On 17/5/25 02:39, Timothy Collinson - timothy.collinson at port.ac.uk
(via tml list) wrote:
> I had a chance to grab a quick look at the latest New Scientist (17th
> May 2025) whilst at work yesterday.
>
> Highlight (or, to be fair, perhaps the most useful(?) for Traveller), was:
>
> *Alien Dyson Spheres may self-destruct before we spot them* - Alex
> Wilkins, p.10
> swarms of satellites would probably be plagued by an avalanche-like
> cascade of collisions - at their orbital speeds that could be disastrous
> even a swarm of fewer, larger satellites or those placed in thin
> orbital shells to avoid satellites crossing paths, couldn't  last much
> longer than a million years unless actively managed,
> arXiv, doi.org/pmg3 <http://doi.org/pmg3>
> we are thus unlikely to come across them after alien civilization has
> gone extinct as they are "only visible while being used"
>
Given an initial, exactly-centred-on-system-barycentre setup for a dyson
_shell_, _any_ perturbation from that results in one chunk of the
structure being a little closer to barycentre, and as a result, the
chunk on the opposite side, will result in a net gravitational force
pulling the closer chunk yet closer to barycentre unless counteracted. 
That's an additional failure mode to consider.

ObTrav: The well-armed lunatics (ie, PCs) come across an Ancient-built
dyson sphere (of whatever subtype) in the process of embuggerising
itself via turbo-Kessler as outlined in that paper you linked to.

> I did have one question though, is it still a Dyson Sphere when it's a
> 'swarm' or is that something else? Oh hang on, I've got Blevins' book
> _Megastructures, that will have the answer.
> And indeed it does:
> Dyson Spheres are made up of three kinds:
> - Dyson Swarms/Rings
> - Dyson Bubbles (statites with solar sails to keep them 'in place')
> - Dyson Shells (what I was thinking of as a Dyson Sphere)
>
> My apologies if you all already knew that.  But it might save someone
> looking it up.
>
> (Also: Highly recommend _Megastructures_!)
>
> My mind still boggles at the amount of real estate any of the above
> would offer a budding traveller.
>
> tc
>
For a prototypical dyson shell around a G2 class dwarf (such as the star
closest to Terra), a 1 AU shell would give 549,316,406.25 x the surface
area of a planet like Terra- call it 550 million times more surface area.

Coruscant-level population (of 1 quintillion - billions being pop code
A, trillions being pop code D, quadrillions being pop code G, and
quintillions being pop code K (skipping I)) on the planet would,
assuming linear scale up with area, result in approx 550 septillion
population (sextillions being pop code N, septillions being pop code R
(skipping O), so hundreds of septillions would be pop code T).

That's assuming only _one_ significant layer of habitation, on both sides.

I leave infrastructure for such a population centre to the reader.

Alex