Monthly Archives: September 2020

They’re Back

Facebook recently announced it will be introducing smart glasses in collaboration with Ray-Ban. This will be the second major attempt at introducing the technology since the failed attempt by Google in 2011 when it introduced Google Glass. For those who might not remember, Google Glass was shunned by the general public and people who wore the glasses in public were quickly deemed to be glassholes. People were generally uncomfortable talking to somebody who could be recording the conversation.

It will be interesting to see if the public is any more forgiving now. Pictured with this blog is Glass 2.0 that is being used in factories, but the first-generation public version was equally obvious as a piece of technology.

In terms of technology, 2011 is far behind us, and since then it’s common for anything done in public to end up being recorded by somebody’s smartphone. But that still doesn’t mean that people like the idea of being secretly recorded, particularly if the new glasses aren’t so obvious as Google Glass.

We still don’t know what the technology will look like, but Facebook will try to brand the new glasses as cool. Consider this video ad that accompanied the announcement of the new glasses – who doesn’t want to wear smart glasses like glasses worn in the past by James Dean, Marilyn Monroe, and Muhammed Ali? Facebook says the new glasses will function by being paired with a smartphone, so perhaps they’ll be a lot less obvious than were the Google Glass.

The glasses are the first step towards virtual presence. Facebook Mark Zuckerberg says his vision is being able to virtually invite friends into your home to play cards virtually. However, this first set of glasses isn’t going to include an integrated display that would be capable of generating or viewing holograms. That means the new glasses will likely include the same sort of features like Google Glass such as being able to record what’s in front of you, using the web to browse for facts, or dipping into the web to call-up information about people you meet. With the advances we’ve made in facial recognition since 2011, that last item is a lot scarier today than it was a decade ago.

I recall the tech industry excitement about Google Glass and other proposed wearables back in 2010. The vision was to seamlessly be able to carry tech with you to create a constant human-computer interface. Google was stunned when the public universally and loudly rejected the idea, because to most people the technology meant an invasion of privacy. Nobody wanted to have a casual conversation with a stranger and then later find it posted on social media.

It’s hard to think that is still not going to be the reaction again today. Of course, as a baby boomer, I am a lot leerier of technology than are the younger generations. It seems that Generation Z is a lot less concerned about privacy and it will be interesting to see if young people take to the new technology. We may have one of the biggest generational rifts ever between the first generation that finally embraces wearables and everybody older.

Google Glass never died and morphed into a pair of glasses to use in factories. It allows workers to pull up schematics in real-time to compare to work-in-progress in front of them. The technology is said to have greatly improved complex tasks like wiring a new jetliner – something we all want to be 100% correct.

I will likely remain leery of the technology. What might eventually bring me around is Zuckerberg’s vision of being able to play poker with distant friends. I’ve been predicting telepresence as the technology that will finally take advantage of gigabit fiber connections. I’m not sure that we need glasses that secretly hide the technology capability to make this work – but I guess this is an early step towards that vision.

Update on the 5G Race

It’s been a while since I checked in to see how the US is doing in the 5G race. I haven’t been following the issue since before the pandemic when the US government was tossing around the idea of buying a controlling interest in Nokia or Ericsson. That idea went nowhere but led to a lot of articles in the business press.

I decided to look anew after seeing recently that the FCC is estimating that it would cost US carriers about $1.8 billion to replace Huawei and ZTE gear in US networks. In June the FCC banned any proceeds from the Universal Service Fund to be used to buy gear from the Chinese manufacturers. The US has been joined by Australia and the UK in banning purchases of the gear. I’m still scratching my head about the requirement to pull out whatever’s been bought in the past. Network engineers tell me it’s not hard to firewall hardware from communicating with the outside, and nobody has yet shown evidence that any of the gear has been transmitting data to the Chinese. It just feels odd to see a trade dispute taken so far as to toss out working electronics.

The real 5G race isn’t about hardware but in the deployment of 5G technology. The cellular carriers are all now bragging about their 5G cellular networks. It’s an interesting marketing claim because from a standards perspective there isn’t yet any cellular traffic that can legitimately be called 5G. From what I can see, the only feature from the 5G specification that has been introduced into networks so far is dynamic spectrum sharing (DSS). This feature allows cellular carriers to simultaneously use the same block of spectrum for both 4G and 5G. This is mostly a preparatory feature that is readying the network for other 5G features – the carriers don’t want to limit future 5G to a small subset of spectrum.

