Nutlessness as a service

Any tech journalist who discusses Malcolm Turnbull's "policy launch" tomorrow without quizzing him on it acting in the service of politics without the chance to examine the policy-maker, and is a fool.

Whatever else we know about Malcolm Turnbull's broadband plans, they represent brains without balls.

Brains tells us that everything you say about broadband in Australia is an attack launched on a government that can't take a trick, even with fourteen trumps in its hand.

Balls?

There once was a man from Devizes

Whose testes were two different sizes

The one was so small

It was no ball at all

The other won several prizes.

When it comes to a broadband policy launch, is Malcolm Turnbull winning prizes?

Don't make me laugh: he's keeping the gig to the morons, partisans, gourmands, time-servers, place-seekers, and dyspeptic dipsomaniac brain-donors of the press gallery.

Having handed a gift of a policy that's already risible – virtue of its red-handkerchief $90 billion – to a witless goon at the Daily Telegraph, Mr Turnbull is now ignoring every approach to journalists with, you know, experience in the telecommunications market.

He's quite free with his advice, criticism and public denigration of journalists that don't agree with him. I've had “why won't you call British Telecom?” (answer: I did) and is happy to tell me what I've misunderstood, forgotten or added up wrong, should I dare to run a spreadsheet over the NBN or any similar project, anywhere else in the world.

I've offered to quiz him at any public press forum he chooses, and never once received a reply.

With 25 years experience in writing about the telecommunications industry, and – if I include my original training at the Overseas Telecommunications Commission – close to 30 years all up around telecommunications, I'm not on the invite list for Malcolm's Great Broadband Policy Launch.

Because, unlike the lazy shit-for-brains that infest the gallery, for whom all politics is reduced to “the government said / the opposition hit back” or “the opposition said / the government hit back”, I am both numerate and familiar with all of the technologies that are proposed for modern broadband.

In other words, people like me – and I have – need to suffer his lame lectures about journalistic technique if we don't agree with him. He'll sit at the door, scratching and whining like last decade's hound that we're not going to the original source (that is, sources that agree with him). But we're not permitted to quiz him in a public forum.

In his broadband launch, Malcolm is going to be an original source – but only if he can share his fleas with other superannuated hounds.

What a lame joke: and this is the best brain in the alternative party of government.

What a joke.

Work that “$90 billion” figure, printed without question, query or even the simplest mathematics by the Daily Telegraph, the utter suckhole of journalistic suckholes in Australia, a place even facts won't use as their hospice because it's better to die in the gutter than have your brains sucked out by News Limited lizard-people with skull-drills and straws.

On today's number of households in Australia, that's well past $10,000 per household.

That number is so stupid – strike that, so utterly stupid – that any journalist worth the price of the headline and a Starbuck's coffee should have asked even one question:

“Since British Telecom is only asking a couple of thousand per kilometre for individual node-to-home connections, on the most inefficient model that human ingenuity can devise, what idiot gave you a convincing estimate of $10k per household for FTTH?”

Or even better: “Why did you believe an estimate that is many times every available international benchmark for FTTH rollouts, based on what's already been spent?”

Or even better: “This is a naked political lie. Put up or shut up.”

News Limited doesn't work that way. Its game is this: we will only run with the naked, unadorned press release, if we get it first. Go off to its IT pages and tell me I'm wrong: but only if you first poll the PR companies who know how the game is played. Give News a 24-hour break; that's all you need.

And it can even be utter drivel as ran in the Telegraph today.

Which is why experts aren't going to be welcome at Malcolm's little piece of theatre. It only took a few seconds with a spreadsheet – but the brain donors don't do numbers, they do “he said / he hit back”. Which is exactly the wank-and-dribble level of intellect Malcolm Turnbull wants covering his policy launch.

The one was so small, it was no ball at all

WiFi NBN not fibre: PLEASE MAKE IT STOP!

Since we're talking about the NBN OH GOD PLEASE MAKE IT STOP the next time you decide to make a serious business decision based on a Business Spectator story, take a look at this story first.

Don't want to? OK, here are the high points: instead of the NBN, build fibre to the street, and do the rest with WiFi.

