Hacker News Reader

Comment by zw123456 | Hacker News Reader

I have tried posting this idea a couple of times before and have been roundly scolded, but oh well, here goes again. I think... that readers of this forum are ideally equipped to find a way around this problem. What I mean to say is that it is universally felt by everyone that they hate their broadband provider, phone company etc. Why can't we, the tech community come up with a competing technology? I know many have tried to do "open source" types of mesh wifi networks, but I think something more robust than that is needed. Everyone agrees current companies are leaving their customers unsatisfied, hence there must be an opportunity there for someone clever enough to come up with a good alternative.

View on HN

It's not a technological problem. The technological barriers have been taken care of. Google Fiber exists and is awesome and affordable. The reasons we don't all have Google Fiber level of service are political: regulatory capture, monopolies and oligopolies, conflicts of interest, etc.

The problem with fiber I think is that a lot of communities are "trench weary" and don't want more fiber providers digging up the streets. I think Google gets around that by hyping that they are coming to town, which is smart, it gets the politicians to back them.

It's only not a tech problem if every service has Google levels of server capacity and performance. Websites still regularly fall over when posted on aggregation sites.

They fall over because they're misconfigured (namely, default Apache config on a $5 VPS; works fine at low traffic, but throw some more at it and Apache spawns more processes than you have RAM, starving the DB and OS filesystem cache, then swapping to death), not because websites are bandwidth-limited.

I'm not just talking about simple static websites. Please, let's not pretend that upgrading everyone to fiber will be sufficient. You know better than that.

I don't know what systemic tech problem you think exists for web hosts. There's no lack of bandwidth on their side of the internet -- heck, if you excluded Netflix and YouTube, over 60% of last mile capacity in the country would be completely unused during peak hours. Those few sites, like video streaming services, that are bandwidth-limited also tie their revenue to consumption -- more bandwidth on the consumer end just widens their potential customer base.

There are many infrastructure changes that are needed in order to increase the performance of the Internet besides faster home connections. CDNs / caching / DNS performance / SPDY / distributed computation and on and on and on. The OP makes it sound like the only thing between us and 0 ms response times on every service and every web page is some politics.

In other words the reason we don't have Google Fiber levels of performance is because 1) we don't have Google Fiber and 2) all the infrastructure to maximize the potential of that last mile connection isn't there. 2) is a problem!

I guess the problem is that all of these (broadband provider, phone company) are regulated by the government. Once you have the government involved, forget about competition, the market won't fix. You have to fix politics.

Ah, but what is within reach of government regulation? Are they willing to, say, shoot down satellites?

Too bad there is not a tech way to work around them. Maybe using some kind of unlicensed wireless?

There is, actually. They're called Wireless ISPs or WISPs. http://www.wispa.org/find-a-wisp

I wonder if an automatic peering arrangement could be built on top of a mesh infrastructure, to give incentive for running more robust nodes (not just for in-network traffic, but for uplink to the internet at large).

Interesting. I was thinking about some of the open source implementations of LTE that have been posted here but maybe the wifi mesh approach needs tweaking. Also, from a business perspective, I think it would be cool if there was a way to make it sort of like a "co-op" so people who host hot spots or base stations would get some profit sharing or something?

FON (https://corp.fon.com/en) tried to make this work, I'm sure there were other attempts.

It's not a mesh network, so it doesn't divorce itself from the last mile carriers.

Very interesting, that is exactly what I was thinking about. I guess something like that would some how have to reach a critical mass.

Reply on news.ycombinator.com