Coming to Terms with Fake News

newsFake news (the “purposely deceptive” kind, not the “I don’t like it” kind) is starting to be recognized in American society for the threat to democracy which it is.

Two major development occurred recently.

First, the Mueller investigation issued 13 indictments against Russian nationals for efforts to subvert the 2016 election. While the indictments are consistent with what American intelligence agencies have been saying for many months, the fact that the announcement comes from the Mueller investigation and actually names real people brings a new level of legitimacy and concern to the issue.

Second, a new Air Force research paper (as reported in the Washington Post) provides many specific examples of how Russian, Islamic State, and other propagandists effectively manipulate Twitter and other social media platforms to sow dissent and falsehoods. Efforts are at the national level — such as support for the Trump or Sanders campaigns — or at the local level — such as effectively heightening racial tensions at the University of Missouri.

While the description of the problems are detailed, the description of effective solutions are not. At least the first step to confronting the issue of fake news is to clearly identify and define the problem.

A Quick Primer on Global Poverty

poverty“Extreme Poverty” is currently defined as an individual living on less than $1.25 per day. Using this benchmark, in 1990 47% of the world’s population was living in extreme poverty. As of today, that figure has dropped dramatically, now approaching 10%. The UN identifies the first of its “Sustainable Development Goals” as the elimination of extreme poverty by 2030.

Much of the progress in reducing extreme poverty is thanks to efforts in five countries: China, India, Indonesia, Pakistan, and Vietnam. These countries alone moved an astonishing 700 million citizens out of extreme poverty between 1990 and 2010. Unfortunately, in Africa during the same period, the number of people in extreme poverty rose from 290 to 414 million people.

The task of eliminating extreme poverty gets progressively harder the closer we get to zero. Those still in extreme poverty are generally in rural or remote areas, lacking electricity, sanitation, transportation, internet, or other fundamental services.

While the number of people in “extreme poverty” is dropping, it is important to remember that most of the planet is still extremely poor. About half of the planet lives on less than $2.50 per day.

Internet.org

internet-org.pngInternet.org is the Facebook funded and directed initiative to increase global connectivity. The name “Internet.org” confuses some, since it sounds like a neutral, non-profit initiative. It is in fact an initiative of Facebook. Current site branding refers to “Internet.org by Facebook”.

On launch of Internet.org in 2013, much of the media attention related to connectivity issues, in particular comparing Facebook’s Project Aquila (drones) to Google’s Project Loon (balloons). Internet.org continues to pursue new connectivity options (although with setbacks, including the explosion of a 2016 SpaceX rocket carrying an Israeli satellite to be used by Facebook in Africa).

More recently, Internet.org is more closely associated with the Facebook Free Basics program, which provides free access to Facebook and other selected content in over 60 developing countries. In these efforts, Facebook is simultaneously lauded for subsidizing access while criticized for restricting content. Facebook employs a system for deciding which third-party content will be included — such as wikipedia, wikihow, education sites, health sites (but not including content from direct competitors). Internet.org also offers access for app developers to the Innovation Lab at Facebook Headquarters in California. The lab mimics bandwidth and device constraints in many global environments.

Controversies around Free Basics became particularly intense in India, to the point where Facebook cancelled the service there in 2016.

Internet.org describes many legitimate motivations for expanding global broadband access. Its impressive progress, however, is also clearly in Facebook’s corporate interest. In many locales “Facebook” is synonymous with “the internet”. For example, Research in 2015 showed that 65% of Nigerians, 61% of Indonesians, and 58% of Indians agreed with the statement “Facebook is the internet.” (Only 5% of Americans agreed.)

Facebook’s penetration in developing countries outside of Internet.org and Free Basics also includes WhatsApp, which it acquired in 2014. In many countries WhatsApp is the dominant app for new users.

Facebooks metrics show that it is having great success penetrating new markets. Facebook grew from one billion daily users to two billion since 2013. Only 41 million of those new users are from the US and Canada.

Sustainable Development Goals and Broadband

sdgIn 2000, the United Nations established the Millennium Development Goals (MDGs), a set of eight development targets to be achieved by 2015. Each goal had associated metrics and timelines.

Progress towards the goals was uneven at both a country and international level. Some goals were not achieved (such as reductions in child and maternal mortality rates), while other goals were actually achieved early (such as global reduction in poverty, mostly thanks to China and India).

In addition to mixed success, the goals prompted debate about whether the best, most legitimate eight goals were chosen. There was a parallel debate around the chosen success metrics.

Despite the shortcomings or disputes, however, the MDGs are widely credited with increasing attention, funding, and coordination around fundamentally important global milestones.

As the end of the 15 year window approached, the United Nations launched a follow-on effort entitled the Sustainable Development Goals (SDGs). Careful to avoid criticism of a hasty selection of targets, the UN considered literally hundreds of possible goals, eventually (and painfully) winnowing down to 17 goals — including 169 “targets” and 304 “indicators”. (The large number of goals, targets and indicators unleashed a new wave of criticism.)

