Gary Kim has been a digital infra analyst and journalist for more than 30 years, covering the business impact of technology, pre- and post-internet. He sees a similar evolution coming with AI. General-purpose technologies do not come along very often, but when they do, they change life, economies and industries.
23 years ago on June 24, 1998, AT&T acquired Tele-Communications Inc. for $48 billion, marking a reentry by AT&T into the local access business it had been barred from since 1984.
Having spent about two years amassing a position in local access using resold local Bell Telephone Company lines, AT&T wanted a facilities-based approach, and believed it could transform the largely one-way cable TV lines into full telecom platforms.
That move was but one among many made by large U.S. telcos since 1994 to diversify into cable TV, digital TV, satellite TV and fixed wireless, mostly with an eye to gaining share in broadband services of a few different types.
By some accounts, TCI was at the time the second-largest U.S. cable TV provider by subscriber count, trailing only Time Warner. TCI had 33 million subscribers at the time of the AT&T acquisition. As I recall, TCI was the largest cable TV company by subscribers.
For example, in 2004, six years after the AT&T deal, Time Warner Cable had just 10.6 million subscribers. In 2000, by some estimates, Time Warner had about 13 million subscribers. That undoubtedly is an enumeration of “product units” rather than “accounts.” Time Warner reached the 13 million account figure by about 2013, according to the NCTA.
Since 1994, major telcos had been discussing--and making--acquisitions of cable TV assets. In 1992 TCI came close to selling itself to Bell Atlantic, a forerunner of Verizon. Cox Cable in 1994 discussed merging with Southwestern Bell, though the deal was not consummated.
US West made its first cable TV acquisitions in 1994 as well. In 1995 several major U.S. telcos made acquisitions of fixed wireless companies, hoping to leverage that platform to enter the video entertainment business. Bell Atlantic Corp. and NYNEX Corp. invested $100 million in CAI Wireless Systems.
Pacific Telesis paid $175 million for Cross Country Wireless Cable in Riverside, Calif.; and another $160 to $175 million for MMDS channels owned by Transworld Holdings and Videotron in California and other locations.
By 1996 the telcos backed away from the fixed wireless platforms. In fact, U.S. telcos have quite a history of making big splashy moves into alternative access platforms, video entertainment and other ventures, only to reverse course after only a few years.
But AT&T in 1996 made a $137 million investment in satellite TV provider DirecTV.
Microsoft itself made an investment in Comcast in 1997, as firms in the access and software industries began to position for digital services including internet access, digital TV and voice services. In 1998 Microsoft co-founder Paul Allen acquired Charter Communications and Marcus Cable Partners.
I remember all these events well, as I began working in the U.S. telecom industry in 1983, just prior to the breakup of the AT&T system in 1984. As a result, much of my work has been in the area of competitive communications.
Gary Kim has been a digital infra analyst and journalist for more than 30 years, covering the business impact of technology, pre- and post-internet. He sees a similar evolution coming with AI. General-purpose technologies do not come along very often, but when they do, they change life, economies and industries.
Consumer broadband access quality (speed, latency, availability) always is an evolving and moving target with distinct profiles that vary by geography and other social measures. Rural areas will always lag urban areas on performance metrics, simply because of the high cost of building networks in areas of low density.
Demand also plays a role, as the presence of larger accounts, business customers and higher-income customers will always be higher in cities than rural areas. And suppliers always respond to higher demand.
So inequalities of network performance can persist, even when demonstrable progress is made. In 2020, for example, at least 37 percent of U.S. home locations could buy gigabit services. 54 percent of locations could buy service operating between 100 Mbps and 900 Mbps. About seven percent of locations could buy service operating at the minimum 25 Mbps speed defined as “broadband.”
The areas of real need are the four percent of locations that could not buy service at the minimum. Notably, though, the percentage not able to buy “broadband” is decreasing.
Keep in mind that most people in the United States live on just six percent of the U.S. land surface, according to the USDA. Unsettled or rural areas are exceedingly common.
About 94 percent of the U.S. land surface is unsettled or lightly populated, including mountains, rangeland, cropland and forests.
The point is that rural coverage is important, and also difficult in a continent-sized country with so much largely unpopulated areas.
That noted, everyone expects household data consumption to increase. The trick is planning for future capacity needs in a way that is graceful, and which does not cause firms to fail because they made unwise decisions about how much capacity to supply.
Hundreds of megabits per second is today’s standard.
