I read an article today that may be interesting for those who have, rather unfortunately, been misled by various politicians and "commentators" into thinking fibre is not the "best" technology for the NBN / will be outdated by the time the rollout is done / will cost a fortune to continue to update in the future / etc:
Pipe goes for 100 Gbps on PPC-1 subsea cable - Telco/ISP - Technology - News - iTnews.com.au
The article is not about the NBN, but rather is about a capacity upgrade currently being performed on PPC-1, one of the high-bandwidth fibre-optic cables that carries data from Australia to the rest of the world and vice-versa. The type of fibre technology used on this sort of cable is different to that being used by the NBN, but the upgrade described in the article still does a fabulous job of demonstrating two of the most important characteristics of fibre - characteristics also shared by the particular type of fibre technology being used by the NBN:
1) Raw speed - the upgrade will increase PPC-1's capacity to 3Tbps (3,000 Gbps / 3,000,000 Mbps), which is about 125,000 times faster than the "best case" for ADSL2+ and about 30,000 times faster than the "best case" for 4G wireless (remembering the real-world speed will be much lower than best case, too) - and this speed is delivered over a grand total of
four hair-thin strands of fibre-optic cable ~6,900KM long. On top of that, the article states that 10Tbps is possible using current technology - so the cable could be over thrice as fast again with upgraded equipment at each end. The fibre technology used by the NBN is nowhere near as fast, but this example still demonstrates that it's possible to cram well over 1+Tbps (1,000+Gbps / 1,000,000+ Mbps) over a single strand of fibre using current technology - and in the NBN model each fibre loop (strand of fibre) is only shared by 32 (or is it 64? Can't remember) houses. That's a
lot of potential bandwidth per house, if it's ever needed in the future.
2)More importantly, upgradability - and "cheap" upgradability at that. The PPC-1 upgrade increased capacity on the cable ~15-fold, all by simply upgrading the equipment that plugs into each end of it. No changes to the cable or any other "infrastructure" components - just plug-in-plug-out changes at each end. No changes to the fibre itself means the upgrade was cheap, too - it only cost "a couple of million dollars", compared to a ~$200 million build cost - so they achieved a 15-fold speed increase by spending 1-2% of the initial build cost. Again, the NBN technology is different, but the principle is the same cause it's fibre - once the cable is in the ground, speed upgrades are cheap. Spending 1-2% of the NBN's build cost in 10-20 years time to increase speeds by 10x+ nationwide, if it's needed by then, seems pretty palatable.