So what does all of this have to do with broadband everywhere? A lot, actually.

It’s quite easy to point to the key role broadband will play in achieving each of the goals. Goal 1, for example, “elimination of poverty”, will be directly impacted due to increased economic growth. Goal 2, “zero hunger”, will be directly impacted through better communications, coordination, and policy implementation.

Some goals stand out as particularly tied to broadband access. Goal 3, “good health and well-being”, and Goal 4, “quality education”, are particularly intertwined with the internet. Goal 8, “decent work and economic growth”, is as well.

Broadband indeed plays such an important role across the SDGs that the UN’s Broadband Commission updated their name in 2015 to the Broadband Commission for Sustainable Development.

Aadhaar Basics

aadhaarAadhaar is the Indian government’s resident identification number system which has registered about 1.2 billion Indians — nearly the entire country. While Aadhaar doesn’t directly relate to internet communications, it is both dependent on and enabling of many online services.

Aadhaar is a random 12 digit number that links to biometric data, typically a photo, ten fingerprints and two iris scans. Soon Aadhaar will include facial recognition services as well.

Registration with Aadhaar is voluntary, but so many government and commercial services are now linked to Aadhaar (welfare programs, pensions, banking services, mobile phone accounts, etc.), Aadhaar is used by essentially everyone in India.

Like any powerful technology, Aadhaar brings both profound advantages and significant risks.

An Aadhaar identification number allows Indian citizens — including its poorest — ready access to information and services heretofore unavailable. It increases efficiency and decreases corruption.

On the other hand, identity services such as this can lead to data breaches, fraud, and abuse.

Many countries are developing or refining their own identification systems. Aadhaar is currently the largest, and in many ways, most impactful personal identification system on the planet.

SpaceX Readies Starlink Test

maxresdefaultTwo down, 12,238 to go.

SpaceX is scheduled to launch the first two of its Starlink broadband satellites, with future ambitions of launching 4,425 Ka/Ku band low earth orbit satellites, and an additional 7,815 V band very low earth orbit satellites. SpaceX remains tight-lipped on the initiative, but FCC documents reveal that initial testing on the network is beginning.

On February 21st, SpaceX is scheduled to launch from Vandenberg AFB a Spanish radar satellite called Paz. A secondary payload includes two test Starlink satellites, called Microsat-2a and Microsat-2b. The satellites, in an orbit of ~315 miles, will test over six months optical communications between themselves, as well as Ku / Ka band communications with ground stations in Seattle (where the initiative is based), SpaceX headquarters in Hawthorne, Tesla headquarters in Fremont, and other sites (including three specially-configured mobile vans).

The satellites reportedly weigh 400 kg each, or about 1/20th the weight of a commercial communications satellite headed for geosynchronous orbit. Falcon 9 has a payload capacity to low earth orbit of about 23,000 kg, suggesting that SpaceX could launch perhaps 50 satellites of this size at a time. (One has to assume that they are working hard to get satellite size down and numbers per launch as high as possible.) SpaceX already launches ten Iridium satellites per launch (which are much larger). India has launched over 100 microsatellites on one rocket. SpaceX plans to sell launch services using the same satellite bus and requirements for its own Starlink network.

Low earth orbit satellites can be much smaller and offer internet latency periods of 25-35 ms, equivalent (or better) than many cable and DSL systems. Current geosynchronous communications satellites have latency of around ten times that level (and cost much more for internet communications — often hundreds of dollars per day). Because Starlink satellites will be much closer to earth, the coverage area of any given satellite will be relatively small — a circle with about a 1000 km radius — requiring a large number of satellites in the network.

SpaceX plans to launch the full complement of over 12,000 satellites by 2025, with the initial 800 being sufficient to start global internet services in 2020. Internal SpaceX documents estimate that in 2025 their launch business might represent $5 billion in annual revenue, while global communications services might represent $30 billion (based on 40 million subscribers). SpaceX estimates satellite lifespans of five to seven years before starting deorbiting procedures — suggesting that launches of new satellites will be required indefinitely.

FCC Chairman Ajit Pai has issued a statement in support of approvals for SpaceX programs, similar to approvals already received by SpaceX competitor OneWeb.

Digital Evolution Index

DEI_reportTufts University, in cooperation with Mastercard, has developed a Digital Evolution Index (DEI) which ranks countries based on their progress in digital development. The DEI includes 60 countries evaluated across over 100 different indicators. In its main ranking, the DEI compares overall digital development to recent digital progress in order to see which countries are innovating and improving most effectively.