When the carriers brag about 5G, what they are really talking about is the introduction of new blocks of spectrum. They’ve labeled every new block of spectrum as 5G and labeled every phone that can receive the new blocks of spectrum as 5G phones. For now, these phones are more expensive than phones that use the traditional 4G spectrum.

A recent article by Geoffrey Fowler in the Washington Post looked at the difference in 4G and 5G speeds all around the San Francisco Bay area. He drove around with six phones so that he could check 4G and 5G performance on the various carriers. What he found will surprise nobody who has ponied up for the new phones – 5G is mostly not faster than 4G. There are places where the signals for one or the other sets of spectrum were stronger, but in logging lots of miles, he didn’t find any advantage for the more expensive 5G phones. Fowler followed up with executives at the cellular companies who admitted that 5G is not yet faster today than 4G.

This is not surprising. Most of the carriers are currently using new lower frequency bands in the 5G offering. The main characteristic of lower frequency bands is that the signal travels farther but there are fewer bits transmitted with the signal. I would have guessed that since fewer people using the 5G spectrum bands that 5G phones might still be seeing faster speeds, but that doesn’t seem to be the case.

The carriers are getting exactly what they hoped for with the 5G phones – they are moving people off of the crowded 4G spectrum bands that were threatening to collapse under the data loads. But unfortunately, just like happened in the early days of 4G, the network performance from a customer perspective is not living up to the marketing hype.

In terms of the 5G race, the US is far behind the rest of the world in 5G speeds. This is again due to the spectrum being used by US cellular carriers. Many other countries have introduced higher mid-range spectrum that they are labeling as 5G, and that means faster cellular speeds. As an example, the average 5G speed in South Korea is more than twice the 5G speeds being delivered in the US.

However, South Korea also offers a cautionary tale about winning the 5G race. The country has deployed well over half of all of the 5G phones being used in the world. However, the South Korean cellular companies are showing no change in average revenue per user – people are not paying more for the 5G experience. And since the experience isn’t actually 5G – they shouldn’t.

 

FCC – Please Focus on Upload Speeds

I wrote a recent blog that talked about how the FCC is recommending to stick with the 25/3 Mbps definition of broadband for another year. In that blog, I mostly talked about how 25 Mbps download is out of touch when the FCC claims that 85% of homes today can buy 250/25 Mbps broadband.

Today I want to look at the second half of the definition – the upload speed. The FCC is proposing, in 2020 – the year when millions were sent home for work and school – that 3 Mbps upload is a sufficiently high definition of broadband. Sticking with the 3 Mbps definition of broadband makes no sense. I contend that 3 Mbps is massively out of touch with the needs of the average home. To make matters worse, the FCC will allow an ISP that offers 25/3 broadband to bid in and win grant funding in October’s RDOF grant – a network which the ISP then has six years to build. The FCC doesn’t just think that 25/3 is adequate broadband today, they think that is okay broadband size years from now.

The pandemic has made it clear to a lot of households that upload speeds matter. Before the pandemic, customers that cared about the upload speeds tended to be folks that sent huge files such as doctors, architects, engineers, photographers, etc. When they worked from home these folks have known for years that the upload speeds on the average home network are inadequate.

All of a sudden this year, millions of homes found out that they don’t have enough upload broadband speeds. Consider the amount of bandwidth that is needed to work from home. There are two uses of upload broadband that are new to most people – connecting to a school or work server and participating in Zoom or other online meetings.

Many home and work servers require the creation of a virtual private network (VPN). A VPN is a dedicated connection – the home connects and stays connected to a school or work server. It generally requires dedicating at least 1 Mbps of bandwidth, but usually more, to create and maintain a VPN connection. This means that somebody working at home on a VPN is going to tie up 1 – 3 Mbps of bandwidth that can’t be used for anybody else in the home.

Zoom calls also require upload bandwidth. The Zoom website says that a home should have a 2 Mbps connection, both upload and download to sustain a Zoom session between just two people. The amount of download bandwidth increases with each person connected to the call, meaning Zoom recommends the 2 Mbps upload, but a 6 Mbps download for a meeting with three other people.

