All our broadband eggs still in one basket

Labour’s Trevor Mallard called Wednesday’s connection numbers “hopeless.”

Latest figures from Chorus, charged by the government to rollout ultra-fast broadband to New Zealand, show that the progress is still slow. The news release is a masterful piece of spin. However, when you start to analyse the underlying numbers, it shows we are a really long way off fast broadband delivered by UFB.

The National Business Review reported that:

“The latest numbers released by ICT Minister Amy Adams, for the December quarter, show Chorus [NZX: CNU] and other UFB contract holders have now rolled fibre past 363,109 potential customers – an increase of 40,630 over the September quarter.

Of those, 19,915 or 5.49% have signed up for a UFB account with a retail ISP. During the quarter, 5667 new UFB customers signed up.” – Source

Let’s analyse these numbers a little more.

The key words here are “rolled fibre past.” In other words, ran UFB past potential customers, didn’t actually connect them. Good luck with getting a connection, with anecdotal reports showing that getting plugged into UFB can be a huge challenge. A good example is in the comments of the article linked above. It shows a typical story of a customer attempting to negotiate a UFB connection and the comical process that occurs. So far, its taken three months, despite the fact that the fibre runs within 30cm of his house.

“An increase of 40,630 over the previous quarter”.

Side by side in a sentence those numbers look good. However, when you consider that there are around 1.6 million households in New Zealand, it starts to look less impressive. Let’s say we’ve now got cable running past 400,000 households. That leaves 1.2 million to go. If Chorus “rolls past” 40,000 households a quarter, then we’ll have UFB 100% available by 2022. Now, I couldn’t find statistics on how many business premises there are. However we know that there are just under 600,000 registered companies in New Zealand. If half of those have a physical premise, then full “roll past” isn’t complete until 2023 or later.

So nine years before roll past is finished.

Let’s have a look at connecting to UFB.

“Of those, 19,915 or 5.49% have signed up for a UFB account with a retail ISP. During the quarter, 5667 new UFB customers signed up.”

This is called spin. It could be read that 5.49% of people have signed up. The reality is that the percentage relates to the total number of properties rolled past. This means that the actual uptake is somewhere near 1%.

“During the quarter, 5667 new UFB customers signed up.”

So by that rate then by the year 2099 everyone will be connected.

Now, I’ve mangled the figures a bit, I’m not a statistician and I don’t have time to get completely accurate but you get the picture. It’s too slow. This is what happens when we put all our eggs in one basket and have inefficient, old, business processes trying to deploy what is new technology. It’s like employing one company to build motorways with no off or on ramps and then employing several companies to try and build those entry and exit points.

Why do we need it faster? Because demand is going to outstrip supply, that means that in the next few short years, even UFB by today’s standards will not be quick enough. So not only do we need that bandwidth now, we need to be doubling it every two or three years.

4K televisions are now very affordable. A new generation are going past the current High Definition (HD) to Ultra HD. Now, one of the biggest drivers of growth on the internet is video content. I can buy a 4K TV today and start to watch Ultra HD movies via the net. As long as I only want to watch two or three a month (if have a data cap of 500GB), I’m prepared to wait for the download. Because, an HD movie is 160GB on average. Those figures are going to drive a large increase in data usage. But wait, there’s more: 8K will be here in a few short years. (As an aside, 8K video streaming requires a connection of 48Gbps, 48 times faster than current UFB.)

The number of devices attaching to the internet is almost growing exponentially. From smart phones, to smart devices, smart houses, to the internet of things, all of these endpoints require bandwidth in various sizes. It is estimated that in the next two years the number of devices attached to the internet will increase to around 18 to 20 billion. My household has a dozen devices that attach to the internet. Now, I’m a technologist but you get the picture.

To cater for that growth, you need to increase speed and bandwidth between 50-100% a year, per household. That means 100Mbs last year, 200Mbs this year, 400Mbs in 2015, and so on. Yet UFB is only currently delivering 100Mbs (if you can get a connection).

I’ve commented before on how we need to get broadband right. Putting all our eggs in a single basket is going to leave us with motorways that no one can use and traffic congestion of the highest order. For a country that is heavily reliant on the Internet given our location and ICT industry, banking on the slowest fibre rollout on the planet is just dumb.

We can only hope that this year political parties get the message, change the policies, and start making some smart choices about how we deploy bandwidth better, faster, cheaper, and using whatever technology can provide, not just fibre. We need new plans and less spin.

Ian Apperley is the director at Isis Group and blogs at Whatisitwellington