Tomorrow’s standards will be in the gigabits per second, in all likelihood. The trick is to scale supply to match demand, without over-investing, too soon. Business failure is the risk when suppliers invest in capabilities people do not wish to buy.
In that regard, much of the capacity demand will be driven by content consumption, as has been the case over recent decades. And content consumption is powerfully affected by the number of users sharing any single connection. So much of the demand for the fastest speeds (roughly equating to volume of data consumed) is generated by multi-person households.
Multi-user households using many concurrent super high-quality video applications (extended reality, for example) represent one possible requirement for multi-gigabit-per-second data rates.
Source: Fiber Broadband Association
An important caveat, then, is that there is a huge difference between projected data demand for households of various sizes. A four-user household might require 2 Gbps. A single-person household using the same applications, with the same intensity, might only require 500 Mbps.
A two-person household using the same apps, with the same intensity, might require 1 Gbps. The problem is that all networks must be built to satisfy the requirements of the most-demanding users as well as lighter users; single-person accounts and multi-person accounts.
The U.S. has an average household size of 2.5 persons. That might imply an “average” per-account speed of 1.5 Gbps by 2030.
source: Fiber Broadband Association
Gary Kim has been a digital infra analyst and journalist for more than 30 years, covering the business impact of technology, pre- and post-internet. He sees a similar evolution coming with AI. General-purpose technologies do not come along very often, but when they do, they change life, economies and industries.
Fixed wireless using 4G and 5G will be important at the margin for connectivity service providers. More than 70 percent of all service providers are now offering fixed wireless, according to Ericsson.
Out of the 311 service providers Ericsson recently surveyed, 224 offered fixed wireless services. That said, fixed wireless will be important for most service providers only at the margin. At present there are perhaps 75 million active fixed wireless accounts in service globally.
There are about 1.12 billion fixed network broadband subscriptions globally, Ericsson and others estimate. So less than seven percent of active fixed network internet access lines are provisioned using fixed wireless.
Of course, fixed wireless matters quite a lot for some connectivity providers who can leverage fixed wireless to sustain broadband operations where fiber to the home is unaffordable or where some providers can take market share.
T-Mobile in the U.S. market, for example, has zero share of the home broadband business generating as much as $195 billion in annual revenues. If T-Mobile can use fixed wireless to take just one percent share, that is an incremental $2 billion in annual revenues.
source: Ericsson
Gary Kim has been a digital infra analyst and journalist for more than 30 years, covering the business impact of technology, pre- and post-internet. He sees a similar evolution coming with AI. General-purpose technologies do not come along very often, but when they do, they change life, economies and industries.
As parts of the world gradually attempt to emerge from the Covid pandemic, business-to-business and business-to-consumer operating practices are in flux. Face-to-face business meetings disappeared, replaced by video conferencing and other tools. Industry exhibitions and trade shows were virtually impossible for much of the past 18 months.
As we recover, there remains much belief that virtual formats will remain more important than in the past. So what do we know, at this point, about what works--and seems not to work--for virtual events?
So far, the consensus of business-to-business trade show professionals is that some aspects of the face-to-face experience do not translate well to virtual formats.
“Virtual exhibit booths do not work for some events,” says a report by the Center for Exhibition Industry Research (CEIR), which surveyed 346 executives globally who are responsible for putting on business-to-business trade shows.
The survey reports “exhibitors are complaining of inadequate ROI,” including lead generation, for example.
Also, “motivating attendees to opt-in to engaging with exhibitors has been problematic,” the survey finds. Similar problems appear in the area of engagement.
“Engagement and networking in the virtual space is difficult,” the study says. “Keeping attendees engaged for extended periods of time is challenging as is getting them to log back in.”
“There is some pushback on the part of attendees to engage via video; this makes networking and engagement with exhibitors challenging,” the study says.
Event organizers have found that virtual event sales cycles are longer.
Gary Kim has been a digital infra analyst and journalist for more than 30 years, covering the business impact of technology, pre- and post-internet. He sees a similar evolution coming with AI. General-purpose technologies do not come along very often, but when they do, they change life, economies and industries.
Definitions always matter, and especially so when considering eligibility for government funds to support whatever business a firm happens to operate. Lots of people disagree that the U.S. federal government definition of “broadband” should remain at 25 Mbps/3 Mbps. And, over time, the definition will be changed. The only question is when, and to what minimum level.
So some advocate a minimum broadband definition of 100 Mbps/100 Mbps. Support for that definition includes some in the U.S. Senate. As always, there are trade offs and business implications.
