Hacker News Reader

Comment by gz5 | Hacker News Reader

The only way to get the last mile competition we need in the US, and make net neutrality irrelevant, is for org-run fiber builds in which the org doesn't offer services. The org operates a multi-tenant fiber infrastructure in which you and I can pick (n) "service providers"...part of what I am hoping is a 2018 full of decentralization and distribution: https://goo.gl/DkpmU5

View on HN

Or just unbundle the local loop as we do in the UK

yes although can be harder to retrofit, e.g. i don't believe many of the US fiber architectures were built to be multi-tenant with (n) service providers behind 1 operator. especially because the duopoly's control most of them. enter the muni-fiber and similar projects...but built to a different commercial model...

Well tough luck mate (you want to play in the big boys game ante up sub) Any how you can use ADSL variants or upgrade to FTC and use VDSL for the last drop.

For the urban, and dense suburban areas fiber is a great option and has much higher potential. But there will always be a large portion of the US that is not profitable to run fiber, and already has phone service. We already have multitenant access in the US for all phone service. Just extending the current regulation to cover DSL is all that is needed to allow for competition and improved internet speed. You can get up to 76mbps with VDSL2 over copper wire, just modern modems and provider side equipment. And there is newer technology that provides even better speeds over the same copper wire.

In the less dense areas the most economical solution for great connectivity is definitely mesh networks. They are even being deployed in very rural areas at a very low cost! http://owni.fr/files/2011/09/Building_a_Rural_Wireless_Mesh_...

Obviously there are better (commercial, prepackaged) mesh network solutions

Mesh networking hasn't really worked out. WISPs, i.e. Wireless Internet Service Providers, do a great job at serving rural (and other) areas.

Hmmm I think it's a bit of a stretch to say mesh networking hasn't really worked out. It just hasn't spread far enough yet.

Have you heard of CJDNS[1]? or BATMAN[2]?

Hyperboria[3] is multiple sizeable deployments of CJDNS. Maybe you heard of Freifunk[4] in Germany, with over 40k access points so far.

[1] https://github.com/cjdelisle/cjdns

[2] https://en.wikipedia.org/wiki/B.A.T.M.A.N.

[3] http://hyperboria.net/

[4] https://en.wikipedia.org/wiki/Freifunk

Yes, I've heard of them and others, but I think dboreham already put the matter conclusively at rest:

https://news.ycombinator.com/item?id=16138402

For buildings (offices, condos, apartments), air fiber solutions are way better and often faster than fiber pulls. 60 GHz microwave, building-to-building if needed, all are in the big cities... why waste time fighting rights to run fiber in dense urban cores?

Define way better. Wireless solutions have many benefits, but they are in no way better in all aspects than fiber.

Fiber has its place, as has wireless.

There's no way there's going to be subloop unbundling after the Supreme Court killer local loop unbundling.

It could happen just for DSL. The big ISP don't really want to service rural areas anyways.

Could happen how? The big ISPs don't want to do it voluntarily and the Supreme Court says they don't have to. Just because the ISPs don't want to serve rural areas doesn't mean they want anybody else to either.

As an illustration of a related inclination, the FCC, pushed by the big telcos—who also tend to be big ISPs—under Ajit Pai killed the use of resellers (non-facilities-based providers) in the Lifeline program, despite the fact that 70% of Lifeline customers used them, often because the facilities-based providers aren't offering them service directly. This got overlooked by a lot of people, because what attention was on the FCC was all about net neutrality.

Already had that in the US until the Supreme Court decided we shouldn't have that in the Brand X case.

It's very, very unlikely to come back.

Yeah there aren't enough suckers willing to invest in firms to "compete" with the monopolists to fall for that regulatory bait-and-switch again.

The FCC did a terrible job at rate setting and ended up killing DSL technology in America. That's why cable companies are typically the biggest ISP in any given location instead of the telecoms.

Open access networks are the way to go. It keeps all the incentives in the right place to allow for capital investments to improve the network, prevent overbuilding, and encourage competition.

Another pie-in-the-sky option: pick a UHF television channel and de-license that portion of the spectrum. The IEEE would develop a standard (like what was done for Wi-Fi) that lets anyone be a local broadcaster or service provider

We already have this. It's called white space technology and has so far amounted to pretty much nothing. The technology is immature and expensive plus it provides very little bandwidth.

What segments of the spectrum allow this technology for public use? How do those compare to what ATT/VZN own?

https://www.fcc.gov/general/white-space

Cool, thanks for the heads up. After wiki-ing around, it looks like current whitespace technology is hobbled by forcing the broadcast device to get permission from an FCC database in order to transmit. With that constraint, I am not surprised it hasn't taken off.

That leaves nothing for those "service providers" to do except skim profit. Why should they get a cut? Because they have a router?

My question is an honest one. A modern ISP needs to only run a router, really. They don't need to provide email accounts. They don't need to provide a Usenet server. They don't need to provide free web hosting space. They don't need to provide hardly any of the things ISPs once did. They don't even need to provide DNS servers. They need to provide DHCP and a route to the Internet. That's it. Where is competition going to occur? Wherever an open network that honestly permits competition exists, one would hope to expect the establishment of such a barebones ISP offering simple access at speed only limited by the customers equipment, with no throttling or restriction (since we're splitting that fiber amongst a potentially infinite number of ISP companies, this will work out great), for maybe 1% over the price they have to pay to the muni for access to the network. And then who competes with them and how?

The trend has been going more towards OTT services, because there's more profit to be made.

I'm going to take a wild guess and say you have never run an ISP.

The answer is people. You need people to "provide a route to the Internet". To answer phones. To fix things when they break.

Running an ISP is far from glamorous and far from just leaning back and skimming the profit.

Reply on news.ycombinator.com