There are other uses for upload bandwidth in the home as well. For example, a telemedicine call can use even slightly more bandwidth than connecting to work or school servers. Upload bandwidth is needed for gaming in the cloud. Upload bandwidth is also used to back-up data files, pictures, etc. into the cloud.

It doesn’t take complicated math to see why a 3 Mbps connection is inadequate for any household that wants to make more than one upload-heavy connection to the Internet at the same time. 3 Mbps is not enough bandwidth for multiple people in a home trying to connect to work and school servers or to make Zoom-like calls. I’ve heard from numerous people this year telling me they can’t have more than one person at a time using their home broadband connection. Many of these complaints came from households using broadband provided by the big cable companies, and many of these homes thought they had plenty of bandwidth until the pandemic hit.

For the FCC to stick with 3 Mbps upload as the definition of broadband is a slap in the face to every family where more than one person wants to connect to the web at the same time. With that definition, the FCC is blessing any ISP that delivers 3 Mbps upload speeds.

Even if the FCC doesn’t want to upgrade the download component of the definition of broadband, they can’t turn a blind idea to the millions of homes trying to make it through the pandemic. If social scientists are right, there will likely be millions of people who continue to work remotely even after the end of the pandemic. This is not a temporary problem that is somehow going to go away.

It’s hard to think that the minimum acceptable definition of upload speeds should be anything slower than 25 Mbps. Assuming a robust WiFi network, that’s enough bandwidth for 3 – 4 adults and/or students to work from home at the time. So FCC, please reconsider the definition of upload speeds. If you stick with 3 Mbps upload as the definition of broadband it means you don’t support broadband networks that can deliver the speeds that the average households need.

Cord Cutting Continues in Q2 2020

The largest traditional cable providers collectively lost over 1.5 million customers in the second quarter of 2020 – an overall loss of 2.0% of customers. This is the smaller than the loss in the first quarter of 1.7 million net customers. To put the quarter’s loss into perspective, the big cable providers lost 16,700 cable customers per day throughout the quarter.

The numbers below come from Leichtman Research Group which compiles these numbers from reports made to investors, except for Cox which is estimated. The numbers reported are for the largest cable providers, and Leichtman estimates that these companies represent 95% of all cable customers in the country.

Following is a comparison of the second quarter subscriber numbers compared to the end of the first quarter of 2020:

1Q 2020 2Q 2019 Change % Change
Comcast 20,367,000 20,845,000 (478,000) -2.3%
Charter 16,168,000 16,074,000 94,000 0.6%
DirecTV 14,290,000 15,136,000 (846,000) -5.6%
Dish TV 9,017,000 9,057,000 (40,000) -0.4%
Verizon 4,062,000 4,145,000 (83,000) -2.0%
Cox 3,770,000 3,820,000 (50,000) -1.3%
AT&T U-verse 3,400,000 3,440,000 (40,000) -1.2%
Altice 3,102,900 3,137,500 (34,600) -1.1%
Mediacom 676,000 693,000 (17,000) -2.5%
Frontier 560,000 594,000 (34,000) -5.7%
Atlantic Broadband 311,845 314,645 (2,800) -0.9%
Cable One 290,000 303,000 (13,000) -4.3%
     
Total 76,014,745 77,559,145 (1,544,400) -2.0%
Total Cable 44,685,745 45,187,145 (501,400) -1.1%
Total Satellite 23,307,000 24,193,000 (886,000) -3.7%
Total Telco 8,022,000 8,179,000 (157,000) -1.9%

Some observations about the numbers:

  • The big loser is AT&T, which lost 886,000 traditional video customers between DirecTV and AT&T U-verse. For many quarters AT&T claimed losses were due to the company eliminating low-margin customers. It seems losses are more likely now due to price increases.
  • The big percentage loser is Frontier that lost almost 6% of its cable customers in the quarter. The Frontier numbers have been lowered for both quarters to reflect the sale of its property in the Pacific northwest.
  • While DirecTV continues to bleed customers, Dish Networks has seemed to have stemmed losses.
  • The most interesting story is for Charter that gained customers during the quarter. The company credits the gains to offering a lower-price package and also to a marketing campaign that is giving two months free of broadband. 329,000 customers took that offer in the second quarter and nearly half of those customers elected to add on cable TV and/or cellular service, both of which were for pay, and not free. Charter has been beating the industry as a whole for cable subscribers every quarter since Q3 2018.

