Re: BT's G.fast copper carcass snake-oil tech.
Finally, someone (albeit anonymous) confirming what I've been saying for years since 2009, given all the backlash I've had over attempting to technically explain BT's copper carcass snake-oil tech.
G.fast is a can of worms to maintain (which end users pay/BT profit from, due to Ofcom's regulatory model of stupidity, of allowing profit from failure), making G.fast more viable on paper, because it's less robust than FTTP from interference, both malicious/non-malicious, i.e. BT paid call-outs potentially rise statistically with G.fast rollout/deployment over FTTP.
Little reported, but the initial field trials used isolated brand new copper cabling (separately run alongside to those used in existing local loop copper), to rose-tint the results of G.fast. Ex-BT are as a routine, parachuted into key Ofcom jobs, you just have to look at their Linked-in profiles to see this.
G.fast may have its uses, multiple occupancies - new town tenement blocks (Marchmont, Edinburgh) are a prime example, but that's about it, but even then the costings are controversial depending on which side of the fence you sit. FTTP v G.fast.
The number of G.fast nodes+mains grid connections (to actively power a G.fast node) required rises exponentially. It's fairly simple maths, which means it soon loses any advantage (if it had any over passive both overhead/ducted FTTP rollout).
Without shortening the copper lengths, G.fast helps no one (being generous here) with copper cable lengths of >500m by length, 250m as the crow flies), that figure is ball-park nearer 300m-350m (150m-175m).
G.fast all has to be ripped out and replaced with FTTP anyhow (so why not start now?) when you've sweated it to the max, speed improvements hit an upper limit/brick wall, like attempting to fold a sheet of A4 more than 5-6 times.