[AR] Re: Freeman Dyson, RIP

  • From: Henry Spencer <hspencer@xxxxxxxxxxxxx>
  • To: Arocket List <arocket@xxxxxxxxxxxxx>
  • Date: Tue, 3 Mar 2020 14:58:31 -0500 (EST)

On Tue, 3 Mar 2020, William Claybaugh wrote:

It looks to me that this scheme does not require a stupid mistake to create a runaway chain reaction.  It appears to initiate upon *any* mistake, ever.

I was addressing a somewhat higher level: the stupid mistake is even attempting such a scheme without incorporating generous margins against such a perfectly-foreseeable problem.

Relying on a complete absence of errors is not realistic -- one of the stupid mistakes we've made in spaceflight is building systems that fail unless everything is done perfectly. What can be done is to fairly reliably catch errors before they do harm, design the system so it can usually tolerate the ones that slip through, have ways of limiting the damage when even that fails, and learn from those cases so repetitions are caught earlier. This is done, routinely, in many real-world technical fields; it is not some unsolved mystery that is beyond our grasp. The fact that we so seldom do things that way in space is our stupid mistake, not a law of nature.

For example, any scheme with so many objects in similar orbits probably wants a fast-response capability to *clean up* the debris from an accidental collision, so it *doesn't* cause a runaway chain reaction. (NB, in solar orbit, "fast" typically means months, not minutes. Even so, yes, this probably means that the first responders need propulsion systems that free them from constraints like launch windows -- it can't be done with chemical rockets but there's nothing impossible about it.)

Kinda looks like a bad bet to me....

I quite agree that if it were built the way we usually build space systems today, it would be a bad bet. But assuming that a project on this scale would never even *try* to do better is silly.

Henry

Other related posts:

  1. Home
  2. arocket
  3. Archive
  4. 03-2020