[raldi]

> Should we really be able to force someone to use their property an a way that serves the public interest?

Yes, that's the price of a government-granted monopoly.

> If so, where do we draw the line?

If you don't have a government-granted monopoly, you don't have to be neutral. For example, that's how the Internet backbone works.

[steamer25]

Instead of trying to get politicians and bureaucrats to thoroughly understand the technology and hoping that they'll steer these monopolies in a favorable direction, why don't we demand that said monopolies be dissolved? I.e., if the terms of the deal are going to be altered, why not declare that there may be two or even several players? Also, why can't this be resolved at the regional level? Why does Netflix have to be as uniformly slow as obscure-and-poorly-optimized-cat-pics.com across the entire country?

[arg01]

One of the problems is they're natural monopolies. You only need the one data link and so it's (arguably) more cost effective to have a single regulated monopoly on the last mile side. You can create hybrid systems (like the UK) where other businesses buy capacity wholesale from the monopoly provider and compete over the same infrastructure.

[steamer25]

The same could be said of the grocery stores or car dealerships in my town. The problem is that as soon as you say it'd be more efficient to have all our eggs in one basket, the human nature of the carrier kicks in and says, "Hey I've got it made now. I can take it easy because no one's allowed to compete with me". Unfortunately, hiring some delegate who's multiple-times removed from the consumer/voter and is likewise unmotivated to improve the situation doesn't change that.

Yes doing the trenching and tunneling through various neighborhoods is disruptive and requires heavy machinery, etc. but that's just to create a hole in the ground. Once some conduit is laid (as a public good by the municipality?) it should be relatively easy for new carriers to come in and pull fiber or what-have-you. You just have to make the channel big enough for e.g., several cables and then you lease the space to the x highest bidders for five years at a time where x > 1.

There are probably better ways to finance things and minimize the disruption, etc. but overall it doesn't seem infeasible to me.