The losses of cable companies continue to mount at dizzying levels for the industry. This is the sixth consecutive quarter where the industry lost over one million cable subscribers. The big providers collectively have lost 3.2 million customers this year, from a starting point of 79.3 million customers at the end of 2019.

It’s especially worth noting that these losses happened during a quarter when the biggest ISPs gained over 1.2 million customers for the quarter.

We’re likely going to have to wait to understand exactly what is happening in the cable industry. For example, a recent large survey from TiVO showed that 25% of US homes have downgraded to less expensive cable packages (cord-shaving). That would mean total revenue losses over and above what would be expected by just net customer losses.

Interestingly, homes don’t seem to be fleeing traditional cable for the online equivalents. Leichtman also tracks Hulu Live, Sling TV, and DirecTV Now and those three companies collectively lost 24,000 customers for the quarter.

CBRS Auction Winners

The FCC held a recent auction for the  3.5GHz Citizens Band Radio Spectrum (CBRS). The auction went for 76 rounds and raised over $4.5 billion for the FCC. This auction was unique in that spectrum was licensed at the county-level awarding up to seven licensed 10 MHz channels in each county. Each PAL (Priority Access License) is good for 10 years.

CBRS spectrum can be used in several applications. The spectrum has good field operating parameters and falls in the middle between the two existing blocks of spectrum used for WiFi. This makes the spectrum ideal for rural point-to-multipoint fixed wireless broadband since it can carry a decent amount of bandwidth for a decent distance. The best aspect of this spectrum is that it’s licensed and will largely be free from interference. For the same reasons, this is also a good spectrum for cellular data.

The biggest winner in the auction was Verizon which spent $1.89 billion on the spectrum. The company landed 557 PALs licenses in 57 counties. The company needed this spectrum to fill-in mid-range spectrum for 5G. Verizon has also recently announced a fixed cellular broadband product for rural homes and this spectrum could provide an interference-free way to deliver that product from rural cell sites.

As expected, Dish networks was also a big winner and will be paying $913 million for CBRS spectrum. As the newest nationwide cellular carrier, the company needed this spectrum to fill in the holes in the cellular spectrum it already controls. The other traditional cellular companies were a no-show. AT&T didn’t buy any of the CBRS spectrum. T-Mobile only purchased 8 PALs licenses in six counties.

The largest cable companies scored big in the auction. Charter bought $464 million of spectrum, Comcast is paying $458 million for spectrum, and Cox purchased $212 million of spectrum. As the newest entrants in the cellular business, Comcast and Charter have been buying wholesale cellular broadband from Verizon – this spectrum will let them shift to their own cell sites for a lot of cellular traffic. There is also speculation that cable companies might be planning on using the new spectrum to launch a fixed-wireless product in the rural areas surrounding their cable properties. Both Charter and Cox have entered the upcoming RDOF auction that is awarding $16.4 billion for rural broadband and the companies might be planning on using this spectrum to cover any areas they can win in that reverse auction.

One of the smaller cable companies, Midcontinent Communications, spent over $8.8 million for PALs licenses. Midco already won sizable rural grants to deploy 100 Mbps broadband in Minnesota and the Dakotas. This spectrum will help the company meet those grant pledges and perhaps allow it to pursue RDOF grants.

There were a few other large bidders. One was Nextlink which provides fixed wireless broadband today in Texas, Oklahoma, Kansas, Nebraska, Iowa, and Illinois. Windstream purchased over 1,000 PALs and the traditional telco is likely going to replace aging rural copper with wireless service, while also possibly be expanding into new service territories with fixed wireless. SAL Spectrum LLC won 1,569 PALs. This company owns numerous other blocks of spectrum and it’s not clear who the user of this new spectrum might be.

The biggest news is that the auction allowed smaller bidders to win licensed spectrum. There were 228 different winners in the auction, most of which are small WISPs, telcos, and electric cooperatives. These entities benefited by the FCC’s willingness to auction the spectrum at the county level. Most previous wireless spectrum was allocated using much larger footprints, which kept small bidders from acquiring spectrum.

