[sethhochberg]

These solutions for things like IP-over-copper-pair are typically designed for situations where you already have the cable buried or strung on poles or whatever - so the choice isn't whether to lay new copper or lay new fiber at roughly equal expense, its really whether to use the existing copper or lay the new fiber at much greater expense

[jagger27]

Thank you, this wasn’t clicking for me until I read this. Of course there are countless places with a dead phone line hanging between them.

[hinkley]

Or a fat old conduit that you can run a new wire through...

[jagger27]

I can't imagine a scenario where you already have a 1km long conduit that doesn't already have copper or fibre in it. Why would you choose to push a kilometre of brand new copper twisted pair line down an existing conduit just to squeeze 10 Mbps out of it? The hard part is already done and fibre equipment is pretty cheap.

If the conduit already has a phone line in it that isn't being used, sure.

[hinkley]

In the context of the current thread, I'd say that's fair. If you're talking urban infrastructure, my understanding is that things are good until the existing (fairly large) conduits fill up, and then everything stops until someone digs up the street for blocks and blocks.

Do you happen to know, for these long conduits is it typical to do a solid run or to put access panels at regular intervals? If you're running a new cable are you actually trying to run a 1km snake through a pipe, a 100m snake 10 times, 50m x20, or something in between?.

I've never seen anyone run a 1km cable, but I've seen buildings be rewired with tens of kilometers of new cables a couple times.