Hmmm interesting but mostly for the point that someone at NBN co has looked at this issue and not arrived at why big downloaders might be paying for slower speed plans.<p>I used to talk frequently to the Telstra techies who used to come out and fix landline issues, but I haven't done so in recent years when for whatever reason NBN Co techs were also being shunted into regular telstra line issues ... every time we had to deal with them my way, often completely clueless, scratch the job as being done but phone lines still down - they were contracted apparently and the deal laid out for them, well it was NBN Co II.<p>The problem I see is there was NBN Co I the original with an aim to do what it was supposed to do, lay fibre where they could, and do away with the ... well words really can't adequately imply how stuffed the copper is in my parts, but in the last decade or two the Telstra techies had worked miracles. Then arrived NBN Co II which was heh for a better word, a foolish endeavour to pretend some of the copper was still good for it, as it would save money, be cheaper over all, and ... changes meant that the new NBN Co techs were told to use "butterfly connectors" which the previous Telsta techs had finally convinced the top level the connectors weren't up to the task, even the best effort usually resulted in corrosion and noise in a matter of two or three years (according to the Telstra tech telling me on the quiet) who also went onto say another large regions where people had been paying for a cheap asdl+ suddenly had a more expensive NBN connection not much better than dial up. I would say that recent changes at NBN Co during the last year, I should probably refer to it as NBN Co III.<p>Of course speed is a pressing issue, it costs more for the company providing it. If the connection turns out to be a bit of a dud, then a customer, after complaining and getting nothing, is inclined to use other services which allows downloading at a slower rate. Since they're paying more in total, some are motivated to get their money's worth.<p>I think what NBN Co will soon find, once they introduce the change, those who've put up, will demand the bandwidth they're supposed to get ... it'll cost NBN Co more in the end if they can't find some new way to redefine the connection plan.
The NBN is a "GPON mostly" national network in Australia. It is also called MTM sometimes, Multi-Technology Mix. That means the highly politicised decision to not only do rural and remote over WiFi and Satellite, but to stop rollout of fiber in the metro areas and cheapen things to use Cable and VDSL. ISPs can become RSPs in NBN lingo and sell/compete for services over a common Layer1/Layer2 fabric. The cost of the national fabric is being recovered by defined bands of charges and costs associated with the sold services.<p>All up it's $50b to $80b of capex and opex.<p>A reminder that GPON networks generally have no strong relationship between the bitstream per house and the cost of service: the NBN decided to commodotize both speed and volume towards (speedy!) cost recovery of the national layers 1-2, the "pits and poles" cost of a national fabric.<p>But the splitters and signal handling boxes usually don't incur more cost simply because people use them.<p>Where G/PON has contention on bandwidth is in the return channel because some designs shaved costs by limiting the opportunities for symmetric bandwidth. Bearing that in mind a "leech" typically is downloading. Its "seeders" who saturate the up-link component.<p>The NBN includes non-fiber components which do have shared return channels too and have higher costs (old cable plan, VDSL) but again, there is very little component of cost here which depends on the amount of data you send or receive.<p>Arguably there is an energy cost, and an encoding cost (especially in VDSL) but thats nothing like the end-user service charge cost.<p>TL;DR this is a charge-model cost recovery business problem and not a technology problem. The technology problem here was deciding to go cheap. Thats a political problem.<p>Rural and remote was always going to be hard. Over the longer timeframe some of them solve by more fiber, but also LEO is changing the shape of the problem.
NBN TLDR: Lets build national fiber network! Who better to lead it than old incumbent telco execs! Oh, but lets reuse old copper links, why waste resources on modern fiber. And lets not forget about data caps on internal traffic, those bits arent flying for free in out umm own err network.