4 ms·

> 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 pe

> 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 going to get better. > I would rather see the expansion of traditional cable or even satellite to rural areas 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 ...

ericmay 1y ago
> 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 while I pay $40 for 1 gig.... With the snide remarks aside, why expand copper or fiber into rural areas when we can just let SpaceX and others launch satellites and provide a potentially better service? I'm sympathetic to a goal of "have really, really fast Internet service" but maybe there is a better regulatory framework for increasing competition both urban/suburban and rural areas.
- baby_souffle 1y ago
  > 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 vast distances since the speed of light is faster in a vacuum compared to glass but for 99.999% of residential ISP needs, fiber-to-the-home is going to offer a more robust pipe that fits more and with less latency. > but maybe there is a better regulatory framework for increasing competition both urban/suburban and rural areas. Almost certainly. Regardless, any better solution necessarily exists only in a world where 100/20 isn't "cutting edge" 30 years after it became technically possible.
  - ericmay 1y ago
    > 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.
    - baby_souffle 1y ago
      > 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...
userbinator 1y ago
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"?
- vjvjvjvjghv 1y ago
  Work from home? Copy large files?
  - userbinator 1y ago
    The majority of WFH setups are either "remote into a server and work there" (RDP uses a few Mbps at most when streaming HD video, and SSH is basically negligible) or "everything you need is already on your local company-owned machine."
- Larrikin 1y ago
  2 younger kids streaming a movie, one parent listening to a podcast in the shower, one parent streaming a YouTube cooking video while making dinner, and an older kid playing any game would completely saturate the network. That is assuming nobody has a phone that is also connected to the network Also its ridiculous to think that is excessive in any way. Imagine what we could have if we had 100 gigabit or 1 terabit. Instead of watching a flat 4k movie, render a full 4k scene in AR.
  - 0x457 1y ago
    2x1080p streams is 10mbit/s 2x4k is 30mbit/s. so below 100mbit/s. Video games don't consume a lot of traffic[] fortnite literally measured in 100s kbps. Video games consume traffic most only when they're updating, then sure you can saturate 100mbit/s, but you also can saturate 10G just as well, so in this case it's about speed, kid can do some chores while it's updating lmao. I initially wanted to skip podcast because...320kbps max? But then thought that not only that, but it's also most likely already downloaded to the device. []: destiny 1/2 consume a lot, but that would mean your older child is over 30 and living with you. That's if you use any streaming service, if you're streaming legally ripped Blu-rays, then yea, 100/20 isn't enough, but those are usually within LAN. And if you're talking seeding/streaming to others, then any asymmetric connection speed will suck.
giantg2 1y ago
"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 state and have higher urbanization. If you want better coverage for an area this size, then it makes sense to include satellite coverage. The gigabit goal excludes them. "Why spend money and time to expand copper into rural areas when fiber is the same cost." They're only the same cost if you're starting from the same location. Copper has better penentration already. Expanding copper might mean adding a couple miles. Expanding to that same location with copper might mean putting in 10-50 miles plus any sort of hub or substation. So yes, equal distance is roughly equal cost, but almost everywhere fiber is put in, it's alongside copper anad thus not increasing coverage nor decreasing costs.
- baby_souffle 1y ago
  > 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.

ericmay 1y ago

> 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 while I pay $40 for 1 gig.... With the snide remarks aside, why expand copper or fiber into rural areas when we can just let SpaceX and others launch satellites and provide a potentially better service? I'm sympathetic to a goal of "have really, really fast Internet service" but maybe there is a better regulatory framework for increasing competition both urban/suburban and rural areas.

baby_souffle 1y ago
> 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 vast distances since the speed of light is faster in a vacuum compared to glass but for 99.999% of residential ISP needs, fiber-to-the-home is going to offer a more robust pipe that fits more and with less latency. > but maybe there is a better regulatory framework for increasing competition both urban/suburban and rural areas. Almost certainly. Regardless, any better solution necessarily exists only in a world where 100/20 isn't "cutting edge" 30 years after it became technically possible.
- ericmay 1y ago
  > 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.
  - baby_souffle 1y ago
    > 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...

userbinator 1y ago

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"?

vjvjvjvjghv 1y ago
Work from home? Copy large files?
- userbinator 1y ago
  The majority of WFH setups are either "remote into a server and work there" (RDP uses a few Mbps at most when streaming HD video, and SSH is basically negligible) or "everything you need is already on your local company-owned machine."
Larrikin 1y ago
2 younger kids streaming a movie, one parent listening to a podcast in the shower, one parent streaming a YouTube cooking video while making dinner, and an older kid playing any game would completely saturate the network. That is assuming nobody has a phone that is also connected to the network Also its ridiculous to think that is excessive in any way. Imagine what we could have if we had 100 gigabit or 1 terabit. Instead of watching a flat 4k movie, render a full 4k scene in AR.
- 0x457 1y ago
  2x1080p streams is 10mbit/s 2x4k is 30mbit/s. so below 100mbit/s. Video games don't consume a lot of traffic[] fortnite literally measured in 100s kbps. Video games consume traffic most only when they're updating, then sure you can saturate 100mbit/s, but you also can saturate 10G just as well, so in this case it's about speed, kid can do some chores while it's updating lmao. I initially wanted to skip podcast because...320kbps max? But then thought that not only that, but it's also most likely already downloaded to the device. []: destiny 1/2 consume a lot, but that would mean your older child is over 30 and living with you. That's if you use any streaming service, if you're streaming legally ripped Blu-rays, then yea, 100/20 isn't enough, but those are usually within LAN. And if you're talking seeding/streaming to others, then any asymmetric connection speed will suck.

giantg2 1y ago

"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 state and have higher urbanization. If you want better coverage for an area this size, then it makes sense to include satellite coverage. The gigabit goal excludes them. "Why spend money and time to expand copper into rural areas when fiber is the same cost." They're only the same cost if you're starting from the same location. Copper has better penentration already. Expanding copper might mean adding a couple miles. Expanding to that same location with copper might mean putting in 10-50 miles plus any sort of hub or substation. So yes, equal distance is roughly equal cost, but almost everywhere fiber is put in, it's alongside copper anad thus not increasing coverage nor decreasing costs.

baby_souffle 1y ago
> 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.