
In order to answer this question correctly we need to first take a step back and understand how internet works and how data is actually distributed.
1. Undersea fiber optic cables form the physical layer providing global internet connectivity. Fiber optic cable endpoints are available in all the continents [1] and countries having coastal regions. The cables are owned by telecom companies, governments etc
2. The internet from the global backbone is then distributed to users via several means. Wireless (using 3G, 4G, 5G and wifi). Wired using data cables (copper and now increasingly fiber optic).
3. All the distribution is done with the help of what are known as feeders/backhauls etc. These instruments serve as gateways or proxies to the main internet line.
Considering the three points above it should be quite clear that per user speed depends upon two factors.
1. The bandwidth available at the internet lines (more lines directly translate to more bandwidth)
2. The efficiency of the last mile distributors. While backhauls perform a very important role in regulating the internet they might themselves prove to be a bottleneck, as it is being observed in 5G deployment. [2]
Now after seeing the map [1] A natural question might arise that if internet cable is available in all the continents, and almost all the countries then why do 3B people (nearly half of the worlds population) have no access to internet? [3]
The reason is that last mile connectivity is expensive. Generating funds to lay down an infrastructure is difficult. Even after funds are available execution is not that simple.
For example take a look at this performance report on Central African Backbone project for connecting CAMEROON, CHAD, CENTRAL AFRICAN REPUBLIC, SÃO TOMÉ AND PRÍNCIPE, in which the world bank invested some $24 M
Click to access ppar_centralafricanbackbone.pdf
So rather than speed the more pressing issue at hand is how to get more people connected.
The good thing is that people who will get internet for the first time will enjoy far better speeds because of installation of a more flexible, future proof infrastructure from the start.
In cities while the internet coverage has been good it has traditionally been routed on copper cables. Copper cables have a low upper limit on the amount of data that can be carried so many places around the world will need to upgrade their underlying cabling as the demand for data increases.
This is a very very expensive process. I’m not talking about expense in terms of money invested, but rather the uphill task of acquiring licence, pleasing the regulators dealing with civic authorities.
Its a miracle that such a big infrastructural undertaking has been successful in the first place. To do it all over again…. I’m not sure if that would be the best of ideas.
On the other hand rural areas that have had poor connectivity will get all the benefits of fiber, when they do from the start.
And for the first time ever, it may be actually cheaper to set up internet in a village rather than overhaul the infrastructure in the city. Because there is a lot more potential for new subscribers as well as less hurdles in implementation, and low competition. This is one of the main reasons that companies like Google and Microsoft (software companies) are suddenly jumping into the rural connectivity, or digitization as they like to call it, business.
Now lets get to the point. How fast is fast enough? What is a good internet speed that operators can sustain for 10–20 years and earn a profit on their investment?
More than anything the speeds depend upon the last mile distribution strategy used. Laying underground fiber is years/decades of work. It may be possible to do it quickly but its going to be expensive. Wireless distribution seems to be the best way to set up cheap and quick internet services.
But then again acquiring cellular frequencies can be costly. Fastest data speeds can be attained on the top of cheapest infrastructure. Otherwise operators will put quota on the data usage, and limit the speed until profits start coming in. And not just that there need to be more local internet service providers that can take the internet from base stations to the home. Handle customer relations, provide support and build a business.
In developing nations there is shortage not only of the internet availability but also of entrepreneurs and skilled professionals who can supply the services to the people.
Even so the prospects are exciting.
On 24th June 2019, for example, [Telxius in partnership with América Móvil](https://telxius.com/en/news-america-movil/) announced [4] 7,300 km, **108 Tbps** cable to connect Puerto San José (Guatemala) with Valparaíso (Chile) with additional landing points in Salinas (Ecuador), Lurín (Peru) and Arica (Chile).
5 developing countries getting a Terrabit internet fiber on their shores. How will this internet be divided among the people its almost impossible to say. But I’ll say this I’d love to stream Game of thrones on any one of the landing points 🙂
**Foot notes and references**
[1] Fiber optic cable map [https://www.submarinecablemap.co…](https://www.submarinecablemap.com/#/)
[2] “*5G backhaul network will need to support hundreds of gigabits of traffic from the core network and today’s cellular backhaul networks are infeasible to meet these requirements in terms of capacity, availability, latency, energy, and cost efficiency*” (https://www.intechopen.com/books/broadband-communications-networks-recent-advances-and-lessons-from-practice/5g-backhaul-requirements-challenges-and-emerging-technologies)
[3] As of 2019 out of **7,716,223,209** people **4,422,494,622** people have access to internet**. 42.7%** of the world is unconnected**.** (https://www.internetworldstats.com/stats.htm)
[4] (https://telxius.com/en/news-america-movil/)
Simple Wave, an inexpensive last mile communication technology
(https://workrockin.github.io/simple-wave)