Live data from Hacker News

FCC to eliminate gigabit speed goal and scrap analysis of broadband prices

arstechnica.com

201–210 of 221 posts

Re: FCC to eliminate gigabit speed goal and scrap analysis of broadband prices

#201

Maybe this will force sites to stop wasting bandwidth and stem the bloat. You should not need a gigabit connection (which I do not even have in all of my home LAN) to browse the Internet. Edit: downvoters, please explain why I need 125MB/s (that's 3 full installations of Windows 95 every second) for normal browsing.

>Maybe this will force sites to stop wasting bandwidth and stem the bloat. You should not need a gigabit connection (which I do not even have in all of my home LAN) to browse the Internet.

Nope. You don't. But why should I have to pay Google/AWS/Azure/Cloudflare/Digital Ocean/etc./etc./etc. to host my small business/vanity/family chat/photo sharing site, or YouTube to host my barbecue sauce unboxing videos, when symmetrical (okay, 300 up/down would likely be plenty for most of that, but once there's fiber, Gb/sec seems good to me)?

I have no interest in propping up huge companies with hosting fees because other huge companies have no competition.

All that said, I get your point. But my use case is not your use case. Do you get mine?

Edit: Clarified prose.

Re: FCC to eliminate gigabit speed goal and scrap analysis of broadband prices

#202
post #79

Earlier quoted context omitted.

You don't need gigabit but resource constraints have never reined in bloat in the past and they won't now.

Apparently you've never heard of the demoscene.

Yeah, constraints are good for art. I was thinking of the commercial world.

Re: FCC to eliminate gigabit speed goal and scrap analysis of broadband prices

#203
post #34

Earlier quoted context omitted.

> Why spend money and time to expand copper into rural areas when fiber is the same cost. it's the people/permits/labor that are $$$$. It makes no meaningful difference weather your expensive hbm crew is pulling fiber or copper and we know that copper doesn't go as fast ... Well they voted for it, so I'll stick to my fiber in my big city and they can fend for themselves and pay $90/month for 10 up 1 down or whatever…

> Well they voted for it Not all of us. > why expand copper or fiber into rural areas when we can just let SpaceX and others launch satellites and provide a potentially better service? Fiber is objectively the right choice for future proofing. Bouncing a radio wave off of cube 300 miles above will _always_ be sub-par compared to a direct fiber connection because the latency is higher. SL May have a slight edge going…

> Not all of us.

I didn't vote for all these highways and suburbs either but we live in a democracy. By and large rural voters* voted to cut programs like this, though, and I think it's very fair to point that out.

> Fiber is objectively the right choice for future proofing. Bouncing a radio wave off of cube 300 miles above will _always_ be sub-par compared to a direct fiber connection because the latency is higher.

Yes and no, it depends on the cost. It might not be ideal but I'm also not sure my tax dollars should go to make sure people who live in rural areas have faster Playstation downloads. Do we actually need to physically pay for and build this infrastructure? I'm not necessarily opposed to it, but where is the economic analysis? If we invest $XYZ on this infrastructure what's the expected ROI versus alternatives?

* Just to be clear I don't have any particular problem with "rural voters" and I don't really like these arbitrary groupings of people into Urban, Suburban, Rural but it is what it is.

Re: FCC to eliminate gigabit speed goal and scrap analysis of broadband prices

#204

Earlier quoted context omitted.

> I think this isn't as bad as people make of out to be. The 100/20 goal is perfectly fine for the vast majority of users 640K was "perfectly fine" for most people, too. 100/20 is barely enough for a household of 3-5 "light" users. The US already has abysmal broadband speed/bandwidth/latency metrics compared to the rest of the developed world and settling for 2010's version of "fast" in 2025 is ... not how we're goin…

"100/20 is barely enough for a household of 3-5 "light" users." It's plenty for my household with a similar number of users. We have people working from home, streaming, downloading documents or games, etc with no problem. "The US already has abysmal broadband speed/bandwidth/latency metrics compared to the rest of the developed world" Because the countries in the rest of the developed world are about the size of 1 s…

> It's plenty for my household with a similar number of users. We have people working from home, streaming, downloading documents or games, etc with no problem.

