I am always baffled by these things. Say there’s a huge company with a monopoly in your area. My first thought is “How did they get that monopoly? What happened to all the other people who must surely have had the idea to compete with them?” But no, these stories are always treating “Hey, let’s start a competing company!” like some revolutionary idea that nobody has thought of before, and that success is assured.
Telephone and cable TV companies were explicitly envisioned as regulated monopolies in most places. Then it was cheaper for them to provide Internet over their existing lines than for a new company to come in.
I didn't think I've ever seen mention of a buyout in these articles. That could be something. Franchised ISP. Maybe Comcast is incapable of servicing an area effectively, so they could say something like "we'll give you x gbps of guaranteed throughout at the datacenter (or however it works) to our main line and teach you how to setup, you cover installation and maintenance". Just because it seems like it would've been easier for these guys to do only the installation and routine maintenance. But idk I guess they don't want to because they make their money anyway
> What happened to all the other...
There's a huge gap between "had the idea" and "had all the technical skills, the $millions in capital, and the managerial ability to actually build it". Then there's the barrier of "and succeed". If you read between the article's lines a bit - these guys had loads of the first 3, yet they're still losing loads of money every month.
But, bigger picture, you have a good point. These articles are obviously cherry-picked stories, with an extremely optimistic "... and the little guy wins!" spin. Ars is writing for an audience of techies who are frustrated with crappy ISP's.
The capital for an ISP is surprisingly low. The main problem is getting a physical connection to your customer's house. And that's such an obvious legal minefield that no networking nerd wants to do it.
>The capital for an ISP is surprisingly low.
Crazy generalization.
>And that's such an obvious legal minefield that no networking nerd wants to do it.
Honestly half the fun.
The capital can be low. Vs. the article notes that these two guys have 75 miles of fiber installed (to 1,500 potential-customer homes) and 15 local employees. Vs. currently monthly revenues of about $10K.
Yeah, obviously these guy's long prior experience - pulling fiber for other ISP's - was another critical cornerstone of their ability to go from idea to build-out.
In the specific case of ISPs I think it's always because you won't make enough money to justify it as a big company, yet the task is too big and complicated to do it as an individual nerd.
The worst part appears to be the physical wiring. If your government has implemented loop unbundling, you're already set (probably need to do some bureaucracy and pay some affordable-at-a-stretch fees to get access to it). Otherwise, or if the loops are just crap, you have to figure out how to physically get a cable to everywhere, a task that is fundamentally laborious and legally fraught, not nerdy at all (unless lawyers are nerds) so nobody wants to do it.
Wireless ISPs are about as popular because of this. Wireless service is always worse, but you only have to install plant (physical infrastructure) at the customer's house and one central location, not all the places leading up to the customer's house. This makes it a whole lot more amenable to individual-nerd or handful-of-nerds setup.
I encourage everyone to at least think about how they would do it.
>Wireless service is always worse
Look, wireless service is almost guaranteed to be worse, but that has more to do with dodgy operators. The technology is fantastic, and when engineered correctly largely undetectable.
That said, in my time, I can count on one hand the number of installations where I was allowed to engineer the service correctly. And I can count on all the hands in a small city the number of times I have been called to rescue something extremely stupid, like shooting a link across a construction site.
> ...you only have to install plant (physical infrastructure) at the customer's house and one central location, not all the places leading up to the customer's house.
In a rural environment, yeah, sure. Based on what I'm seeing in San Francisco, in an urban environment, you're going to be negotiating for roof space for many transceivers on many separate roofs. (I do absolutely agree that even that annoying tasks is certainly way less work than dealing with a local or state government that wants it to be impossible to run fiber through or along streets and sidewalks.)