100gig, but 30gig for streaming

Actually, it's a cost/profit/marketability problem.

Imgine the sunk cost in launching a satellite, running and maintaining around 20 ground stations with connections to the internet, and then having the staff to solve problems and do software/hardware maintenance.

Now imagine the number of terminals and the price to sell it at that's required to make it a profitable and sustainable business... Needs to be lots and lots of people, and there's still limited internet resources - the bandwidth you can actually pass through said satellite at any given time and still stay relatively operational, or:
    Total Bandwidth / # users = bandwidth available to each user

I think we once figured out that all you need are about 8000 people (400 per ground station) running an HD stream at any given time to bring the whole system to it's knees. Given that there are millions of subscribers, that's not hard to do.

What I'm trying to explain is that there are 2 immutable facts:
1. The system has limited resources that can't be changed without increasing the capacity of the satellite.

2. The system will always have a minimum 500mS lag time as long as it's geosynchronous, and due to other technical reasons, is not optimum for streaming, or any kind of real time communication (video or audio) when trying to apply terrestrial technologies that demand better responsiveness.

Bottom line, in order to contain the problem, we need less people streaming and eating up resources, not more.

They are already involved in LEOs and will move to it when it's ready for 2-way communications. Right now all you have is receive/only (e.g., SiriusXM) and that's because the orbit (actually a constellation of satellites) is constantly moving so it can stay aloft. Imagine having to have a tracking gimbol on your current dish to track a moving satellite just so it can transmit to it.