Penalizing Bad FCC Broadband Reporting

It’s universally understood throughout the industry that the broadband data reported by ISPs to the FCC is full of big problems. Some of the problem in the database can be blamed on the FCC, which allows an ISP to claim an entire Census block as having good broadband even if only one customer in the Census block can actually get that faster speed.

However, in looking in detail at counties all over the country, this seems to be a relatively minor part of the overstatement of broadband. For example, the issue crops the in Census blocks near to a town that has cable broadband, and the FCC reporting system usually assumes that some homes past the end of the cable network can get fast broadband. The FCC has proposed to fix this by asking ISPs to draw polygons around customers, and if the ISPs serving towns do that right, this issue would disappear.

The much bigger problem in the FCC database come from ISPs that overstate broadband coverage, broadband speeds, or both.  I’ve seen entire counties where the FCC database claims broadband coverage that doesn’t exist.

Part of this problem is due to a poor interpretation of the FCC rules. A lot of ISPs interpret the FCC rules to  mean they should report the fastest speed they advertise instead of the fastest speed they can deliver. I’ve seen numerous places where the big telcos have claimed 15 Mbps or 25 Mbps download on DSL, when speed tests can’t find anybody in the area getting more than 5 Mbps download. I’ve looked at counties where WISPs claim speeds of 50 Mbps up to 300 Mbps when customers largely have speeds under 10 Mbps.

Even more aggrevating are ISPs that claim broadband coverage that doesn’t exist. For example, I’ve seen WISPs that claim coverage of an entire county when they are only located on one or two towers. But this isn’t done only by fixed wireless providers, and the FCC is finally talking about fining an ISP for faulty reporting.

The FCC is threatening to fine Barrier Communications Corp. from New York that markets under BarrierFree. In 2017 the ISP made big news when they falsely claimed that they were providing fiber broadband to 62 million customers that largely didn’t exist. The FCC went published an annual report to Congress that included the imaginary broadband, which led the agency to crow about the big nationwide improvement in broadband coverage. The FCC got egg on their face when the issue was brought to its attention, and the agency was forced to reissue the annual report to Congress. As part of that process, the FCC warned BarrierFree to cease the overreporting.

Apparently, the ISP is at it again because the FCC is now threatening a fine of $164,000 for BarrierFree for continued overreporting. The FCC says that’s the maximum penalty allowed by law. There were supposedly substantial overreporting in both the September 2019 and March 2020 data. To the best of my knowledge this would be the first fine due against an ISP due to false reporting in the 477 process. The FCC has threatened fines against Verizon and a few other ISPs for falsely reporting rural 4G cellular coverage, but I’m not aware of any fines being levied.

The idea of levying fines against ISPs for blatant broadband overreporting is long overdue. There can be huge consequences when ISPs can freely claim broadband coverage that doesn’t exist. The biggest current consequence of such overreporting is that it can block eligibility for grants. The FCC used the faulty 477 data when determining the areas that are eligible for the $16.4 billion in RDOF grants that will be awarded in October. I know of counties where no RDOF grants are being offered due to the FCC data falsely showing counties to already have adequate broadband. There are many rural counties where at least some portion of the county has been incorrectly excluded from RDOF grant eligibility due to ISP overreporting of broadband speeds and coverage.

I have to believe the FCC when they report this is the biggest penalty allowed by law – but it’s not nearly high enough. How large should a fine be if an ISP keeps tens of millions of grant dollars from coming to a county? That question is even more pointed if the overreporting ISP gains a market advantage by keeping out grant funding. In my mind, if an ISP blatantly overreports broadband and keeps $10 million of grant funding from benefitting a county, then that ISP owes that community $10 million. I’m sure there are ISPs that are glad I’m not an FCC Commissioner.

Loving to Hate Our Big ISPs

The American Customer Satisfaction Survey (ACSI) was released earlier this summer that ranks hundreds of companies that provide services for consumers. Historically cable companies and ISPs have fared poorly in these rankings compared to other businesses in the country. The running joke reported in numerous articles about this survey is that people like the IRS more than they like their cable company (and that is still true this year).

