> How do those untrenchable locations receive water and sewer? If water and sewer can be installed underground, why not fiber?
I am not a civil engineer, I only drink with them occasionally. I'd imagine there are some big differences between heavy pipes in various states of liquid internal pressure above the frost line and bundles of relatively lightweight cables, but I'm very much just guessing at that point. I do know that there are lot of processes in place to plan that sort of work and it can get very expensive. (I see some of my sewer bills. Though the city is also unfortunately in the middle of several huge combined-sewer-overflow rework projects.) So yes, cost would be a big factor, but I think the cost is partly such a big factor because of the physics of all of it and planning it in safe ways that don't damage the neighborhoods, at least that is my very lay understanding of it.
> Google apparently didn't even want to spend the expense running it on utility poles, which frankly, I would naively think would be the cheapest option.
That's the part that still most boggles my mind. That was by far the cheapest option than a bizarre R&D for "exotic new epoxies" and all the money spent in digging up and then replacing asphalt. At one point (after the epoxies failed) they pawned a lot of that cost of the replacement part off on the city by picking up the city's normal pothole/repavement maintenance plans and dredging their nano-trenches just before, which was "smart" cost cutting in an evil sort of way.
Not only was using the utility poles the cheapest option, it was the planned option for years. It was the option that the city spent millions of tax dollars litigating for free for Google's part to make it even cheaper, about as cheap as it could possibly legally be. Google didn't pay a dime for those legal challenges. The city won those court cases! They had the go ahead to Google to do "just about whatever they wanted" with city utility poles and Google decided to do "Plan B" at the 11th hour after the go ahead.
I guess some impatience comes in to play, because it did take years to get that "go ahead" and I guess Google thought a "productive" use of those years was to experiment with silly glues and then they accidentally sunk cost themselves with R&D money into going with that Incompetent Plan B even though Plan A was still on the table and extremely viable and cheap. It's still hard not to also call that impatience its own form of incompetence, though, because that R&D didn't seem to take much reality into account. (Glues? In streets? Streets that experience weather? Streets that Americans drive heavy cars, trucks, and SUVs on? Really? They wasted how much money on that?)
