BigPipe IPv6

Forums › Spark New Zealand (including Skinny and BigPipe) › BigPipe IPv6

1 | 2

aumouth

73 posts

Master Geek
+1 received by user: 11

#1343419 15-Jul-2015 00:44

However, if you can make a compelling reason why we should prioritise it (other than 'I like tinkering with stuff') I'm very open to hearing it.

Both Microsoft (Win 7+) and Apple (OS X 10.10+ and iOS 8.4+) mainstream IPv6 over IPv4
In fact, in the very latest Apple versions due next month - OS X 10.11 and iOS 9, Apple are specifically delaying usage of IPv4 A records and waiting for IPv6 AAAA records before using IPv4. (Refer: https://www.ietf.org/mail-archive/web/v6ops/current/msg22455.html) Most network engineers agree that latency and throughput is better on IPv6, against direct IPv4 and CG-NAT.
Lack of support for IPv6 at the ISP level will mean users are getting a reduced experience and response times, not only from better latency and throughput but also because their Apple devices are delaying and waiting/trying IPv6 first.
I expect Windows will also very shortly follow the same principle in the next WIn7 and Win8 SP's.

Zeon

3919 posts

Uber Geek
+1 received by user: 752

Trusted

#1343421 15-Jul-2015 01:22

While there might not be a business case for it until you start seeing IPv6 only services or service that break badly under CG-NAT.

In saying that I think there is a moral responsibility for ISPs to make the investment (as well as content providers). At the end of the day the internet is how they make money and their business runs and the principles of the internet are around collaboration at a technical level (with which the internet is possibly one of the best example ever for global collaboration). It needs IPv6 and the lack of adoption now will hurt more in the future - from the content provider who bears extra cost to get IP space to provide that content to the person designing their service to get through CG-NAT and spending months of extra time on that...

Show leadership Bigpipe.

Speedtest 2019-10-14

BigPipeNZ

1170 posts

Uber Geek
+1 received by user: 777

Trusted

BigPipe

#1345277 16-Jul-2015 11:00

Zeon: While there might not be a business case for it until you start seeing IPv6 only services or service that break badly under CG-NAT.

In saying that I think there is a moral responsibility for ISPs to make the investment (as well as content providers). At the end of the day the internet is how they make money and their business runs and the principles of the internet are around collaboration at a technical level (with which the internet is possibly one of the best example ever for global collaboration). It needs IPv6 and the lack of adoption now will hurt more in the future - from the content provider who bears extra cost to get IP space to provide that content to the person designing their service to get through CG-NAT and spending months of extra time on that...

Show leadership Bigpipe.

I take your point, but please bear in mind that it isn't just about the business case, it's a case of opportunity cost. We have a (extremely) limited amount of money AND people available to us. If we put them on IPv6, that means other projects must suffer. Things that are crucial to just keeping the business running, or things that are important products (like UFB). And it's not just the cost of implementing it, there will naturally be ongoing costs to managing it too - troubleshooting issues etc.

I found this today in a reddit thread

http://www.ipv6.org.nz/2015/01/27/ipv6-support-across-nz-isps-january-2015/

According to that, no residential ISP in NZ is fully supporting IPv6 right now. A few have trials, and the likes of Snap have it available but not in production support - so even that's not full implementation yet.

bigpipe.co.nz
https://www.facebook.com/BigPipeNZ
https://twitter.com/BigPipeNZ

aumouth

73 posts

Master Geek
+1 received by user: 11

#1345485 16-Jul-2015 14:12

BigPipeNZ: And it's not just the cost of implementing it, there will naturally be ongoing costs to managing it too - troubleshooting issues etc.

