Well we saw this coming years ago!
Look, technically speaking most of you are right. Network contention at the RSP level was always going to be the critical problem.
I've been saying since day one the entire thing is flawed because of the incredible capital cost of the rollout which made it impossible for everyone to make money, even at the currently ludicrously low bond rate plus a bit. Someone had to lose here, the taxpayer, nbnco, the retail ISPs, the customers or all of the above.
It only takes a calculator to work out many tens of billions, perhaps a hundred billion in infrastructure costs for 25M people was madness - especially when we'd already seen the idiotic duplication of the metro HFC networks from 1995 onward with $2B+ price tags, for almost exactly what the NBN was offering at startup.
But this situation isn't new. We were paying $5-$10 per hour in the early-mid 90s for a dialup account and could pretty much max out our modems at whatever speed they connected at. Then along came flat rate ISPs who realised they could oversell their 128kbps ISDN Microlink to the network - or a E1/Macrolink if you were really lucky - to hundreds or thousands of flat rate subscribers. Why? Because the cost of backhaul was incredibly expensive. Then people were pissed because in peak time anything outside the ISP's local network was unusable, if you could actually get through and grab a circuit in the model pool at all!
All the incessant hype, particularly aimed at the non-tech market, just pushed expectations of what the nbn could do for their lives into the stratosphere. For heavens sake, it's just an internet connection, and just like in the dialup days, people don't understand network contention or backhaul. Joe Average will always pay the least they can for a consumer product, because unlike we tech peeps, that's how they think of it. And with all that nonsense and PR ... of course they think 'it's the 'NBN' so as long as I've got it, it has to be incredible.'
Don't tell me if it were all fibre it would be different, it wouldn't - just even more expensive. This is the crux of the problem and no amount of fiddling with CVC numbers is going to make a hoot of difference. The biggest players will always be able to screw the little guys on volume, especially when there is next to zero differentiation in product. This is just basic economics; we've created a residential access monopoly so by default the biggest will win, who also just happen to have pervasive LTE-A and soon 5G networks at gigabit speeds. So now we're stuffed. The customer's confused - even many technical customers - so they start blaming anyone because they just don't understand the real issue.
It's very simple. Just-in-time always wins, while massive spend technology projects inevitably fail. Why? Because innovation moves faster than the rate at which you can physically implement it.