NBN: the next 1000x and why it matters

commentary We've got 1000x the internet we had in 1993. What's the point of another 1000x by 2033?

1000x. That's how far internet speeds have come in the past 20 years. The way we use the online world has changed so much, but for some reason, we keep finding ourselves thinking it is what it is and it won't ever change much again. With that fixed thinking, it's easy to feel that the future looks a lot like today, and that this future doesn't need an internet much different from the one we have today. Why should we spend a lot of government money replacing all that copper with so much fibre? What possible changes can that bring to the internet we already know and love?

The problem with this shift to fibre is that it has all come so easy until now. From low-baud acoustic couplers to dial-up modems to ADSL2+, we've won that 1000x improvement with a minimum of obvious fuss. Existing copper infrastructure, plus fancy new little boxes in our homes, has equalled consistent leaps in speed at quite nicely spaced intervals.

That first 1000x shows us a few key stages of evolution in how we understood the internet and what it was good for.

Dial-up

At first, it was about asynchronous tasks like email and web pages. One person pushes and another person pulls at their convenience. As basic dial-up got a little faster, we even took the extra time to share (small) pictures and (low-bitrate) music, and we had our first taste of real-time interaction as instant messaging let us chat with friends in little bursts of text. Yahoo search. Alta Vista. Eudora. Geocities.

For many, this was all very good, but it seemed like it would be enough. This internet would do just fine.

ADSL1

As early, highly throttled and highly capped broadband arrived, we started to see two new experiences: things arrived sooner, and wait times started to disappear. Waiting was a central part of the internet experience until now. Web browsers added tabs to let us open a few things at a time and read some things while other things were loading. We could start to do two things at once. Download something while reading other things. Search, open in new tab, keep searching, keep loading, keep exploring. The balance shifted in favour of consuming over exchanging. We could receive much more than we could give. Google. Napster. Flash.

For many, this was all very good, but it seemed like it would be enough. This internet would do just fine.

ADSL2+/HFC

Broadband got better. The dynamic shifted. Now we could do things in "real time". The idea of websites that updated while you looked at them arrived. The static came to life and the web became an operating environment. Video emerged as a viable part of the consumption model. Voice calls arrived. Video calls arrived. Social interaction was able to move out of the push-pull of email and the basics of IM, and shift into full real-time networking with vast numbers of friends old and new. Skype. YouTube. Myspace. Facebook.

This is where we are now. We could say we've travelled a couple of steps further, but it's been incremental and the basic model has been maintained. For many, this is all very good, but it seems like it will be enough. This internet will do just fine.

Faster horses?

So why do we need to spend so much money to get from here to there if it didn't cost so much to get from there to here?

While step one to two to three above seems like such an organic flow of concepts, in hindsight, it was a shift from text to rich media. Print to colour television are both forms of media, but we know our experience with each is utterly different. But somehow, we've made this all so seamlessly integrate into our lives that we've already found a comfortable way to just take it all for granted even as it changed us all so deeply.

So where does the next 1000x come from? The foreseeable future of copper offers another 5x at best. If we get that out to 10x, we'll be doing very well in the next 10 years. We simply do not get to 10 gigabit and beyond with copper networks in the picture. To find another 100x after that, we need to get fibre involved. Fibre delivers that 1000x over today's speeds to businesses already. We'll just be waiting for the trickle down in prices to make it reach consumer affordability.

Perhaps more important is a question of why that next 1000x is important anyway. Is it just faster horses? Aren't we consuming enough stuff already, and is pulling more stuff really going to make our lives better?

Access and opportunity

One subtle part of the back story above is the nature of access. From day one, the internet changed our relationship with information. Library and post office opening hours no longer controlled when we could access new information. Stores didn't have to be open for us to buy things.

But it also opened opportunities for those who saw the potential. Yahoo. Google. Skype. YouTube. Facebook. They're all still here, and they were all started by clever users who saw a new way to use the internet. At each step, someone with an idea gave people a new way to use that extra capacity. We didn't know we wanted it until we had that new reason to use it.

None of those companies had to ask for permission. They just made something and pushed it out onto the internet for users to bring to life.

Push is the creative force where pull is for consumption. And more than anything else, it is the push that has been deeply limited for home users. With the next real leap in broadband performance, it is push that becomes more usable. What opportunities could appear when we can share as easily and seamlessly as we can pull?

When the network disappears

There are real applications today that shift from clever to revolutionary when the wider internet experience becomes as fast as the local.

With the question of wait times removed, private personal cloud becomes a reality. Network-attached storage devices stop being a local backup device and start giving us remote access to all of our own most important files, photos and videos, without entrusting our storage to US corporations. Synology's outstanding DiskStation range is essentially made up of private servers in search of better networks, giving you a real computing experience anywhere you can get fast and reliable access. But that requires fast upload speeds out of your home to be able to have a fast download experience from your server in those remote locations.

Where earlier, we loved being able to do more than one thing at a time online, higher bandwidths will allow for dozens and hundreds of things to happen at once around our home. Each device streaming high volumes of content. An Internet of Things sending information up, down and between many dozens of other new connected devices. The "and" of the internet — not this or that, but this and that ... and that and that and that — comes to life when some tasks don't have to slow down to accommodate others.

At its simplest but most potent, the network disappears. As speeds up and down the pipe increase, the movement of files becomes as transparent as from one hard drive to another across a local network. The train changed the world because speed changed our relationship to distance. This, too, the internet has already achieved, but has more yet to achieve. The cloud is not yet local, home servers are not yet global.

The only thing that will then control someone's ability to produce new forms of content, applications or services will be their own ability to produce, to code, to program.

That's not quite true. Control is the other issue. When the networks are controlled by corporate interests, there will always be limits on what you can use your bits to achieve. Even Google is telling US users of its gigabit fibre service that they cannot run their own servers, because home servers threaten the Google business model of charging for access to cloud servers. Live broadcasting from your home studio may be considered an "excessive usage pattern" by a provider that has links to traditional broadcast service providers.

We need the government to ensure that the next 1000x improvement comes with no strings attached. Open, accessible, equitable, with bits that are just bits, no matter what they are used for.

That next 1000x matters. We don't need all of it tomorrow. We need it arriving 10x, 20x, 50x at a time. The view from here isn't clear, but the view from 2033 will undoubtedly show us that the internet in 2013 is not the natural order of the internet.

There is so much more to be done, and so many new ideas to be created by the next generation of clever young entrepreneurs just waiting for that next wave of speed, that next three orders of magnitude, to ride in on.