“WiFi is getting to the point where it could be used for the last bit of the network that connects households and businesses to the fibre. This is what should be the focus of an "open and transparent debate”.

Did I already yell “please make it stop”? Yeah, no need to repeat that bit.

The short version.

1. WiFi hotspots are shared: securing an individual's traffic is not trivial.
2. WiFi hotspots are shared: one user can swamp the network.
3. WiFi hotspots suffer interference from other WiFi hotspots – they use shared, unmanaged spectrum.

Read on:

“The NBN fibre could be brought to smaller neighbourhoods and then connected to each home and business using new powerful WiFi options, specifically little things called "picocells”.”

A picocell isn't a WiFi hotspot (PLEASE MAKE IT STOP!), it's a little mobile phone base station designed to connect to the fixed network and eliminate in-building cellular blackspots. That is: if there's a blackspot inside (say) a shopping centre, Telstra or Optus or Vodafone might decide it's worth bringing in a picocell so people can still use their smartphones.

It's got nothing to do with providing last mile wireless access.

Now, if you have the IQ of a fence-post and spare cash, go out and spend a couple of billion on a WiFi last-mile network, on the advice of Business Spectator, and go broke. I'll be laughing in the wind-up proceedings.

FTTN vs FTTC: my methods and decisions

Since it’s been criticized, here are some personal notes on the methodological choices I made in estimating the rate of BT’s FttC rollout for the ABC in an Australian context.

Why use exchange areas instead of population?

1.    Verifiability – While it would be possible to verify BT’s household-coverage claims, it would need someone willing to part with around $10k to buy a data license and carry out the analysis.

The number of cabinets installed by BT provides no base for comparison: Australia doesn’t have a design yet.

On the other hand, the number of exchanges serving VDSL2 is published in the UK, and ADSL2+ in Australia.

The only assumption I made here is that a hypothetical FTTN would need to replicate ADSL2+ coverage to satisfy the public.

2.    Geographic equalisation – Using exchange areas equalises the vexed question of the geo-demographic differences between Australia and the UK, an issue which I’ve worked on for six years in various projects.

Britain and Australia have nearly the same number exchanges – 5,000 or so – yet the UK’s exchanges serve 40 million households, Australia’s 7 million. The UK’s exchanges are bigger (in terms of lines served) but smaller (in terms of geography served).

Why did I select 2009 as the start date for the UK FttC, and 2010 for the NBN?

That’s much easier to answer: the deployment of commercial pilots provides an equivalent date for the two projects.

The NBN has proceeded very much in the public spotlight – the implementation study, the establishment of the company, the design decisions, the maps, the appearances in Estimates and so on. We knew a lot about the design process before the launch of commercial services in 2011.

BT’s process is a black box, by comparison. We do know that its first commercial pilots occurred in 2009, but between that date and the official launch of commercial services, we know very little about how much work went on. For example:

-    How many cabinets were ordered, delivered, and put through acceptance testing?
-    How many locations went through site preparation so as to enable rapid installation when cabinets were available?
-    What resources were devoted to network design, site selection, training, contractor selection, and ramp-up planing?

And so on.

Without an understanding of the pre-launch effort, I think it’s quite reasonable to use the commercial pilots as “Year Zero” of BT’s rollout since I can be certain that all that “underground” work was going on at least as early as 2009, prior to the commercial service launch in 2010.

Since everything else was based on public sources linked in the article, I don’t feel the need to explain the other sources further.

About that British Telecom Fibre-to-the-cabinet case study...

Since British Telecom’s fibre-to-the-cabinet rollout is the example Mr Turnbull favours, let’s take a look at it.

Question: Does the rollout deliver high speeds?

BT currently claims a maximum of 80 Mbps over copper. As always, with a copper-based technology, “your mileage may vary”.

Question: Can customers upgrade to fibre?

Answer: Yes, in some cases.

Question: What’s the fibre-to-the-cabinet reach?

Answer: Currently around 60 percent of the UK population.

Question: How many exchanges are required to deliver that reach?

