
I had a rather unexpected encounter with my own past last week. While looking through some articles I wrote during my time with Cable & Wireless and Mercury Communications, I rediscovered a Technology Watch paper I wrote in 1992/93 about Low Earth Orbit satellites — LEOs. I had completely forgotten about it.
Given how much I have written about Starlink and satellite communications in the Falkland Islands over the last few years, it surprised me to find I was writing about essentially the same technology more than thirty years ago.
But rereading it was even more interesting. The satellites, speeds and companies have changed almost beyond recognition. Some of the questions haven’t.
2,400 bit/s sounded exciting
One of the systems I discussed was Globalstar, which was promising worldwide voice and data communications. Its proposed digital data speed was up to 2,400 bit/s.
Yes, 2.4 kbit/s.
Today Starlink can deliver Internet speeds measured in hundreds of megabits per second. The difference illustrates just how much the surrounding technology has changed. I also wrote about Orbcomm, whose experimental satellite was intended for low-speed data and specifically wasn’t designed to carry high-speed data, video or voice.
Nobody in 1992 was talking about sitting in a remote house and streaming television through a small satellite terminal.
But the bigger questions sound strangely familiar
What surprised me wasn’t how primitive some of the technology now appears. It was how familiar some of my concerns sounded. I wrote:
“It is not clear at present what the exact roles of system owners, service providers, satellite operators, are to be.”
More than thirty years later, that remains an important question. Indeed, Starlink has taken vertical integration much further. SpaceX builds the satellites, launches them, owns and operates the constellation and provides the service to customers.
I also worried about American dominance:
“The European space industry has been very slow to get involved in these systems, leading to the possibility of dominance by US equipment manufacturers.”
That doesn’t sound particularly dated either.
And I worried about monopoly
Perhaps the passage that surprised me most concerned spectrum and competition.
I wrote that it was:
“entirely possible that the first system deployed could create monopolisation…”
The technical circumstances today are very different, and I certainly won’t pretend that in 1992/93 I somehow predicted Starlink. I didn’t.
But the underlying question is remarkably similar. What happens when the enormous cost and scale required to build a global satellite network give one operator an overwhelming first-mover advantage? That question matters particularly to the Falklands.
I have strongly supported allowing Islanders to use Starlink. It has transformed Internet access for many people and demonstrated what modern connectivity can provide. But I have also repeatedly argued that the Falklands should not replace dependence upon one telecommunications provider with dependence upon another.
Apparently I was worrying about much the same problem 33 years ago.
I was right to be sceptical about the economics
I was enthusiastic about the possibilities of LEO systems but distinctly cautious about their commercial prospects.
I wrote:
“The reported investment cost of big LEOs, in the range of 1 to 4 billion US $, leads to a good deal of scepticism about the real commercial feasibility of such projects.”
And about Iridium:
“The announced Iridium handset cost and tariffs would seem to limit usage to users who are not concerned with costs. This would severely limit market growth.”
That concern proved justified. The original Iridium service launched in 1998. Less than a year later, the company filed for bankruptcy protection. The engineering was extraordinary. The original economics weren’t.
The idea wasn’t wrong — it was early
That may be the biggest lesson from rereading the article. People understood the fundamental attraction of LEO satellites more than thirty years ago.
But the technologies surrounding them weren’t ready. Launches were extraordinarily expensive. Satellites were expensive. Terminals were expensive. Electronics and digital processing were far less capable than they are today. And the enormous worldwide appetite for Internet connectivity that supports today’s satellite broadband market barely existed.
SpaceX didn’t invent the idea of LEO communications. What it did was bring together advances in rockets, satellite manufacturing, electronics, antennas and networking that finally allowed the concept to operate at enormous scale.
Full circle
I finished my old article by saying:
“Big LEO projects encompass a great deal of vision and possibly represent the ultimate in personal communications.”
I wouldn’t use quite those words today. But perhaps they weren’t too far off.
What I certainly couldn’t have imagined was that, more than three decades later, I would find myself involved in a campaign about the use of a huge LEO constellation in the Falkland Islands—and writing repeatedly about the opportunities and risks that came with it.
And I certainly didn’t remember that I had been asking some of those questions since 1992/93.
Sometimes old files are worth keeping!

Chris Gare, OpenFalklands, November 2026, copyright OpenFalklands
