The Lateline debate held on Monday between Malcolm Turnbull & Anthony Albanese was met with both anticipation & trepidation from National Broadband Network supporters. We were all excited to finally see the NBN take centre stage for a debate, & were cautious of both Turnbull’s history of becoming belligerent when challenged, & Albanese’s being so new to the job.

It seems, both were well founded, with Turnbull resorting to arrogant overtones early on, & Albanese getting some basic figures wrong. Emma Alberici, the host, attempted to maintain control, however the utter disrespect Turnbull has for anyone who asks pertinent question prevented any semblance of a rational debate forming.

IN THE BEGINNING

The debate started with a simple question from Alberici: why go fibre?

Albanese did well here, listing the many positive outcomes from deploying an FTTP network, although this is where a serious error on Albo’s part happened. Albo quoted “100Mbps” as the current peak download speed of the NBN, & 400Mbps as the peak upload. I can see that Albo was intending to say 1000Mbps, but failed to do so.

The key point stressed by Albanese, & anyone who agrees with the FTTP NBN plan, is that fibre looks to the future. It’s something the debate misses due to glib lines like “we can’t possibly know X will  happen”. Yes, we can actually, CISCO make a business out of predicting data trends & pricing their products accordingly.

The same question was asked of Turnbull. The tired old “get the right mix of technology” line was trotted out. I wonder why no one has said “well, the right mix is FTTP, Wireless & Satellite”. The funny thing is, Turnbull points to the 20% that NBN Co’s contracts cover as being part of his plan.

No Mal, that 20% wouldn’t exist if the ALP hadn’t started the FTTP NBN. Do not claim ownership over it now, it’s just that there’s nothing you can do about it without costing (a now revealed) $4.8b to renegotiate the contracts. Rampant misinformation from Mal in his first answer.

Mal then goes on to say 100Mbps FTTN is available at 400m from the node. VDSL2 is NOT being delivered at 100Mbps at 400m from the node, in fact, without vectoring, a single pair service will not get 100Mbps at 100m. Mal is basically quoting this:

[VDSL2 Vectoring] takes full advantage of existing resources by making conditions in the field as close to ideal as possible, so each VDSL2 line can operate at its best — and deliver higher bit rates. Downstream speeds of 100 Mb/s can be achieved at distances of up to 400 m, and 40 Mb/s can be supported with loops as long as 1000 m.

Unfortunately, Mal missed this in the same article:

Next-generation xDSL advances began in 2007, when FTTx deployments combined deep fiber with VDSL2 to provide speeds up to 40 Mb/s at distances up to 400 m.

I see nowhere in Mal’s plan that vectoring is mentioned, & no provision for the extra $4.5b. Malcolm Turnbull’s recent youtube ad even states, between €250 (AUD$363) & €350 ($508) per premises just to upgrade VDSL2 to vectoring.

As per Alcatel Lucent’s own white paper, these are actual tests on 400m of 0.50mm copper in a 24 pair binder (most are 100 pair):

The key is, Mal is trying to whiteant the NBN. By claiming to be able to get 100Mbps @ 400m, he can say “look, there’s no difference, & ours is cheaper”. He’s now claiming that ALL loops will be less than 500m. I’m not sure if Mal understands: the shorter the loop, the more expensive the VDSL2 deployment. In fact, if Mal does want 500m loops, that’s around 300 000 cabinets. YES, 300 000 cabinets. Where do I get my info? The same place Mal got his info on vectoring prices for his ad: Deutsche Telekom.

There’s this continual idea that comparing current streaming rates to future rates is valid, Turnbull refuses to give up the idea that ALL video streams at a max of 6Mbps. Firstly, to only look at what’s currently available is folly. With 4K TVs starting to dip below $1000 in the US, we can expect a flood of cheap sets to hit our shores soon. So what then? Foxtel will look shit on them, FTA will look shit on them, & your 25Mbps service (as there is no guarantee of 100Mbps) will barely handle a single stream.

The key with streaming is: you never want your connection to be close to the required speed to stream. You want your connection to be 2 – 3x the speed of the stream. Why? Oh, I don’t know, maybe you want to click on a link on a website… oops, movie has to stop, buffering time. This idea you need 6Mbps to stream HD content is folly. For a 6Mbps HD stream, 12Mbps makes it comfortable. For a 25Mbps 4K stream, you want at least 50Mbps.

THE MEAT IN THE SHIT SANDWICH

From here it degraded into a funding slinging match, with neither man coming up roses. The reality is:

NBN = $37.4
NBN = (cost)

Fraudband = $20b + $13b+ + $5b + $11b + $5b
Fraudband = (cost) + (Telstra’s network) + (contract renegotiations) + (monies already spent) + (vectoring upgrade)

When we compare the two plans, we can see much of Mal’s plan hinges on not counting the backhaul that has already been completed, AND it assumes the network is free, AND doesn’t count in any of his actual claims since. In the end, the FTTN upgrade will cost more than a straight FTTP, as they’ll still have to spend $30b (as per Mal’s own background documents) in today’s money to get to FTTP in the end. Not sure how that is “cheaper” but I digress.

One thing that always strikes me is that Turnbull uses NBN Co’s lack of transparency to justify changing the technology. That just doesn’t make sense in my book. I agree, transparency is vital for a project of this size, however I don’t agree that the technology is to blame. In fact, if Mal thinks he’d do a better job, which he may due to his previous experience running OzEmail & working for Goldman Sachs, then make that your policy: we’ll run the FTTP rollout better!

Heaven forbid a positive policy happen, am I right?

