Hacker News Reader

Comment by jeswin | Hacker News Reader

> We're taking an AWS-like platform approach to our tech stack, our goal is to deliver deep APIs, network integration points, edge-caching and much more.

The website wasn't convincing, but this doubly so. What is your technology (which scales) that enables 10x savings, given that ISPs are operating on razor thin margins already?

Also, 20k*100 = $2million. If the technology is proven, wouldn't you be able to just raise it without a lot of trouble? I want you guys to succeed. It's just that you might need to provide better arguments here.

View on HN

90% of the cost of being an ISP comes from laying the core underground fiber network that serves as the backbone of the system. This usually costs anywhere from $30k - $50k per kilometre to lay because it involves digging trenches, getting permits, etc. It takes about $50M to deploy a basic network in a city like Bangalore for example - we know from our existing ISP partners that this is a reasonable figure.

The fact that we are using FSOC to replace the underground fiber is what creates the 10x in savings. It only takes a few hundred FSOC units to do the job of the fiber as the core network and is much, much cheaper.

I've added a reply further down the page that addresses why we'd like the public to be involved, I'd be happy to elaborate more if you've got more questions.

>The fact that we are using FSOC to replace the underground fiber is what creates the 10x in savings

Then why aren't other ISPs doing it?

The tech is extremely new and other telecom companies and ISPs are testing it to plug holes in their network, this gives them only limited ROI on the tech because they've already got fiber networks and their entire organization is geared towards the deployment of fiber.

Also, it's worth noting that most ISPs/telcos are manpower and regulations based resource extraction organisations. Their moat is being able to handle local bureaucracy and pump in the heavy capital required to deploy the network. They like to buy and use proven tech that comes with SLAs and support from large tech vendors. They will wait for FSOC to be mature before they do anything meaningful.

We get the largest ROI from FSOC tech simply because we're building the network from scratch and are targeting the low cost market that is currently underserved. Most other ISPs are aiming for the premium market.

Oh come on!

Free space optics isn’t some new technology. The real reason FSO isn’t used is because despite almost half a billion USD in R&D over the past 20 years the reliability issues haven’t been solved.

FSO has taken a leap in the last 3 years. We've got 9 months of data with thousands of live customers being served that says otherwise.

Without divulging any trade secrets, are there any research papers or topics you would recommend learning more about? I'm really interested in learning more about these FSO improvements.

[Citation needed]

Way back in 2009 there was a company in Chennai that tried something like this https://www.equitybulls.com/admin/news2006/news_det.asp?id=5....

And AFAIK Hathway (another ISP in India) too tried something like this but didn't succeed and they went back to traditional fibre.

I don't know anything about the tech but it seems like it is not so easy and I do hope Wifi Dabba succeeds.

Reply on news.ycombinator.com