NBN Co says Melton test not valid, likely uses processes developed there anyway
Reg gets its claws into leaked report of FTTP rollout experiment
Far from being an “unrepresentative” document based on bad data, the NBN Co internal presentation which leaked last week was the outcome of five months' work, instigated to create a benchmark for future fibre rollout areas.
The furore – Vulture South could even go as far as to call it a kerfuffle – began with the leak of an internal review to the Sydney Morning Herald. That review detailed a set of metrics derived from work in a Fibre Servicing Area Module (FSAM) in the Victorian town of Melton.
(In response to comments on our previous story, The Register notes that the FSAM was not covering the whole of Melton.)
As The Register noted, NBN Co felt obliged to issue a media release (here) refuting the Herald's conclusion that the fibre rollout could be “delivered faster and for less money than originally forecast”. NBN Co's refutation claimed “the work underway in Melton delivered no such conclusions.”
It's important to note that neither the Herald nor anyone else claimed that the Melton trial showed fibre would be cheaper than the “multi-technology model” – only that NBN Co had assessed how the fibre rollout could be improved, and obtained specific measurements.
The Register has now viewed a copy of the document, and put the following questions to NBN Co:
- On all but two of the 12 scorecard items agreed between NBN Co and stakeholders as criteria for assessing 3MLT-10, that FSAM was considered a success, and on three scorecard items relating to cost, 3MLT-10 was between 23 per cent and 43 per cent lower than the 3BRA average. Are either of these assertions in dispute?
- Regarding NBN Co's claim “The efficiencies that our construction crews had applied to construction in Melton - such as smaller diameter cables and smaller multiports (or splitters) - are already being employed in the NBN build across Australia”, why did the “3MLT-10 Deployment Trial” document make five recommendations for broad or national adoption of work practices and / or technologies from the deployment trial?
- NBN Co said “the area of rollout in Melton had some unique characteristics that inevitably would have brought down the cost of the build there relative to the project as a whole”. Why, then, was Melton considered suitable to serve as “an internal baseline / benchmark” as stated in the document?
- NBN Co claimed the Melton work “also required more oversight and more resources than usual to make it a "success".” Is this relevant, given that on all price benchmarks, 3MLT-10 was cheaper than existing averages?
Here's how NBN Co's press peeps responded to our questions about the document:
“... that we have identified efficiencies is not in question. The power point that you are referring to is the issue. It is not endorsed by the business - the methodology and metrics from one unrepresentative rollout site had a range of shortfalls that meant the presentation did not pass muster.
We look for efficiencies at every opportunity, across all technologies. It's not only our obligation to taxpayers but it will ensure our success delivering the NBN within the time frame and budget envelope we've been given. We are committed to doing this regardless of technology.”
In other words, rather than address specific questions, NBN Co has chosen to denounce the document.
The Register is in no doubt that the document in question is “not endorsed by the business”. The business is horrified that the document escaped. However, Vulture South is unconvinced by the argument that the presentation “did not pass muster”.
The leaked document makes it clear that NBN Co devoted more than three months to work in the area designated “3MLT-10”, and that the project was designed to test specific benchmarks covering not only cost, but other characteristics such as technologies in use, work practices, work management software.
It would also seem to The Register that NBN Co's claim that it has “identified efficiencies”, and that its efficiencies are “already being employed” across Australia lend authenticity to the document, since the document in question explicitly recommends that efficiencies be deployed on a national basis.
It's also clear that the 3MLT-10 trial was conducted for a specific purpose, one that pre-exists the document in question. It was initiated, the document states, “to create an internal baseline / benchmark” for various aspects of fibre work (including cost).
It may well be that the network builder was unhappy with the presentation. However, if the FSAM in question is “unrepresentative” of the country as a whole, it raises a new question: why was that location selected as a benchmarking pilot if it could not be used as a benchmark?
This saga still has legs in it. ®