Hacker News Reader

Comment by collinf | Hacker News Reader

Is Google Fiber still expanding? Last I heard it was pretty much cut.

I don’t think Google Fiber is expanding, but it did spur the rest of the ISPs.

AT&T seems to be expanding its gigabit footprint rapidly (https://m.att.com/shopmobile/internet/gigapower/coverage-map...) and not just to most affluent sections of the metro area where I live. Comcast is trying but they are still overpriced compared to AT&T and the upload speeds are abysmal- 35Mbps

FWIW, Google Fiber is actively being rolled out here in Louisville.

Yes, and I recently had an AT&T fiber line installed in my house because of it. I'm 100% certain that if Google Fiber hadn't started deployments in Louisville I would not have a fiber connection today. Even if it's not from Google.

Thank you Google!

Google Fiber is still being built in Huntsville, AL. It straight up made Comcast, WOW, and AT&T all start rolling out gigabit. The only other local provider, Mediacom, had already begun rolling out gigabit before Google even announced its plans for the area.

There is an assumption among folks in the tech community that there is no competition between providers serving different markets--i.e. what Verizon does in Maryland cannot affect what Comcast does in Delaware. However, the idea that Google Fiber is "spur[ring] the rest of the ISPs" to roll out gigabit is accepted without question, even though Google Fiber serves very few markets. Verizon, for example, upgraded FiOS to gigabit even though Google Fiber serves no Verizon territory and has shown zero willingness to overbuild another fiber provider, and after Google Fiber announced it was downshifting deployment.

One of those premises has to be wrong.

It's hard for big telcos to move fast (it was hard for Google to move fast in this area). What telcos are doing today is responding to what happened 5 years ago. Where they're doing it is responding to what happened 6-12 months ago.

Because Google wasn't able to get traction on their construction in San Jose, AT&T had time to do street surveys and order custom multistrand fiber with drops for a splitter at every pole, and then get the fiber actually on the pole before Google got much of anything done. Existing carriers can run new cable on poles they already have access without a lot of hassle.

Otoh, Google primarily wanted to show it could be done and spur others to, faster eyeballs without capital costs is a win. If you can't get 10mbps for YouTube on a gigabit connection, something is clearly broken.

Edit: pole instead of poll (thanks unmonk)

Did you mean to write "pole" instead of "poll"? (Genuine question - I'm not sure whether "poll" is a term with some meaning in the telecom / isp industry.)

It did spur competition where Google Fiber was announced. The ISPs were tripping over themselves to match Google in those cities.

But as soon as Google announced the demise of Fiber, they mostly put the brakes on those expansions. They may still expand their own fiber initiatives now, but it's nowhere near as fast, and you won't get as cheap 1Gbps fiber as you would've gotten with Google Fiber around.

AT&T's gigabyte service is $70 a month.

Nationally?

From what I can tell, the price for new customers is $80 a month for the first year and $90 a month thereafter.

I can get 100mbs (both down _and_ up) from at&t for $50/mo. That's plenty for my home. Much better than comcast

I have gigabit internet and my house is wired for gig-e. But I agree, it is overkill. Even 4K streaming content is only 15Mbps and I’ve never seen anything in real life come to close to even 100Mbps except for the VPN to my job in the middle of the night, occasionally upload speeds to AWS, and when I’m uhh, downloading Linux ISO’s from bit torrents.

The only advantage I see of gigabit Internet is that it’s uncapped and I could easily host some side web projects from home.

> There is an assumption among folks in the tech community that there is no competition between providers serving different markets--i.e. what Verizon does in Maryland cannot affect what Comcast does in Delaware.

Grass-is-greener effect is real. While there may be no competition to spur changes, other things might such as PR and loyalty when facing potential future competition (i.e. in a more competitive future or just by people moving). If the marketing efforts are any measure, customer retention during a move is a big target.

We should realize that they compete for mindshare locally and regionally and there are overlaps, even if they don't affect decisions as significantly as direct competition would.

Tell that to people in smaller towns....

