Tuesday, 6 November 2012

NBN: Turnbull asks some good questions at CommsDay Conf.

Mr Turnbull, in his Commsday address, does raise a number of good question around the NBN Implementation and Accountability, problem is, he's so busy heckling and hectoring, that anything good he says is lost in the noise. The Oppositions' main role is to hold the Government, and by inference NBN Co, accountable: asking the hard questions

This is an attempt to not lose those questions amidst noise and irrelevancies.
... But if we simply divide through the $2.2 billion in accumulated capital expenditure on the FTTP local and transit network projected by mid-2013 by the 341,000 premises passed by fibre projected by mid-2013, the cost per premise is $6,400. ...

Is this metric unfair, given it lumps together transit and local access capital expenditure? Possibly.

But if you don’t like it, then here is a suggestion: tell me precisely how we should instead measure NBN Co’s performance and cost structure. 
Potentially good questions: where's the money going?
The Accounting and Commercial reality that Mr Turnbull has ignored is: costs cannot be known or reported before the fact, only afterwards.
This is exactly why we know exact costs/premise of completed deployments overseas.

It seems reasonable to ask NBN Co to release per-premise metrics after deployment milestones are finished. They need only report premises passed, not active, as that will take 2-3 years to stabilise. NBN Co, to be transparent, should release two sets of costs: overheads not directly attributable to services in a deployment and direct costs related to per-premises deployment.

Still, Senator Conroy might take note and address this.
.. questions about selection of CEO, Board and their seeming lack of Telco experience. [not quoted, too long]
Seemingly good questions, but I think not useful.

Mr Turnbull criticises the NBN Co CEO for "only" having Vendor experience and never responsibility for a large deployment. And the Board for only having one member who's ever been involved in Infrastructure companies.

The Opposition should be holding the Government and NBN Co accountable for Governance, competence and selection of key individuals: a richly deserved question.

In the start-up phase, the NBN project has nothing to do with operating Infrastructure, even building it. Initially their job was only about "doing the numbers", letting tenders and negotiating contracts with vendors and the regulator. Seems the Board was perfectly chosen for that.

As we move into a deployment and operational phase, then relevant experience within large Infrastructure businesses is increasingly important. We have precisely zero local enterprises with large-scale Fibre Broadband rollout or operational experience, so candidates with non-Telco backgrounds are needed. One board member has been appointed with experience in Water Infrastructure.

Seems a good move to me, and that over time we should see more appointees like this.

As for Telco experience, we just have to cast our minds back a couple of years to Sol and his three mates... That went well for Telstra, didn't it? The share price collapsed, employee relations became toxic and customer service became worse than useless. From the promises unfulfilled and the damage left behind, you'd wonder why the four were ever imported and allowed to carry on... They did leave very happy men with outrageous pay-packets and terrific line-items on their CV's, but with no friends amongst the shareholders, possibly within the Industry, but I'm not privy to that circle.

I think any Australian Telecommunications Minister whom chose to repeat this decision in the next twenty years would be on very thin ice with the electorate. How can Mr Turnbull forget that?
And of course, under FTTN, the most fault prone parts of the copper network—the bundles of copper that feed into exchanges, not individual access lines—would be replaced by fibre. Grahame Lynch, Commsday, 22 Aug 2012. (attributed by Richard Chirgwin to Dr Paul Brooks, ex-CTO TransAct).

I have never seen this point acknowleged by the likes of David Braue, Nick Ross, Renai Le May or the other so called specialist commentators in this space. Or by Alan Kohler or John Durie.
It's a good point if true, but sounds wrong... Changes the economics of his FTTN proposal a little.
He's saying the big bundles in the most protected environments are the most fault prone.

I'd have thought the smallest cables with least protection, literally the last mile, would be the most fault prone. I've only my personal experience to go on and that's hardly representative or definitive.

These would be very good points to confirm or refute:
  • Is copper-line fault data available?
    • Preferably current, but even historical.
    • Summary data is better than nothing.
  • 75% of subscribers are over 2.5km from the Exchange,
    • or for 15,000 out of 20,000 subs on a Local Exchange, there are more than 2.5 kms of "big bundle" cables (100-500 pair).
  • What's the maximum distance from Exchange to Endpoint? 8Km max?
    • What's kind of cable-plant reduction for an 0.8km rule are we talking?
  • Are there good Telstra ducts leading back for the cheap deployment of Fibre to replace the large cable bundles?
    • There's no significant cost advantage if many km of trenches have to be dug to lay new fibre for all those nodes. 
Perhaps Senator Conroy and David Thodey could jointly release this data. It is quite pertinent to the debate.

No comments:

Post a Comment

Note: only a member of this blog may post a comment.