But something interesting happened in this year’s survey and the ranking for cable companies collectively improved by 3% and consumer confidence in ISPs climbed 5%. There is no easy way to understand a national satisfaction survey, but those trends are interesting to contemplate.

Let’s start by looking at the numbers. Consumers still rank cable TV providers as the least liked group of companies in the country across all industries, joined at the bottom by ISPs. The ACSI ranks each company and each industry segment on a scale of 1 to 100. The top-rated industries are breweries (84%), personal care and cleaning products (82), soft drinks (82), and food manufacturing (81).

By contrast, cable providers are ranked the lowest at 64 followed closely by ISPs at 65. Joining these companies at the bottom are local governments (65.5), video-on-demand providers (68), and the federal government (68.1).

The overall ranking for cable providers grew from a 62 in 2019 to a 64 in 2020. I can only speculate why people like cable companies a little more this year. This could be due in part to huge growth in cord-cutters who no longer watch traditional cable TV and who might perhaps no longer rate a product they don’t use. Or perhaps folks have come to appreciate the cable product more during the pandemic when people are going out less, and likely watching TV more.

The cable providers at the bottom of the rankings continue to get low satisfaction ratings, with Suddenlink (56), Frontier (58), and Mediacom (60). Just above these companies are two of the largest cable providers – Charter (60) and Cox (61). But all of these companies had a slightly improved satisfaction ranking over 2019. The highest-ranked cable providers continue to be Verizon FiOS (70) and AT&T U-verse (70), now relabeled as AT&T TV.

ISPs didn’t fare much better. It’s worth noting that this list contains many of the same companies on the cable provider list, but consumers are asked to rank cable services separately from broadband services. The overall satisfaction for ISPs grew from a 62 in 2019 to a 65 in 2020. The same three providers are at the bottom – Frontier (55), Suddenlink (57), and Mediacom (59). At the top are the same two providers – Verizon FiOS (73) and AT&T Internet (68).

Part of the explanation of the change in approval ratings for the industries might be little more than statistical variance within the range of sampling. The rankings of individual ISPs vary from year to year. Consider Charter, ranked as an ISP. The company was ranked highest in 2013 and 2017 at a 65 ranking and lowest in 2015 (57) and 2019 (59). This year’s increase might just be variance within the expected range of sampling results.

What matters a lot more is that our cable companies and ISPs are generally consumer’s least favorite companies. This has always benefited smaller ISPs that compete against the big companies. One of the most common forms of advertising for smaller ISPs is, “We are not them”.

People don’t rate cable companies and ISPs so low due because they deliver technical products. Other technology sectors have much higher satisfaction ratings such as landline telephones (70), cellphones (74), computer software (78), internet search engines (76), and social media (70). Consumers are also like electric utilities a lot more than cable companies and ISPs – electric coops (73), and investor-owned and muni electric companies (72).

It’s always been somewhat disheartening to work in an industry that folks love to hate. But I’ve always been comforted by the fact that my smaller ISP and cable clients generally fare extremely well when competing against the big ISPs and cable companies. I have to assume this means people like small ISPs more than the big ones – or perhaps hate them a little less. That’s something every small ISP should periodically consider.

Will Cable Companies Ignore Pleas for Faster Uploads?

One of the biggest impacts of the pandemic on broadband networks has been that homes suddenly care about upload speeds. Homes that largely lived off of downloading video transitioned to having adults and students at home and simultaneously trying to connect to remote work and school servers. People who were happy with their broadband speeds pre-pandemic suddenly found their broadband connection to be inadequate. Industry statistics show that huge numbers of people have upgraded to faster broadband products hoping to improve the home broadband performance without realizing that their performance bottleneck is due to inadequate upload speeds.

The cable industry has largely ignored upload bandwidth in the past. DOCSIS technology that operates the cable broadband networks allows as much as one-eight of total bandwidth to be dedicated to uploading. However, many cable broadband connections are configured with something less than that, because very few homeowners, other than perhaps photographers or others professionals who routinely send big files have ever cared about upload speeds. To make matters worse, the cable industry generally has squeezed the upload data stream into the portion of a cable network spectrum that has the most noise and interference. That never mattered to most people when sending files, but it matters when trying to maintain a steady connection to a work or school WAN.

