I wouldn't go so far as to say sole, the US DoD has plenty of their own very high costly and capable geostationary satellites in addition to the commercial transponder capacity they purchase. I'm referring to the AEHF, WGS, UFO and similar programs. And they'll definitely want some belt and suspenders type redundancy.

If and when Kuiper actually becomes a real thing, or oneweb (the UK government investment and ownership stake is encouraging, there), I don't see them sole sourcing to starlink either.

But all I can say is that I'm glad my 401k doesn't include any geostationary operators in it.

> If and when Kuiper actually becomes a real thing, or oneweb (the UK government investment and ownership stake is encouraging, there), I don't see them sole sourcing to starlink either.

Starlink has far greater capability than Oneweb, simply due to having a much bigger constellation. Oneweb can't compete with Starlink in terms of constellation size, because unlike SpaceX they don't own their own launch vehicles.

What is the motivation for having a second-source to Starlink – technical, economic, political? If it is political, Oneweb may have the disadvantage of being a UK-headquartered company not a US-headquartered one, so that gives the US military a political reason to prefer Starlink (on top of its technical advantages). Unless the US is particularly interested to gain the UK government's favour – and the Biden administration doesn't appear to be particularly interested in that

OTOH, Oneweb has a good chance of getting the UK government and military's business, being UK-headquartered and partially UK-government owned. But the US government/military is a higher value customer than the UK government/military – the US military's annual budget is over 16 times that of the UK's.

I think Project Kuiper being US-owned has a greater chance of getting US military business. But they haven't launched anything yet, and it isn't clear when they are going to start. Their launch costs are likely to be higher than SpaceX's, and SpaceX has a big head-start in the market, which put them at a disadvantage.

I mostly agree with all of this - but the US government is already a fairly big user of o3b. Oneweb, in terms of what their terminals will look like, the slightly higher orbit, and the fact that they'll only be selling to enterprise/carrier/government users and ignoring the consumer end user, is already somewhat similar in business plan to o3b. I wouldn't be surprised at all if they make a serious effort to capture a good portion of US gov business from SES.

It is also highly possible that oneweb may sell the equivalent of a geostationary SCPC 1:1 transponder lease, or an o3b dedicated-capacity circuit, which might not be something that would be technically feasible with starlink. If you're coming from an environment of geostationary looking at $1200 per Mbps per month, anything looks good by comparison.