In 2017 the latest report was issued. It includes a number of key findings (nicely summarized in the Executive Summary), including:

  • Three countries in the “stand out” category are making particularly strong progress: Singapore, New Zealand, and UAE. Each has employed active government programs worthy of emulation by other countries;
  • A number of countries in the “stall out” category, such as the US, South Korea, and the Scandinavian countries, are ranked highly but not showing many continued improvements. This is understandable — it is hard to improve forever — but also may indicate complacency;
  • Countries in the “break out” category, such as China, Kenya and Bolivia, are showing great momentum which bodes well for the future;
  • Countries in the “watch out” category, such as Egypt, Pakistan and Peru, are at a low level of development and not showing marked progress;
  • A focus on mobile infrastructure brings the biggest return on investment for least advanced countries;
  • Government policy and initiatives play a major role in the relative success of digital development in a country;
  • Offering access at reasonable pricing isn’t sufficient. Technology providers and government agencies need to provide the privacy, security and accountability to engender trust.

Tufts also provides details about methodology, background papers, and access to the underlying data.

A Quick Primer on the Radio Spectrum

radio-spectrumThe radio spectrum is the range of the electromagnetic spectrum between 3 Hz and 3 THz (a frequency difference factor of a million). Because the radio spectrum is so valuable for many forms of communications, it is heavily coveted, used, and regulated.

Below 3 Hz the capacity to send information is very slow. Above 3 THz the earth’s atmosphere absorbs most of the energy, rendering transmission impossible until frequencies approach those of visible light (which are around 300 THz).

Countries establish national laws regulating the radio spectrum. Those laws are then coordinated through an international body, the International Telecommunications Union (ITU).

In 1947, the ITU defined nine bands for the radio spectrum, giving them impressively descriptive names. For example, the top six bands in the ITU hierarchy are called “High Frequency”, “Very High Frequency”, “Ultra High Frequency”, “Super High Frequency”, “Extremely High Frequency”, and “Tremendously High Frequency”. (I’m not making this up.)

Different frequencies exhibit very different characteristics. The high frequency bands can carry much more information (since they are higher frequency…). Furthermore, different frequencies exhibit different propagation properties: some can travel through water, some can diffract around large objects (like mountains), some can bounce off the ionosphere. The four bands between “High Frequency” and “Super High Frequency” are those competed for most by telecommunications services.

Other nomenclatures for the radio spectrum also exist. The IEEE has a competing set of designations. The EU / NATO / US have yet another nomenclature.

There are many ongoing ongoing policy negotiations around spectrum usages and ownership — often complicated by the rapid change of technology, which allows new approaches to using and sharing bandwidth (through technologies such as spread spectrum transmissionfrequency reusedynamic spectrum managementfrequency pooling, and cognitive radio).

Satellite communications present a new set of policy challenges. Low Earth Orbit (LEO) satellites can share bandwidth with terrestrial sources, so there are many guidelines to be decided to guarantee that satellites appropriately share spectrum with both ground-based and other satellite-based sources.

Update from Davos

wefIn preparation for the World Economic Forum’s annual event in 2017, the International Telecommunications Union (ITU) published a working paper “Connecting the Unconnected” which offers current data and program updates regarding global broadband. It’s a very useful document — the endnotes are particularly impressive — and author Imme Philbeck deserves credit.

The document outlines the many reasons why 53% of the world’s population (~3.9 billion people) is still offline:

  • Access: 1/3 of the world’s population is further than 100 km from a fiber connection (although 84% do live within range of 3G);
  • Infrastructure: Many of the poorest on the planet don’t have electricity — a prerequisite for internet access;
  • Cost: 57% of the world’s population can’t afford internet access as offered;
  • Education: Only 44% of the world’s population has a secondary education or higher — a clear predictor of internet usage;
  • Relevance: Many in the poorest countries don’t see the relevance of online services (and may be actively opposed).

The document outlines that progress is strong in more urban environments, but significant obstacles still exist in rural areas. New technologies (satellite, balloons, drones) may help, but are still several years from implementation. The issue of relevance and cultural acceptance may in fact prove to be the most daunting for the poorest regions.

The Link Between Broadband and Economic Growth

econDoes broadband spur economic growth? The intuitive answer is “of course” (unless new users do nothing but watch cat videos…). Two representative research studies take a more rigorous and quantitative approach to this question.

The first, from the World Bank, is “Exploring the Relationship Between Broadband and Economic Growth” (2016). The author, Michael Minges, reviews a number of studies that collectively show increases of GDP in the 1-2% range for every 10% increase in national broadband penetration.

The second, from Ericsson and Imperial College, is “How Important are Mobile Broadband Networks for Global Economic Development?” (2017). The study reviewed data from 135 countries and concluded that a 10% increase in mobile broadband penetration causes a 0.6-2.8 percent rise in Gross Domestic Product (GDP).

Both of these studies confront difficulties in researching this topic: securing timely and accurate data, teasing apart correlation and causation, trying to implement randomized controlled study design. Nonetheless, the preponderance of evidence suggests that broadband adoption has prompt and significant impact on GDP.

 

Addendum: A new study from the National Bureau of Economic Research concludes that the arrival of broadband to 12 African nations was responsible for 10 percent of the rise in employment.