Enable IPv6

Well... this might come sooner than later... Apple is now requiring all apps for iOS 9 to support IPv6. Obviously IPv4 support is still required too, but it shows how fast things might be moving, and how quickly Sonos might get left behind by not supporting it...

http://www.internetsociety.org/deploy360/blog/2015/06/apple-will-require-ipv6-support-for-all-ios-9-...

Well... this might come sooner than later... Apple is now requiring all apps for iOS 9 to support IPv6. Obviously IPv4 support is still required too, but it shows how fast things might be moving, and how quickly Sonos might get left behind by not supporting it...

http://www.internetsociety.org/deploy360/blog/2015/06/apple-will-require-ipv6-support-for-all-ios-9-...
Thought I'd also add this post from Akamai's blog... being a major CDN, they're in a unique position to see all the traffic from end-users that support IPv6. This post includes a nice graphic showing significant increases in end-user IPv6 requests over almost the past two years from a number of countries around the world.

https://blogs.akamai.com/2015/06/three-years-since-world-ipv6-launch-strong-ipv6-growth-continues.ht...

ARIN's got about a month (based on historical numbers) before it runs out of IPv4 addresses. ARIN provides IP address space to companies and organizations based in North America, and is the largest regional IP registry as far as registered address space. More info, along with some history of IPv4 depletion, and some info on how NAT has been used in growing numbers to extend the life of IPv4... An Update on IPv6

APNIC (Asia-Pacific region) ran out of IPv4 addresses in April 2011... RIPE (Europe and Middle East) ran out in September 2012. In May 2014, LACNIC (Latin America and Carribbean) ran out of IPv4 addresses.

Obviously companies can barter addresses they aren't using (an article from Network World talks about the private IPv4 address market), which will help to extend the life of IPv4 a little longer... but with four out of five global IP registries out or nearly out of IPv4 addresses, Sonos should really be starting to work on this.

There is a lot of news around IPv6 lately, and some interesting articles, thanks for sharing MikeV. When it comes down to Sonos and IPv6, keep in mind that we only connect to an incredibly small slice of the Internet. Sonos players only connect to our home servers, music services, and Internet Radio streams. That’s about all of the net which our devices will ever see. Those locations are all major and will, in all likelihood, remain IPv4 accessible long into the future.

Your local network will remain on IPv4, as the home is never going to have enough devices that you need to go to IPv6. Again, the players will continue to function perfectly fine.

We are constantly monitoring the market and technology trends that are important. Should there come a time when Sonos players need to connect to an IPv6 server, we’ll make sure that you can continue to listen to music in your home. Whether this means making players IPv6 compatible, or finding a solution which will work somewhere in the middle.

You are completely missing the point and are part of the problem with this mindset.

> Your local network will remain on IPv4, as the home is never going to have enough devices that you need to go to IPv6. 

Seriously?  You think this is what IPv6 is about?  You think that the "home network" is the only thing running out of space?

Here is a very real scenario for your product.  

A few months down the road, this hip new company starts offering a service that is novel and your customers are clamoring for.  But because IPv4 address space is exhausted, they are struggling to get enough v4 addresses to support their business and the address space they have is getting increasingly more expensive to keep due to greater demand and zero new supply.  So, they deploy their service dual-stacked and simply say, "sorry, for the problems over IPv4, there is nothing we can really do about it.  It is putting us out of business.  Use IPv6 for the best experience."  

Now, your product is SOL because you think your own integration are going to be the big kids from now and forever more.  Next, you as a company scramble to implement support. but because you were too bullheaded to head our requests, you know nothing about implementation.  You have no expertise in building out a v6 network, so you struggle for a while getting the basics working, all the while that competitor of your which wasn't so short-sighed, is getting your business.

If you think this is far fetched, you deserve to be out of business.

You are completely missing the point and are part of the problem with this mindset.

> Your local network will remain on IPv4, as the home is never going to have enough devices that you need to go to IPv6. 

Seriously?  You think this is what IPv6 is about?  You think that the "home network" is the only thing running out of space?

Here is a very real scenario for your product.  

A few months down the road, this hip new company starts offering a service that is novel and your customers are clamoring for.  But because IPv4 address space is exhausted, they are struggling to get enough v4 addresses to support their business and the address space they have is getting increasingly more expensive to keep due to greater demand and zero new supply.  So, they deploy their service dual-stacked and simply say, "sorry, for the problems over IPv4, there is nothing we can really do about it.  It is putting us out of business.  Use IPv6 for the best experience."  

Now, your product is SOL because you think your own integration are going to be the big kids from now and forever more.  Next, you as a company scramble to implement support. but because you were too bullheaded to head our requests, you know nothing about implementation.  You have no expertise in building out a v6 network, so you struggle for a while getting the basics working, all the while that competitor of your which wasn't so short-sighed, is getting your business.