Answer: Around 1,300, if this list is accurate: http://www.samknows.com/broadband/exchanges/bt/fttc.

Question: How long did the rollout take to reach 1,376 exchange areas?

Answer: Around four years. Call it 340 exchanges per year – at which rate, Australia would have VDSL2 services replicating current ADSL2+ services in, oh, about 8 years. Give or take. Your mileage may vary…

The absolutely inescapable point of this little conceit is simple: Australia is not the UK. Pack 60 million people into Victoria; concentrate them tightly enough that you still have farmland and (in Scotlant) tundra; deliver broadband in a way that your rural users complain they can’t get it, and then use that as a model for Australia.

As I said, if Australia were to match the UK’s rollout rate, we might match its KPIs, but the actual Liberal Party target for fibre-to-the-whatever would take nearly as long as the NBN.

But I don’t blame Malcolm Turnbull. I blame the supine, spavined, sleepwalking slackness of journalists who by hell should be smart enough, Google-savvy enough and care enough to actually test the case study. The research took me a little bit of down-time, not enough to justify asking someone to pay me for – which is why this is a blog post instead of an article.

Off yo' asses, my colleagues. Research or retire.

Will the coalition’s FTTN get people off dial?

This is odd: There’s a considerable gap between two different ABS measures of the number of dial-up Internet customers still in Australia.

According to the bureau’s Internet Activity, Australia, there were 379,000 household dial-up subscribers at the end of 2011. However, according to the Census data (collected in August 2011), there were only 235,000 households reporting themselves as dial-up Internet users.

I suspect that there are a lot of people who kept a backup dial-up account at some point, and who haven’t gotten around to cancelling it.

This is an aside, however. 

This post is in the context of Malcolm Turnbull's initiative to survey users on broadband speeds.

I was interested that Turnbull's concerns include the dial-up customer (which led me to notice the data discrepancy). To quote Tony Eastley’s introduction* in AM, “the opposition says close to half a million households still use dial-up technology, and they’re being forced to wait too long for the NBN.”

Given the speed of dial-up’s decline in Australia, I doubt it. The dial-up sector lost nearly 100,000 customers in the six months to December 2011 anyhow – if that rate is maintained, it will die of its own accord in a couple of years.

(I would guess that most of the remaining 90,000-ish businesses with dial-up accounts use them only for backup, by the way. That application is being replaced rapidly with mobile connections.)

My other point is best illustrated by a rough map:

This shows the Census SA1 areas in which more than ten percent of the households still have dial-up connections. I haven’t calculated exchange distances, but the illustration should tell you enough.

Dial-up is at its most stubborn in remote places – areas that would be the slowest, most expensive and difficult to upgrade to FTTN.

*Note: it's perfectly possible that AM misunderstood something Mr Turnbull said, but that the program didn't put to air. If so, my apologies to Mr Turnbull.

Bleating about the NBN opt-out again

-->

(Cue background music: “One of These Things is Not Like The Other, from Sesame Street)

Everyone’s up in arms about the “opt out” NBN, it seems. Here’s some snippets.

“The three biggest states have opposed the move by NBN Co to install fibre-optic connections to homes and businesses without asking permission” – The Australian Financial Review.

“The opposition communications spokesperson Malcolm Turnbull said the shift to an opt-out policy confirms that the NBN business plan depends on compulsion” – The Australian.

“Victoria’s … technology minister has confirmed that the state does not support the introduction of Labor’s preferred ‘opt-out’ model for the National Broadband Network” – Delimiter. 

“If Australian consumers want a fixed line for telephony or Internet access, they are going to have to use the NBN’s line” – Malcolm Turnbull on his blog. 

Customers “will be forced into connecting to the national broadband network (NBN) unless they ‘opt-out’” – The SydneyMorning Herald.

How many of these stories have a 2012 dateline? One. The rest are from 2010.

For whatever reason, The Australian Financial Review treated a years-old government policy as a Major News Announcement, duly handing a stick to three state governments, pointing to a piñata called “The NBN”, and letting them swing.

I wonder how the premiers would feel if the Federal Government challenged their water utilities’ rights to fix sewer mains running under private backyards?

