Hard News by Russell Brown

It would be madness

When a Ministry of Research, Science and Technology discussion paper earlier this year recommended that the government provide at least part of the funding for a gigabit-speed Advanced Network, it seemed hopeful news indeed. More so because MORST plumped for a relatively open model where not only universities and research institutions could stump up to join the gigabit club, but some private sector businesses.

An Advanced Network - aka a research network or next-generation Internet - is not only much faster than the "commodity Internet", it is different in kind. Typically, it will have deployed Internet Protocol Version 6, which offers huge improvements in security and integrity of service. All other OECD countries - not to mention such technology titans as Costa Rica - have such networks at various stages of maturity.

They are rapidly becoming essential for certain kinds of research - notably those involving the transport of very large files and the sharing of computing resources across distance - and, indeed, there is an apparently well-founded fear that unless something is done, certain kinds of research, particularly in the biotech field could become unviable in New Zealand.

Many such networks in other countries have fairly tight acceptable use policies and will not accept straight commercial use. But in those places there is usually a commercial alternative (often right down to consumer level - Verizon in the US has just started selling 30Mbit/s to the home for $34.95 a month). In New Zealand, there is not. So there is a logic in allowing a few private sector organisations to share the load of building and operating this kind of network (whose costs might eventually run to $250 million). Yet by the time a Network Use Policy was published in September, that vision had all but disappeared.

While publicly-funded tertiary institutions and CRIs have an automatic right to join, others in the "innovation sector" will have to depend on a partnership with a member of the non-profit Advanced network Company itself. It is possible that access to the network will be even further confined by the time the current RFI process feeds into a government spending commitment.

To understand why this is happening, you need to grasp that networks like this militate against the traditional telecommunications company model, in which telcos sit in the middle of their networks, managing traffic and access and clipping tickets. The new model is variously referred to as a "stupid network" (because all the intelligence is at the edge, rather than the centre) or "dark fibre". All its participants want is access to the basic commodity of fibreoptic cable, to run their own services in such a way that once a connection fee has been paid, bandwidth is effectively free and limitless.

Can you see why Telecom New Zealand would regard this as the end of the world? And why, as a consequence, Rod Deane and Theresa Gattung have been lobbying senior government ministers to confine access to this thing as tightly as possible? The Screen Council - whose members would love to be able to move around very large files at reasonable cost - is lobbying in the other direction, but there are no prizes for guessing who gets more attention.

Telecom is, on the other hand, offering to provide parts of the Advanced Network - which, out of necessity, will be cobbled together as much as possible from existing fibre - but doing so in a way that is effectively a non-sequitir. It is not offering dark fibre (and it will probably be a cold day in hell before it does), but a 200Mbit/s "managed service". Users will take its network management value-adds even if they choke on them. Which they probably will.

In truth, Telecom would be delighted if this thing just went away. After a few drinks had been consumed at the party for the Tuanz broadband conference this week, a colleague and I sat down to hear someone from a telecommunications equipment supplier and someone, I believe, from Telecom, vilifying the whole idea and the people behind it. A little later, the intensity of the politics around this thing was underlined when a representative of one of the universities started hissing abuse at a member of the Advanced Network steering committee.

In part, this is because all the envisioning has been done. The big geeks went into their two days of sector-group discussion at the conference with five and 10-year plans in hand. As a colleague of mine pointed out, they didn't have anything to do but argue about the politics.

There are valid questions about the Advanced Network proposal, including what it will really cost to build and whether it can really be cobbled together from a scattering of existing fibre installations. But at least one of the challenges is also a crucial opportunity.

A number of research institutions are not very adjacent to the planned network. For Industrial Research Limited, the "last mile" to its base in the Hutt Valley is about 16km, and it faces considerable cost in connecting. But that fibre run will pass schools, municipal buildings and business. Could they also contribute - whilst not being users of the core Advanced network itself? Could IRL be what Canadian Bill St Arnaud described on the conference's opening day as an "anchor tenant" for new community fibre infrastructure?

St Arnaud, who works for Canarie Networks, made a compelling case for all the things we don't do here: most notably the separation of services from infrastructure. He touted the concept of MUSH (Municipal, University, School and Hospital) networks, which are funded or loaned money to become the anchor tenants. Does it? Well, Canada is third in the world for broadband penetration (and that's real broadband), while New Zealand is 25th and falling. Costs there are a third of those here. And they've overcome some very significant geographical issues to get there.

Another thing makes me trust the big geeks: they've been right before. The people behind the Advanced network proposal are largely the same people who brought and built the Internet here in the first place. They had the good sense to clear out and let it be commercialised in 1996 and I'm strongly inclined to let them build the basis of our next network.

I am not a reflexive Telecom-basher. I did not believe, on balance, that full unbundling of Telecom's network was appropriate. This far down the line it would have been a worrying encroachment on Telecom's property rights. But this is different. Right now, Telecom is lobbying to prevent or confine a hugely important service that it is not prepared to provide itself.

The government really needs to show some guts and vision over this. Because the alternative is that the interests of every other business in the country - and especially those in the sectors the government has identified as crucial to New Zealand's economic development - will be sacrificed to those of a single incumbent. It would be madness.

PS: You'll note that Public Address is presenting the second Flying Nun pub quiz on occasion of the release of the Second Season DVD compilation, on Wednesday November 24, at the King's Arms, from 8pm. Details are in the ad on this page (reload if you don't see it). It will be very good fun and you should feel free to join us. You might even win yourself a special festive prize!