But suddenly upload speeds matter to a lot of households. Some of the current pressure on upload speeds will be mitigated as K12 students eventually return to the classroom, but there seems to be widespread consensus that we’re going to see more adults permanently working from home.

It’s going to be interesting to see how the big cable companies react to the upload crisis. I’ve not seen many of them talking about the issue publicly and I suspect they are hoping this will go away when the pandemic ends. The cable companies have to know that they will be competing against technologies that offer faster upload data speeds. AT&T built fiber in the last few years to pass over 12 million homes. Telcos like CenturyLink and Windstream are quietly building fiber in some communities. Verizon says it’s going to pass 30 million homes with its fiber-to-the-curb technology using millimeter wave spectrum. And private ISPs are edging fiber into cable markets all over the country.

The cable companies have possible solutions on the horizon. Cable Labs recently announced the release of the DOCSIS 4.0 standard that will allow cable companies to offer symmetrical bandwidth. The gear for this technology ought to hit the market starting next year, but industry tech writers who interview cable company management seem to agree that the big cable companies have no appetite for paying for a new round of upgrades.

The cable companies all upgraded to DOSCSIS 3.1 in the last few years that added the capability for a gigabit download product. The web is full of pronouncements from the CTOs of the big cable companies saying that they hope to get a decade out of this last upgrade. Are any of these companies going to be willing to make a major new investment in new technology so soon after the last upgrade? In many markets the cable companies have become de facto monopolies, and that inevitably leads to milking technology upgrades for as long as possible.

There are a few other technology upgrades on the horizon that could provide relief for upload speeds. There has been a move by several vendors to explore expanding the bandwidth used inside a coaxial cable. A coaxial cable network acts like a captive radio network that uses a big range of different frequencies. Cable systems historically used as much as 1 GHz of total spectrum. In recent years with the advent of DOCSIS 4.0 that’s been expanded to 1.2 GHz of total spectrum. The technology now exists to upgrade cable networks to 1.8 GHz. That would provide a huge additional pile of spectrum that could be dedicated to bandwidth. But such an upgrade would require changing out a lot of network components such as amplifiers, power taps, and modems. Such an upgrade might be nearly as expensive as a shift to DOCSIS 4.0.

The bottom line is that any significant changes to create more upload bandwidth inside cable networks will cost a lot of money. I bet that the big cable companies will stick with the current technology migration plan that would keep DOCSIS 3.1 for the rest of this decade. Likely the only thing that might prompt cable companies to upgrade sooner would be competitors mass deploying technologies that are marketed for having faster upload speeds. The most likely future is that the big cable companies will introduce DOCSIS 4.0 selectively in the few markets where they are feeling competitive pressure, but that most of households are not going to see the upload speeds that people now want.

Breakthrough in Video Compression

Fraunhofer HHI, Europe’s largest research organization recently announced a new video codec, H.266, or Versatile Video Coding (VVC). This represents a huge breakthrough in video compression technology and promises to reduce the size of transmitted video by 50%. This is big news for ISPs since video drives a large percentage of network traffic.

Codec is an acronym for compressor/decompressor. Codec software is used to prepare videos for streaming over the Internet. Codec software compresses video signals at the sender’s end and is used at the viewer’s end to decompress video. The decompressed video file you watch on your TV, computer, or smartphone is much larger than the video file that is transmitted to you over the Internet.

Codec software is used to compress video signals of all types. It’s used by online video vendors like Netflix and YouTube TV. It’s used by networks like ESPN that broadcast live sports. It’s used by online video games. It’s used in online chat apps like Zoom. The codec is used to compress images from video cameras that are transmitted over the web. Any video you receive online has likely been compressed and decompressed by codec software. Fraunhofer claims its codec software is included in over 10 billion devices.

Reducing the size of video files will be a huge deal in the future. Sandvine reported in October of 2019 that video represented over 60% of all downloads on the web. We know the amount of streaming video has exploded during the pandemic, aided by massive cord-cutting. Cisco predicts that video could grow to be 82% of downloaded web traffic by the end of 2022.

The new H.266 codec standard will replace earlier codec software H.264 and H.265. Interestingly, the H.265 codec reduced the size of video files by 50% compared to the predecessor H.264 codec. Fraunhofer says the software is particularly well-suited for transmitting 4K and 8K streaming video for flat-screen TVs and for video with motion like high-resolution 360-degree video panoramics.