Blowout? What blowout?

-->

Here’s a forecast: there will be more delays in the NBN project – not because the project is impossible, nor due to the incompetence of anyone, but because it’s a very large project, and construction delays are almost inevitable.

Not that this will help Mike Quigley or Senator Conroy in any way: as with the budget, timing is now considered nailed down by a press whose project management skills don’t have to go far beyond the next deadline.

As for the relentless use of the word “blowout” – spotted in the wild at Fairfax, News Limited, the ABC and even TheConversation – what a bunch of dills. The new figures from NBN Co diverge by just 3.9% from the company’s forecasts: in the world of accounting, this would be considered a high accuracy forecast.

In looking at the NBN corporate plan, please keep in mind that there’s a difference between “premises” and “households”. The NBN will pass 12.7 million premises – according to the forecasts in the corporate plan – but that’s not 12.7 million households.

Let’s take a guess that there’ll be 11 million households by the end of the build (there’s about 7.7 million now). The rest of the premises passed are not households – they’re businesses, schools, hospitals, and so on.

This might seem like pointless pedantry, except for this: if some enterprising analyst sits down to try and run their own projection of the NBN’s income based on 12.7 million households, their numbers will be out by around 15 percent. That kind of puts the 3.9 percent everyone’s worried about in the shade.

(Actually, the error will be bigger than that, since the 1.7 million premises getting miscalled households would actually be the kind of premises that spends more than a household. But I’m not about to try and put estimates to that!)

Fixed line decline: real but over-stated

-->

This post arises because of a Twitter conversation, in which people were speculating about whether political polls have become skewed because of the decline in the landline. If the pollster can’t call you, you’re left out of the sample, and there’s so many no-landline homes now it must destroy the accuracy of the poll, right?

Wrong, I suspect.

Because it’s a Sunday morning, and because nobody’s paying for this, I don’t propose even trying to start crunching the numbers. But I believe that the “landline decline” is frequently overstated. So here are some thoughts on the factors that are left out of the story.

Telstra’s PSTN decline

There has been a fairly dramatic loss of basic PSTN services revealed in a decade’s worth of Telstra annual reports – but that’s not a reliable guide to the number of “no home phone” households. Most of the services disconnected since 2002 represent the consolidation of multiple lines down to a single service. The drivers for this are outlined below.

Broadband – once ADSL became widely available, nobody needed a second line for dial-up Internet any more. That drove a five-year drift in the number of services, starting about 2001-2002.

Faxes – there are still more fax machines out there than the people think, but they have been declining steadily for a decade. A business that disconnects a dedicated fax line contributes to the loss of basic PSTN services.

Business IP telephony – SMEs routinely bought multiple incoming phone lines (I suppose many still do). However, each time a business buys a data service connecting to its IP telephony server, it eliminates an unknown number of individual PSTN services.

Naked DSL – if someone abandons their Telstra landline service and buys a naked ADSL and home phone combo from iiNet, TPG or whoever, they fall out of the Telstra data. However: they might still have a phone number visible to the PSTN.

Cable telephone services – Consumers who connect a PSTN phone to their (say) Optus HFC service don’t count as using a basic telephone service, so there’s a few hundred thousand services that are often forgotten by analysts prepping numbers for the media.

The Naked DSL and HFC services exemplify another difficulty: if you have no Telstra fixed line service, but you do have a PSTN-visible phone number, how should you be counted?

With all of these factors coinciding, it’s quite feasible that Telstra could shed a couple of million basic PSTN services without a single household opting to abandon the fixed phone completely.

The all-mobile individual

Most surveys I have seen identify all-mobile individuals – and there’s a fundamental problem with using this to estimate the number of households without a PSTN service: the individual isn’t the household.

It’s quite feasible for someone to say “I don’t use a fixed line” while still living in a home that has a fixed line – because they’re not the person paying the bill. This is especially true in a world in which young adults no longer leave home at the first opportunity.

The reason I can’t be bothered properly crunching the numbers on a Sunday morning should now be clear: anyone who wants to work this out properly needs a week or a month.