More practical assuming that Starlink does not go out of business or get destroyed. Giving them the benefit of doubt I assume perhaps incorrectly that a panel of people risk ranked such things. Starlink is a military target and a single point of success. Their first customer was the US Army and the only reason it actually exists at all as Starlink would not be profitable otherwise. A few countries have already been saber-rattling to take them out, no idea how they would do it.

Diversifying into cable, DSL and fiber could mitigate some risk for the government employees working from home. I am biased of course because the deployment of my fiber in a very rural area was partially funded by the US government. Not every ISP will use that money correctly and without fraud. That would be my bigger concern is that someone is validating results and that money won't just end up in executive pockets.