been natural customers for this address space, since the UK IDD code is 00 44. I would gladly pay $50 per IP for what is almost the only virgin address space going, if I had been offered. Diligent address space title management including some genuinely tough identification requirements, managed by any top tier corporate actions specialist, could have multiplied the value several times immediately, and then there would have been increased inflation rates applicable to the range precisely because of this management.
nixgeek 6y ago
Its been appreciating slowly for the last 5 years, I’ve seen deals in the $20-25 range throughout that time, and a few larger contiguous blocks going for a little more but not $50/ea. You can monitor secondary market prices through RIPE fairly easily. What’s leading you to be sure it will increase 400% in the next 24 months?
- DaiPlusPlus 6y ago
  > What’s leading you to be sure it will increase 400% in the next 24 months? Honestly? Just a hunch. But also looking at trends on IPv4Auctions.com (now auctions.ip4.global), a few years ago blocks would go for $7/address, now they're $25/address - and as demand for IPv4 addresses won't be going away[1] and supply will only decrease (barring rare gluts of supply like AMPRnet's when a major /21 or /16 is broken down into smaller blocks to auction off). I think $100/address would be an upper-bound though, because at that point it's likely cheaper to finally invest in IPv6 deployment, including issuing new CPE than it costs to retain IPv4. [1] If an ISP could have deployed IPv6 by now then they would have done 5+ years ago. Any ISPs that haven't already deployed IPv6 by now (or even 5 years ago) must be facing some massive internal hurdles, such as incompetent management who see it as a massive and unnecessary technical expense when CG-NAT is good-enough for the majority of their customers - or have a hetereogenous network that would be a nightmare to bring up-to-code on any reasonable timescale). The only way these ISPs will upgrade is when the cost of operating an IPv4 network becomes unbearable - hence why I want to see IPv4 address costs skyrocket :)

elihu 6y ago

Yeah, that's been my theory for awhile: that IPv4 addresses are effectively a currency, and that a move to IPv6 would devalue that currency. A disproportionate amount of that currency is held by companies and institutions based in the United States, and in many cases they're the same companies and institutions that one might otherwise would expect to be responsible for much of the work of transitioning to IPv6. Meanwhile, Asia has a relative scarcity of IPv4 addresses and so is (as far as I know) ahead of the curve because their incumbents have less to lose.

laurent92 6y ago
Sorry for the stupid question but each time I got allocated an IPv6 address, I was allocated a /124 subnetwork (16 IPs) and not a single IP. How is IPv6 not going to be in shortage if even a small random actor like me gets allocated many IPs for each server?
- dragonwriter 6y ago
  The IPv6 address space has 2^128 addresses. If you assign 2^4 addresses to every actual node, that still supports 2^124 nodes. There's about 2^33 people on the planet, so that still supports about 2^91 nodes nodes per person. A shortage isn't likely.

mindslight 6y ago

You've got the incentives backwards. An organizational push for IPv6 would allow a company to free up their IP addresses and sell (/lease) them, actually realizing that valuation.

wmf 6y ago
So far it looks like companies are mostly selling IPs that they have never used so they get the money and they get to stick with IPv4.