There are hundreds of small regional and muni broadband companies popping up deploying fiber networks. In my area I know of a couple of smaller ISP that only service my state

This on of the reasons Verizon and ATT are moving to improve their services, not Google Fiber

Google Fiber proved there was a demand for faster fiber based wireline services in an era where companies like ATT and Verizon were saying wireline service was dead and the future was 100% wireless...

But like the false narrative of the PC being dead and all computing will be 100% on 5in mobile phone, the idea that wireless will ever replace wired services is pipe dream in the mind of Verizon's CEO

Verizon or any other ILEC rolling out last mile fiber also removes them from antitrust restrictions on traditional copper, so viewing such actions as 'competitive' in nature should not be assumed.

Comcast may have little incentive in a remote city to make these changes, but if Comcast even believes that Google Fiber may be looking to expand to one of its areas, it is in Comcast's best interest to preemptively improve service in that area.

Preemptive retention is much less costly than having to reacquire customers after they've switched to Google Fiber.

>the upload speeds are abysmal- 35Mbps

It's funny how regional this is, even only looking at major cities. In my part of Los Angeles, you can't get a residential line with more than 20Mbps up for any money.

Any money? For a couple millions dollars I'll personally drag a cable to your location.

It's actually true. I know of people who did have the money and asked for a quote to lay about 2km of line, no matter how high. Quote refused.

I mean, it is likely just access to deploy the new wires. LA is far larger and likely has much more cost/hoops to jump through for companies to deploy anything. Ex: "No deploying new fiber unless you do it to the entire area simultaneously, also you must use this labor to do it, coordinate with this many more entities while working in the areas" etc compared to sparser and generally business-friendlier Southern/Midwest Cities

5G can't come soon enough.

Google fiber is dead internally. My friend works there said the org has been reduced and employees repurposed to other orgs.

That is indeed why I left Google, but that was almost two years ago at this point. Last I heard they changed their mind and are starting things back up with brand new people.

Being an ISP is just too appealing a business model. It's 95% profit once you've paid off the loans for the infrastructure. And when you're Google, you don't need loans.

> It's 95% profit once you've paid off the loans for the infrastructure.

This is FUD, and it doesn't even pass the smell test.[1] Maintaining broadband infrastructure is incredibly labor intensive (and the required labor is high-skill and often unionized). To use Chattanooga's EBP as an uncontroversial example: https://static.epb.com/annual-reports/2017/media/EPB_2017_Fi.... In FY 2017, it's fiber optic division had revenues of $150 million and expenses of $122 million (not including interest on the underlying fiber optic system, which it looks like is accounted for on the electricity side of the business). That's an operating profit margin of about 20%,[1] which is actually pretty good. Verizon's wireline division in most years has an operating profit margin of under 5%. (EBP, being in a right-to-work southern state, almost certainly has lower labor expenses than Verizon, with its primarily north-east footprint.)

[1] And that's before taxes. A telco can't pull a double Irish with a Dutch sandwich because all its revenues and expenses are domestic. Verizon's effective income tax rate prior to the Trump cuts was 34%: https://247wallst.com/telecom-wireless/2018/01/23/verizon-ea....

you're right. they're not significantly expanding. i don't think it was really paying off in any useful way.

The rule change won't necessarily spur more Google Fiber deployment, since the ISP has other financial problems and has largely stopped expansion of fiber into new cities.

I wonder if Google could pivot to offering legal services for establishing municipal broadband. There has to be some way they can recoup the costs by acting as the legal 'spear' for cities that are unaware of how to offer greater competition and allow smaller ISPs in (if not acting as their own).

Google grew up in a services era by replacing a hugely human intensive service with an algorithm.

Navigating local laws might be doable with their current expertise, but would definitely be human intensive, absent a real breakthrough in AI.

OTOH, there is that automated portal to fight traffic tickets. I wonder if Google tried to automate and A/B test more complicated court or municipal filings how it would do in the long run.

What portal to fight traffic tickets are you referring to? Never heard of this but sounds interesting

https://venturebeat.com/2016/06/27/donotpay-traffic-lawyer-b...

In Austin, at least, though we might be among the last. In February of 2017, they microtrenched our streets and installed fiber. On Tuesday of this week, they put up No Parking signs on the street again, ostensibly to do whatever else they need to actually bring us live.

Reply on news.ycombinator.com