The new codec won’t be introduced immediately because it has to be designed and installed into the network gear that transmits video and into all of the devices we use to watch video. Hopefully, the new codec will hit the market sooner than its predecessor H.265. That codec software was announced on a similar press release by Fraunhofer in 2012 and has just recently been implemented across the network.

H.265 got embroiled by a number patent disputes. The new H.266 codec might encounter similar problems since the team working on the codec includes Apple, Ericsson, Intel, Huawei, Microsoft, Qualcomm, and Sony. Fraunhofer is trying to avoid disputes by implementing a uniform and transparent licensing model.

There also might be an eventual competitor for the new codec. The Alliance for Open Media announced a new codec in 2015 call AV1 which is a competitor of the current H.265 codec. This is open-source software and free and is supported by Google, Microsoft, Mozilla, and Cisco. (Note Microsoft is backing both sets of codec software). This group has been working on a forward-looking codec as well.

Even should everything go smoothly it’s unlikely to see the H.266 codec affecting consumer video for 3-4 years. Carriers could deploy the codec on network gear sooner than that.

The Reverse Donut

A lot of rural areas are going to get fiber over the next five years. This is due to the various large federal grant programs like ReConnect and RDOF. New rural broadband is also coming from the numerous electric cooperatives that have decided to build broadband in the areas where they serve rural electric customers. This is all great news because once a rural area has fiber it ought to be ready for the rest of this century.

These new fiber networks are going to revive and transform many of these areas. People who want to work from home will move to existing rural homes and build new homes. It doesn’t take a lot of high-paying jobs to revive a rural economy. Rural communities are also hoping that fiber can slow the drain of people migrating to cities to find work.

However, nothing this transformational is without consequences. I’m already starting to see some of the consequences of what happens when rural areas get fiber but the towns in a county don’t. I’ve been referring to this phenomenon as the reverse donut, where all of the rural areas around a county seat or mid-sized rural town have fiber but the town doesn’t. Today, most of rural America has better broadband in towns than in rural areas, and maps of broadband in most counties looks like only the donut holes have broadband.

Every broadband grant program in the country is aimed at rural areas that have little or no broadband, and that’s how it should be. But I’ve worked in dozens of counties where everybody just assumes that broadband in towns is okay, particularly if a town is served by a cable company.

In many cases, this is not true. Small rural towns often don’t have the same quality of broadband as larger towns. DSL in smaller towns is often of the oldest vintage and delivers speeds under 5 Mbps. Small town cable systems often underperform. Such systems might have been built in the 1970s and have been largely neglected since then. Aging and deteriorated coaxial cable performs even more poorly than old telephone copper since the network acts as a huge radio antenna and attracts spectral interference through any open cable splice point. It’s also not unusual in smaller communities to find neighborhoods that don’t have cable broadband. The houses might have been built at a time when the local cable company didn’t have the money to construct new cables.

When my firm helps communities to do speed tests, it’s not unusual to find a significant percentage of cable subscribers in small towns with download speed far under 100 Mbps, and in some of the worst cases under 10 Mbps. It’s a big mistake to think that cable company technology translates to good speeds because when the network is poorly maintained this is often not true. It does no good for a cable company to jam new technology upgrades on top of bad copper.

There are going to be some predictable consequences of communities with the reverse broadband donut. New housing construction is likely to occur outside town instead of inside of in town. That means that over time that the demand for government services will shift. Most counties have geared services like law enforcement, school transportation, trash services, and numerous other government services and programs around serving the county seat. Over time property values in towns will dip compared to newer homes with fiber in rural areas.

This is not to say that any of these changes are bad – but having better broadband in rural areas instead of towns will definitely change communities over time. I talk to people in rural county seats all of the time and many of them are incredulous that there is no grant money to help them get better broadband. The good news is that it’s often possible to build a profitable fiber network in small towns without any need for grants. But that is not going to happen unless towns take a proactive approach to attract an ISP willing to invest in their community or even decide to build their own fiber network.

It’s my belief that a county or community is not done the job until everybody has great broadband. Counties that will be getting rural fiber are lucky if their towns already have good broadband. But many counties will look up in a few years and see the consequences of having the reverse donut.