If you think this is far fetched, you deserve to be out of business.
We know the local network isn't going to run out of IPv4 address, so the only case we need to be worried about is in regards to the external servers we connect to. You do bring up a good point Jon, and we've talked about the possibility of such a service around here to be prepared just in case.

Any music service that wants to become main stream today will need to have IPv4 access. Should your situation come to play where it's simply impossible for them to get an IPv4 server connection in the future, and yet they still get the level of support you suggest, we'll work on a solution with them. 

You are completely missing the point and are part of the problem with this mindset.

> Your local network will remain on IPv4, as the home is never going to have enough devices that you need to go to IPv6. 

Seriously?  You think this is what IPv6 is about?  You think that the "home network" is the only thing running out of space?

Here is a very real scenario for your product.  

A few months down the road, this hip new company starts offering a service that is novel and your customers are clamoring for.  But because IPv4 address space is exhausted, they are struggling to get enough v4 addresses to support their business and the address space they have is getting increasingly more expensive to keep due to greater demand and zero new supply.  So, they deploy their service dual-stacked and simply say, "sorry, for the problems over IPv4, there is nothing we can really do about it.  It is putting us out of business.  Use IPv6 for the best experience."  

Now, your product is SOL because you think your own integration are going to be the big kids from now and forever more.  Next, you as a company scramble to implement support. but because you were too bullheaded to head our requests, you know nothing about implementation.  You have no expertise in building out a v6 network, so you struggle for a while getting the basics working, all the while that competitor of your which wasn't so short-sighed, is getting your business.

If you think this is far fetched, you deserve to be out of business.
The last thing I'll say on this, at least for the immediate future, is that I feel this line of thinking is extremely short-sighted.

Over the past day, numerous scenarios have popped into my head about how Sonos loses by not supporting IPv6. I could write a few pages about it all.  But I won't. My opinion on this matter is clear, and there's not much more that I can say.

I think it's a mistake to ignore this. It's the future, and you know it's coming. Take the time now to get it right, rather than be rushed down the road when someone - an ISP or a music service - makes a change that suddenly prevents your products from operating properly, affecting your customers' ability to use them.

You are completely missing the point and are part of the problem with this mindset.

> Your local network will remain on IPv4, as the home is never going to have enough devices that you need to go to IPv6. 

Seriously?  You think this is what IPv6 is about?  You think that the "home network" is the only thing running out of space?

Here is a very real scenario for your product.  

A few months down the road, this hip new company starts offering a service that is novel and your customers are clamoring for.  But because IPv4 address space is exhausted, they are struggling to get enough v4 addresses to support their business and the address space they have is getting increasingly more expensive to keep due to greater demand and zero new supply.  So, they deploy their service dual-stacked and simply say, "sorry, for the problems over IPv4, there is nothing we can really do about it.  It is putting us out of business.  Use IPv6 for the best experience."  

Now, your product is SOL because you think your own integration are going to be the big kids from now and forever more.  Next, you as a company scramble to implement support. but because you were too bullheaded to head our requests, you know nothing about implementation.  You have no expertise in building out a v6 network, so you struggle for a while getting the basics working, all the while that competitor of your which wasn't so short-sighed, is getting your business.

If you think this is far fetched, you deserve to be out of business.
Even if it's years away, at least you'll be ahead of the game.

Sonus is correct, in that nearly every home broadband ISP will find a way to provide customers with local IPv4 addresses that work via NAT (even while the infrastructure of many of them is already moving to IPv6 for transport.)  Similarly, people are today using IPv6 on the mobile phones and don't even know it - the mobile service provider uses IPv6 to phone, and there's dynamic mapping at both ends so IPv4 only applications work.  Google reports that nearly 20% of US users are accessing Google via IPv6 (and this up from 10% just a year ago)

Of course, the IPv4-only devices and applications are slower than those using IPv6, and are prone to overloading and jitter issues, as your audio/video stream is being dynamically mapped to/from IPv4 (along with everyone else) at some data center in the service providers network...  For example Facebook noted that devices supporting IPv6 are loading content from them 20 to 40% faster than those just using IPv4.

So, the real question is "can you build consumer home devices and services that only support IPv4, and still manage to get by as the Internet moves IPv6?"   Yes, you definitely can.    Are you at risk of having competitors show up with better performing solutions, and this situation becoming much more noticeable to your users over time?  Alas, that's also true – whether it's an issue depends on each companies competitive situation, user focus, and product planning timeline.

/John

John Curran 
President and CEO
American Registry for Internet Numbers (ARIN)

Another thing to consider is that sooner or later (and my money is on sooner) we're going to reach a point where enough content is on IPv6 that the economics of continuing to provide IPv4 as a free component of "Internet Access" will no longer make sense to some of the larger eyeball providers. (The economics will, in fact, drive the largest ones to this brink fastest). Lee Howard has provided some excellent data in this regard.

Once that time comes, you may face a loss of customers that don't want to pay extra for IPv4 just to get to Sonos.



FWIW, I won't be subscribing to Sonos until I can get there over IPv6.