What sorts of support will Starlink customers require? The Chilean pilot study might provide some clues.

Five companies are developing low-Earth orbit (LEO) broadband satellite constellations, but, as of now, only SpaceX is planning to market directly to consumers. What sorts of support will they require? A pilot study of Starlink connectivity in remote Chilean communities may provide some answers to that question.

The Chilean regulator, SUBTEL, has authorized a year-long pilot study of Starlink connectivity in remote, rural communities and is committed to supporting them during the year. In an earlier post, I speculated on the applications the communities might want, and in this one, I’ll speculate on the sorts of support they will require, but to set the stage, we need to know what the communities are like.

Sotomó, the first pilot site, is in a forested area on a fjord in southern Chile, and twenty families live in the town. While Sotomó is only about 35 miles from the nearest airport in Puerto Montt, the trip takes about three hours driving or driving plus a ferry ride. The second pilot site is in Caleta Sierra on the coast about 1,200 miles north of Sotomó, and SUBTEL says there will be others.

A diesel generator supplies power 12 hours a day during the winter and eight during the summer in Sotomó, and they have “patchy” cell phone coverage, but it is used mainly for calls and text since data is expensive. Still, the people have had mobile Internet experience and therefore have some familiarity with Internet services and applications.

Elon Musk quipped that Starlink setup is simple—plug it in and point it at the sky, in either order. He didn’t mention the part about requiring a clear view of the sky, and antenna installation and cabling may be difficult. In Sotomó the Starlink terminal is at the school, headed by professor Javier de la Barra. Someone had to attach the antenna to the roof of the schoolhouse where the terminal is located and run a cable to a WiFi access point in the building. A clinic shares the building with the school and they may have run a cable into the clinic.

The Starlink connection gives the users access to online storage, but many applications—like interactive courseware—will have to be stored on a local server that would have to be acquired, configured, operated, and maintained. Since each test site will need a similar local-area network (LAN) and server, they can jointly agree on a standard configuration, setup, and operating procedures.

Professor de la Barra and his counterparts in Caleta Sierra and other test sites in Chile and other nations will be able to share experience and ideas using the collaboration-support tools and services on the Internet and that may require some training or setting up some application to facilitate their collaboration.

For now, Sotomó has just one terminal in a single building, but if demand grows, they may want to create a community LAN, which will require establishing wired or wireless links between buildings and perhaps integrating multiple Starlink accounts. If they decide to connect multiple buildings, they should also evaluate the offerings of Starlink competitors OneWeb and Telesat, which plan to be offering service in 2022.

The above sounds like a job for a part-time system administrator—someone local who will need training initially and subsequent online support as needed. For the pilot study, that training and support may be provided by SpaceX, SUBTEL, or the regional government. The system administrators will also be able to help each other.

Application training and support will also be needed. For example, an educational consultant specializing in online teaching and curriculum development or a consultant familiar with online medical resources and services and telemedicine experience may help Professor de la Barra and the clinic staff. These specialists would be available for consultation, to recommend Internet services, training resources, databases, etc., and to facilitate collaboration and experience sharing among the community of users—teachers and nurses—in their area.

The community members are users, not developers, and during the course of the pilot year, application software will have to be written. For example, community members will want to download software, educational material, entertainment, news, etc. Since the community will be sharing a single link to the Internet, that will require a system for requesting and queuing material for subsequent unattended download to the local server. The scheduling-downloading system will have to be designed and programmed. That is one example of supporting software that will be required, but my guess is that community members will come up with others during the pilot year—perhaps in support of teaching, remote health care, or for Sotomó‘s local industries—aquaculture, tourism, and farming.

Antennas and hard drives will occasionally break. If Internet access catches on, they may be able to justify more fuel for the current community generator or a dedicated generator at the schoolhouse—one more thing to support.

LEO broadband ISPs will be serving remote, hard-to-reach locations like Sotomó, and each operator will have to deal with the mundane issue of support. SpaceX plans to sell directly to end-users, and if Starlink succeeds, there may be 50 or 100 communities like Sotomó in Chile and thousands worldwide. User and system administrator support groups will form and be valuable, but SpaceX will have to play a role in support. I’ve offered some guesses as to the kinds of support needed, and we will learn more about what remote communities need during this pilot study.

Update Sep 13, 2021

For a lightly edited machine translation of this post into Spanish, click here and let me know of corrections to the translation.