Turnbull then went on to justify his $94b figure. I’m sure if we added 50% to the per premises build of your network, Mal, & added in all the funding required to deliver your promises that you’ve made after the policy was announced, it would take your policy to well over $100b. There’s just no way you can justify such childish unfounded rhetoric. I think it’s telling that 30 pages of a 34 page background document are used to talk about the current policy. Is your policy so poorly thought out that you can’t actually deliver real background information?

Mal became incensed when Albo claimed the $94b fibure was “plucked out of a Coco Pops packet”. He may be right there, as there’s no evidence to show any of the assumptions or misinformation is true. In fact, Mal’s only defence is:

We put them out more than four months ago and nobody has been able to say that our assumptions are unreasonable, or, that given those assumptions, you would not end up with a $94 billion figure.

Apparently every technology journalist worth their salt doesn’t count when it comes to Mal’s “facts”.

The funny thing is, Mal is claiming that there is a rise from $22 – $62 p/m, as per the LNP’s background document. This figure is laughable as is taken as an average price. The key difference between the LNP plan & the current NBN is that, as per Steve Jenkin’s excellent investigation, the top 25% of users generate most of the profit.

If the ARPU is $62, looking at the spread now, the wholesale price for 25/5 (the equivalent to Mal’s plan), the pricing would be well below the $38 VDSL2 wholesale price Turnbull has predicted, without reference, in their plan.

Looking deeper into the calculations, it’s clear that the maths they have done is shoddy at best, relying on childish assumptions like everyone on a fibre network pays the same, that business users won’t pay exceptionally more for “five nines” service. All these assumptions are ignored in favour of the bullshit “$62 wholesale price”. If anything, the NBN Co corporate plan clearly shows that prices will FALL (compared to today’s dollars), not increase over time. This is the nature of bandwidth, the more we use the cheaper it gets.

Another troubling aspect of this part of the debate was Turnbulls derision of KPMGs work auditing & warranting NBN Co’s corporate plan. Turnbull deriding the plan is already on thin ice, but to try & tar KPMG’s name because they don’t agree with you is childish at best.

This is made even worse by the fact that Turnbull references KPMG repeatedly in his own background document & uses KPMG’s ‘Implementation Study for National Broadband Network’ as a central piece of evidence as to the current network’s topology. Surely a company that can’t do a good job auditing a simple corporate business plan wouldn’t have any hope of writing such a study for the most complex network in Australia.

ROUNDING OUT

The rest of the debate degraded from there, with the old “connection cost” argument coming up again. For this, I will just quote NBN Co’s FAQ (point 4, first line):

A standard installation of NBN Co equipment is free of charge

Not sure how much clearer that could be.

Aside from this, I find it laughable that Turnbull is now trying to equate privatised comms infrastructure to socialism. In a way he’s correct, the sell off of Telstra is but one example: privatise the profits, socialise the losses.

I already covered the 1Gbps dedicated bullshit Mal tried to pull, & I challenge him to deliver even dedicated 25Mbps services over VDSL. Having a 1:1 contention rate would cost $1000′s per month to run. To be frank, if I could get a 1:1 contention rate 1Gbps service (dedicated) for $20k p/m, I’d get it & set up a datacentre, as I’d have one of the fastest non-ISP datacentres in Australia I reckon.

Another point Mal got wrong, to be an investor in the company & not know this is worrying, was that France Telecom are deploying VDSL & not FTTH. Nope mate, there’s actually ARCEP regulation (from 2008) that disagrees with you:

Last week, the French Parliament and the French regulatory authority ARCEP took major steps in defining a regulatory framework that aims at achieving widespread deployment of Fibre to the Home (FttH).

This includes, among others, obligations on all operators to meet reasonable requests for access to in-building fibre (at an access point to be defined), an ambivalent position on fibre unbundling beyond the private property portion, a formalised civil infrastructure access obligation on France Telecom, and a decision not to mandate wholesale broadband access over fibre.

While the last provision doesn’t force providers to deploy fibre, if requested, they must make all reasonable efforts to deploy fibre. In fact, France Telecom are committed to 60% of France being connected to FTTH by 2020. With ARCEP even stating clearly that “only a modest minority of customers will benefit” from the small amount of VDSL2 that France Telecom is deploying.

I think the funniest line comes from when Alberici offered Mal this figure:

You talk a lot about other countries, Mr Turnbull, but I would point out that the Nordic countries are pioneers and in fact just in the last few months, Sweden, the Swedish Government announced that its aim was to have 90 per cent of households by 2020 with fibre to the premises.

Mal’s response was so immature I thought he’d been replaced with someone from the Sydney Grammar School Debating Team: Well a lot of governments set those goals in very grand ways, but they’re very rarely achieved.

To brush away every infrastructure project with “but it is rarely achieved” is laughable. Yep, we don’t have the Snowy Mountain Scheme or the Sydney Harbour Bridge, they were lofty goals & were never achieved by governments. What kind of imbecile comes up with this stuff?

FINAL NOTES

The debate was a travesty, held together by Alberici the best she could with an arrogant technically illiterate Turnbull, & a shaky unprepared Albanese. I can’t say I was impressed with the format or the time frame. 30 minutes is far from enough to debate such an important issue, especially when you consider that both policies will have lasting impacts for decades to come. One positive, the other negative.

For what it’s worth, I think Mal is at the end of the line. He’s making bigger & bigger promises every week, which in turn is driving his policy final price up to the $60b target. No, you won’t be doing 1Gbps Mal, & no, you won’t deliver 25Mbps to 71% of the lines by the end of the 2017 financial year, & no, you will not be able to get lines to 500m long without building 20x the cabinets.