I recognize that "no problem" and "as fast as possible" are not the same. 100/20 works for you but once you've seen 1000/20, you really do notice things taking ~10x longer than they need to. It all adds up.

> Copper has better penentration already.

Excluding 56k, twisted-pair copper has absolutely crap throughput unless you're ~500m or less from the 'head-end' which is where your DSL is being turned back into laser pulses. Coax does have better performance/distance compared to twisted pair but now you have a non-trivial network of amps/taps/power-injectors to maintain. DOCSIS really does not like it if there's any issue w/ the coax so you're going to need a small army just to keep the hard-lines in good shape.

I have never seen a PON network get it's throughput cut in half just because somebody didn't screw the cable _all the way_ in to the tap. I see ingress degrading DOCSIS networks all the time though.

At least as far as rural northern CA goes, ATT runs fiber out to plants and then twister pair from there to the customers which are - at most - a few km away. Fiber is already 85% of the way to the customer... let's just finish the job instead of giving everybody a starlink account.

Re: FCC to eliminate gigabit speed goal and scrap analysis of broadband prices

#205

Earlier quoted context omitted.

> Well they voted for it Not all of us. > why expand copper or fiber into rural areas when we can just let SpaceX and others launch satellites and provide a potentially better service? Fiber is objectively the right choice for future proofing. Bouncing a radio wave off of cube 300 miles above will _always_ be sub-par compared to a direct fiber connection because the latency is higher. SL May have a slight edge going…

> Not all of us. I didn't vote for all these highways and suburbs either but we live in a democracy. By and large rural voters* voted to cut programs like this, though, and I think it's very fair to point that out. > Fiber is objectively the right choice for future proofing. Bouncing a radio wave off of cube 300 miles above will _always_ be sub-par compared to a direct fiber connection because the latency is higher.…

> Do we actually need to physically pay for and build this infrastructure? I'm not necessarily opposed to it, but where is the economic analysis? If we invest $XYZ on this infrastructure what's the expected ROI versus alternatives?

I wish I had a clear answer and good data on this. My gut tells me that giving more people faster and more reliable access to information is almost certainly going to be worth it in the long run. A few hypotheticals that come to mind:

Some highly paid engineer can live out in the boonies but still work for their high wages. That excess cash is probably going to make it's way into the local economy.

Likewise, rural schools are not know for their performance. That almost certainly could change if better content/curricula was easier to access and distribute. There is a non-zero chance that some breakthrough cure for cancer comes from a smart kid that grew up in a rural area. With crap internet access, that kid almost certainly would not get the education to match their potential.

(or maybe it isn't a cure for cancer, it's the next John Carmack...)

Fiber isn't the only way to level-up access... but it is the easiest to maintain. Other than the very ends of the connection, it's all passive.

No need to to inject power for amplifiers. No need to keep building/launching satellites and corresponding ground-stations.

Once it's installed, it's more or less maintenance free until a wild backhoe shows up...

Re: FCC to eliminate gigabit speed goal and scrap analysis of broadband prices

#206

Earlier quoted context omitted.

Assuming you’re referring to Chevron/Loper, I fail to see the relevance to this case. Also, it’s important to remember that Chevron wasn’t “however they want” or to “reinterpret their own authorizations”. It was a doctrine that if the agency (staffed by domain experts responsible for resolving the ambiguity) had a reasonable interpretation of an ambiguity in the law, even if the court thought it had a better opinion,…

> It was a doctrine that if the agency (staffed by domain experts responsible for resolving the ambiguity) This is not an accurate description of what agencies are meant to be experts in. Their expertise is meant to be in how best to act within their bounds. Which is distinct from deciding what those bounds are.

First, are the courts actually any better? Constitutionally they have limited jurisdiction in the first place to adjudicate administrative matters, only getting some authority in 1948 from Congress to do such administration.

> In 1974 the Supreme Court stated that deference depends on an administrative interpretation being consistent with the agency's other statements and being consistent with the congressional purpose

So it wasn't that the agency got to set its own bounds - it had to be consistent with all their other statements and with the congressional purpose of the body. So for example, Congress said the EPA had the power to regulate pollution sources but did not define what meant sources or specifically what regulations they could enact. Chevron said "Congress intended the EPA to figure this out so they get to do that". Loper said "actually, Congress needs to spell everything out" which is actually quite a radical idea for administrative legislation which with a stroke of a pen undoes a huge amount of legislation that had been done with Chevron in mind (& before that it was understood to be the case anyway since judges didn't even have power over administrative powers).