Capital investment requirements; efficient use of scarce capital; user behavior; willingness to pay; current and anticipated usage profiles all are important. Cable networks have an advantage in basically having upgraded virtually all U.S. plant to gigabit speeds downstream. Upstream improvements are much more difficult, costly and time-consuming.
How the upgrades can happen within the feasible limits of today’s business models, for all would-be suppliers, is key. Government subsidies are important at the margin, but most of the upgrade activity has to be undertaken by private suppliers.
And that means there must be a clear understanding of how the upgrades fit the business models. In that respect, the upstream definition will be more challenging than the downstream definition.
Up to this point, the key element in the broadband definition has been downstream speed, as that remains arguably the most-important single numerical indicator of “quality.” But all observers agree that upstream speeds now are more important. And that is the rub.
Telcos and independent internet service providers can rip out copper and replace it with optical fiber access at a faster pace, to be sure. But the business model is challenging. Were it not so, they’d already have done so.
5G, fixed wireless and satellite networks also would be challenged to supply 100 Mbps upstream, though there is a path to incremental downstream upgrades that do not break the business model.
For the majority of U.S. households and locations served by cable TV networks, the 100/100 standard would be troublesome only in the upstream direction, but still would require reworking of most of the physical plant.
For telcos the challenge would be far greater, requiring a replacement of copper access facilities. For every fixed network operator save Verizon, the 100/100 definition would require ripping out and replacing a majority of physical plant.
Rural areas of low housing density would be especially troublesome.
Though cable operators might have a more-graceful upgrade path, telcos generally must rip out the existing legacy network and replace it with entirely-new infrastructure to meet the 100 Mbps minimum upstream standard.
Estimates vary, but a huge telco capital investment would be required to meet a 100/100 minimum. Customer demand is an issue, but less an issue over time, as most consumers now buy faster services than they used to, and often pay more money than they used to, for broadband service.
The 100 Mbps downstream goal is more realistic. About 49 percent of U.S. residents buy fixed network service operating between 100 Mbps and 200 Mbps.
Nearly 32 percent buy services running faster than 200 Mbps.
But a significant percentage choose to buy services operating at lower speeds. Some 20 percent of all customers purchase services running no faster than 75 Mbps, according to Openvault data.
We can argue that 80 percent of the market already buys service at speeds as high, or higher than, the proposed 100 Mbps minimum definition. But 20 percent of the market does not do so, possibly because they cannot buy anything else, but many also might choose not to buy service at 100 Mbps.
Setting a higher minimum definition will happen. But it matters what the definition entails. Virtually no platforms could meet the 100 Mbps upstream definition quickly. FTTH networks could do so, but only at a cost that stresses the current business model.
And that matters. If 54 million U.S. homes are served by fiber to premises networks, and there are about 138 million total U.S. homes, then fully 61 percent of telco passings would have to be replaced to meet the 100/100 standard.
One might argue that fixed wireless, 5G or some other network could meet the 100 Mbps downstream speed. But none of the other networks are engineered to support 100 Mbps in the upstream path.
More than anything else, it is the impossibility of practical mass market networks hitting the 100 Mbps upstream speed that is the key problem.
Gary Kim has been a digital infra analyst and journalist for more than 30 years, covering the business impact of technology, pre- and post-internet. He sees a similar evolution coming with AI. General-purpose technologies do not come along very often, but when they do, they change life, economies and industries.
According to Elmar Rode, Oracle Global Industry Solution Lead, up to 60 percent of Oracle leads come before the involvement of any sales personnel. In other words, potential customers are doing their research online, before contacting potential suppliers.
According to Marc Halbfinger, CEO of PCCW Global, the firm gained 800 accounts within the last year completely without the involvement of any sales personnel at all.
All that suggests something fundamental has happened in business-to-business sales operations. At the very least, we will see more automation, more use of artificial intelligence, more remote interactions.
But what else will change?
Rode and Halbfinger are among the subject matter experts who will speak about such issues on a PTC webinar July 21, 2021, at 14:00 HST. The series is restricted to PTC members on that date, but the event will be viewable by everyone about 30 days after initial screening. You can view episodes here.
Gary Kim has been a digital infra analyst and journalist for more than 30 years, covering the business impact of technology, pre- and post-internet. He sees a similar evolution coming with AI. General-purpose technologies do not come along very often, but when they do, they change life, economies and industries.