You forgetting the cost of not implementing it....
Connection delays from user with Apple devices (and Windows shortly too, IMHO)
A competitor provides it first, leaving BigPipe in catch-up mode and not a market leader - just a follower, and losing a users/market share as BigPipe would be majority early adopters
IPv6 services only coming online being un-supported, leaving BigPipe again in market catch-up.

lxsw20

3590 posts

Uber Geek
+1 received by user: 1983

Subscriber

#1345492 16-Jul-2015 14:23

BP don't seem to be disagreeing with you, but IMO you're being a bit hyperbolic about the whole thing really.

sbiddle

30853 posts

Uber Geek
+1 received by user: 9992

Retired Mod

Trusted

Biddle Corp

Lifetime subscriber

#1345494 16-Jul-2015 14:24

aumouth:
BigPipeNZ: And it's not just the cost of implementing it, there will naturally be ongoing costs to managing it too - troubleshooting issues etc.

You forgetting the cost of not implementing it....
Connection delays from user with Apple devices (and Windows shortly too, IMHO)
A competitor provides it first, leaving BigPipe in catch-up mode and not a market leader - just a follower, and losing a users/market share as BigPipe would be majority early adopters
IPv6 services only coming online being un-supported, leaving BigPipe again in market catch-up.

I don't see any of those as a compelling reason.

BigPipeNZ

1170 posts

Uber Geek
+1 received by user: 777

Trusted

BigPipe

#1345502 16-Jul-2015 14:58

aumouth:
BigPipeNZ: And it's not just the cost of implementing it, there will naturally be ongoing costs to managing it too - troubleshooting issues etc.

You forgetting the cost of not implementing it....
Connection delays from user with Apple devices (and Windows shortly too, IMHO)
A competitor provides it first, leaving BigPipe in catch-up mode and not a market leader - just a follower, and losing a users/market share as BigPipe would be majority early adopters
IPv6 services only coming online being un-supported, leaving BigPipe again in market catch-up.

Those might become good reasons in the future, but right now they are nowhere near as important as the other stuff we have the network team working on. Sorry.

bigpipe.co.nz
https://www.facebook.com/BigPipeNZ
https://twitter.com/BigPipeNZ

Quic Broadband

Move to New Zealand's best fibre broadband service (affiliate link). Free setup code: R587125ERQ6VE. Note that to use Quic Broadband you must be comfortable with configuring your own router.

freitasm

BDFL - Memuneh
79608 posts

Uber Geek
+1 received by user: 38034

Administrator

ID Verified

Trusted

Geekzone

Lifetime subscriber

#1345602 16-Jul-2015 16:56

I rather see peering and better support than IPv6 these days.

Please support Geekzone by subscribing, or using one of our referral links: Quic Broadband (free setup code: R587125ERQ6VE) | Samsung | AliExpress | Wise | Sharesies

Geekzone and Quic social @ DataVault Auckland 18 Oct 2025 11AM - 2:30 PM

aumouth

73 posts

Master Geek
+1 received by user: 11

#1345750 16-Jul-2015 20:17

you're being a bit hyperbolic about the whole thing really.

Hindsight and foresight, never the same really.
Followers and leaders, never the same either :)
Only time will tell whether the planning and road-mapping for IPv6 should be done now, or left until others do it and you follow.

sbiddle

30853 posts

Uber Geek
+1 received by user: 9992

Retired Mod

Trusted

Biddle Corp

Lifetime subscriber

#1345757 16-Jul-2015 20:25

aumouth:
you're being a bit hyperbolic about the whole thing really.

Hindsight and foresight, never the same really.
Followers and leaders, never the same either :)
Only time will tell whether the planning and road-mapping for IPv6 should be done now, or left until others do it and you follow.

Point 1: The harsh but very true real world reality is that 99.9% of internet users neither care for IPV6 or even know what it is. Providing Trademe, Netflix, Google and YouTube work they're happy.

Point 2: You clearly fall into the 0.1% that care. That's a very small minority.

Point 3: Right now IPV6 is a solution for a problem that doesn't actually exist. Yes it can offer all sorts of benefits, but it's not without some major issues that will arise.

Point 4: Until something breaks with point 1, return to point 1.

It doesn't matter of you're first or last to IPV6 - and there are many, many, many reasons why an ISP would want to be last.

If you want IPV6 there are plenty of options out there that will mean looking somewhere other than an ISP targeting the bottom end of the market.

1 | 2