Re: FCC to eliminate gigabit speed goal and scrap analysis of broadband prices

#207

Earlier quoted context omitted.

> I think this isn't as bad as people make of out to be. The 100/20 goal is perfectly fine for the vast majority of users 640K was "perfectly fine" for most people, too. 100/20 is barely enough for a household of 3-5 "light" users. The US already has abysmal broadband speed/bandwidth/latency metrics compared to the rest of the developed world and settling for 2010's version of "fast" in 2025 is ... not how we're goin…

100/20 is barely enough for a household of 3-5 "light" users. What the hell are you doing that 100/20 is "barely enough"?

> What the hell are you doing that 100/20 is "barely enough"?

Many things. A partial list in no particular order:

My day job is all SSH so I don't need a wide pipe so much as I need a short pipe. If the pipe is also quite wide, I can reliably stream content while I work. This isn't just music, this is watching the daycare camera feeds.

Every once in a while I need to docker pull a multi-gig image. It's nice when that doesn't take more than 30 seconds. I don't worry about background tasks moving/syncing/backing-up data; there's plenty of headroom!

Some of my hobbies involve _lots_ of data and the less time I spend waiting for that to download is more time I get for actually doing hobbies.

I also have a few scripts that backup other consequential data stores; the biggest one is google takeout for a few accounts. This results in ~ 800 gigs of data from a few different accounts.

The work product from these hobbies and backups is something that I like to back up off site and a 20mbps connection is trivial to saturate for hours as a time.

Every once in a while I'll make some of my excess compute/GPU available to friends so they can render things. The assets involved are not small and a big pipe means that everybody gets the desired result _faster_.

If you give me a faster/wider pipe, I _will_ find a new use for the extra room. I may not be representative of the typical user but I know for sure that I would not have developed my skills to the extent that I have if I had crap internet access.

Re: FCC to eliminate gigabit speed goal and scrap analysis of broadband prices

#208
post #135
post #14

I think this isn't as bad as people make of out to be. The 100/20 goal is perfectly fine for the vast majority of users. The only thing really supporting the old goal of gigabit connections was fiber. I would rather see the expansion of traditional cable or even satellite to rural areas. Fiber plans tend to be expensive and mostly available in the areas that already have usable high speed options. If we really want t…

Coax can come close of the previous FCC gigabit goal. The FCC wanted 1 Gb/s down/500 Mb/s up. Xfinity is currently offering 2 Gb/s down over coax to most of their residential customers. They can only do 200-300 Mb/s up though.

> They can only do 200-300 Mb/s up though.

This is changing with newer DOCSIS standards! It's still up to the network provider to choose how many channels they allocate for up vs down. Even in perfect lab conditions, DOCSIS 4 is still about 3:1 ratio down/up (about 5/1.5 gigabit). Fiber is the only medium that makes a 1:1 down/up possible at any meaningful speeds.

Re: FCC to eliminate gigabit speed goal and scrap analysis of broadband prices

#209

Earlier quoted context omitted.

How confident are you in the authenticity of posts and comments on these "liberal and conservative sites"?

What a strange accusation. One of the liberal places I read is here, where the discourse usually isn’t that far from Reddit Politics just with more self-righteousness (sometimes, most of the time it’s okay). Why would you doubt that I read legitimate conservative sites? I read enough to know how brigaded Reddit Conservative is, that’s not real. I spent a lot of time with an actually balanced source of news.

You're outting yourself as an unreliable narrator - HN isn't liberal, it's typical moderate neo-lib.

Youre assuming HN must be liberal because it's somewhat critical of Trump. You're reverse-engineering your position.

HN isn't critical of Trump because they're libs, they're critical of Trump because he's bad and the policy is bad.

I mean, raising the deficit by trillions in exchange for A REDUCTION in public services? How can any conservative American, anywhere, possibly defend that? They didn't vote for that!

Post reply on HN