Any Port in a Storm | Hacker News Reader

Any Port in a Storm

cringely.com

51 pointsrfreytag19 comments

This comment on the article particularly struck me:

    There was a time in the U.S. history could have launched
    a dozen or so satellites in its sleep. “Need another GPS
    communication satellite? It’s on order. We’ll build it by
    Friday and launch it by Wednesday. We’d do it faster, but
    it’s Presidents Day on Monday, and you know how those Unions
    are with holidays” I read some NY Times articles about the
    U.S. response to the Pearl Harbor attack. How in months, we
    built giant cranes, uprighted ships, and rebuilt our fleet.
    Now, we can’t even build another measly two track tunnel
    under the Hudson river without political posturing.

http://www.cringely.com/2010/11/any-port-in-a-storm/comment-...

This isn't a fair comparison. The United States has always been able to muster its resources with great efficiency in an emergency. That hasn't changed with time. You'll remember about 9 years ago when we had a similar situation and we managed to establish a military presence, mobilize and to a certain extent complete a military operation half way around the world in just a few months.

But for a government to move that fast politicians have to act rashly and that can lead to some nasty side effects like putting Japanese citizens in internment camps. Preventing that rash behavior is why society becomes litigious under normal circumstances.

I'd be the first to admit we've become too litigious but I can't see that stopping us if we made the space program as much of a priority as Pearl Harbor or 9/11 was.

Edit: Just to prove the point there were apparently 40,000 deaths by industrial accident in the first year or WWII alone. So clearly there was a cost to that speed of production.

http://www.historykb.com/Uwe/Forum.aspx/world-war-ii/3992/de...

> Just to prove the point there were apparently 40,000 deaths by industrial accident in the first year or WWII alone. So clearly there was a cost to that speed of production.

How many industrial accident deaths were there in 1936-39?

I feel obliged to point out the limitations of LORAN. Current-generation systems are only accurate to within a few hundred metres and they are just as vulnerable to geomagnetic storms as satellite-based systems. LORAN-C operates on 90 to 110khz, so it is vulnerable to multipath interference caused by skywave propagation. At present it also has relatively limited coverage, most obviously in North America. LORAN is a wonderfully simple, rugged aid to maritime navigation, but it is no substitute for GPS.

Modern radio systems are so slick and mature that it is easy for us to imagine that they are a magical force akin to electronic telepathy, but they are reliant on simple electromagnetic radiation and highly susceptible to interference. Beyond the obvious issues of space weather, propagation and accidental interference, most radio systems are trivially easy to deliberately interfere with and we are now at a stage where intersatellite conflict could cause massive destruction. We have no realistic means of either defending satellites or of removing orbital debris. GPS is tremendously vulnerable, particularly to malicious attack and we are best to assume that it will fail completely at some time in the immediate future.

"It would fry satellites, overload power grids, destroy all our computers, and possibly put the lights out for most of us for months..."

If this happens, I think losing GPS would be the least of our worries.

Would a faraday cage protect equipment? Obviously the power grid would be down, but combine solar with low-power electronics and it wouldn't be too hard to equip oneself with a running computer after the event.

Yeah, as long as you don't run out of tin-foil hats you should be fine.

What about simply having computers and other machinery powered down? If we situated several probes closer to the sun, they could warn us like a canary in a coal mine. The satellites would simply transmit a heartbeat and status signal along with electromagnetic observations data. When one fo the satellites fails, we'll have a couple of minutes to shut machinery down and get the most sensitive equipment into Faraday cages.

Wasn't sure if the EMP pulse would be picked up regardless of power state due to induction in wiring or casing.

If you're right about powering electronics down, a early warning system definitely is a no-brainer.

Do you have any idea how quickly the corona ejection would travel, and hence how long it would take to hit earth? It would take a bit over 8 minutes for any message from a probe near the sun to reach us, and if the ejection is travelling at (random number) 0.5c, it would only give us a bit over 8 minutes to react and get the word out. Obviously, if it was slower (0.1c), we would have a lot more time. The worst case scenario would be it travelling at >0.9c, giving us under two minutes to react and do everything.

It's perhaps a much, much better solution to keep a large (and I mean really fucking large, since it might be all we had to bootstrap ourselves back up to our current level of technology) store of equipment in a large Faraday cage. Probably several, distributed for redundancy, really. This could be combined with probes, of course.

I've made this point on HN a few times, but "space weather forecasting" has become an industry in itself. A lot of scientists have made predictions for the next cycle. The big problem is that cycles last 11 years, and the predictions are of the form "Very strong cycle", or "Abnormally weak cycle". You can see why "This researcher has correctly predicted the last two cycles" can easily be attributed to luck. Given that a research career spans at most 4 such cycles, it's hard to demonstrate one's ability.

There's a few respected solar scientists who have written papers to the effect that such forecasting is way beyond our capabilities (there's hints of chaotic behavior in the sun). But these papers don't get any citations, because if they're right, most of the money in this field will dry up. No one wants that to happen.

With that said, Cringely speaks of a two-year forecast, and I don't know if that is short-term enough to be feasible. Still, I'm skeptical of the impending doom predictions, because it takes only one known scientist willing to beat the drum for the government to worry about its space-bound investment (and quite reasonably so).

Someone could probably make money by launching 6 or so observation satellites around Mercury and at Mercury Trojan points. Since mass ejections are directional, we should be able to quickly confirm that as the cause of failure for a satellite with the others, and send out a digitally signed warning via the networks. This should give us a couple of minutes time to shut machinery down.

I think we already have a couple of satellites deployed. (Source is a documentary I watched last month)

According to Wikipedia, CMEs have a top speed of 3200km/s. The earth is 147Mm from the sun (at its closest), so we'd have a minimum of 12 hours warning.

How vulnerable are electronic devices that are powered off to CMEs?

I imagine that while we have sufficient warning, there is also considerable variation in CME speed (and size?). It might be difficult to coordinate "turn stuff back on" day without functioning communication systems.

Just to note: Mercury is incredibly hard to get to, and it's almost impossible to maintain a stable long-term orbit there, at least in terms of our fuel technology. Forget Lagrange points, even just orbiting it is hard. http://en.wikipedia.org/wiki/Exploration_of_Mercury

The U.S. has a probe on the way: http://www.nasa.gov/mission_pages/messenger/main/index.html

NASA posted this today:

http://science.nasa.gov/science-news/science-at-nasa/2010/08...

The Omega system was shut down years ago, to the joy of local BASE jumpers.

Assuming we have a corona ejection, how long a warning would we get?

Between 18 hours to a few days.

Flares can take as little as 15 minutes (with early warning from satellites in L2).

For high energy x-rays there'd be no early warning through other means than forecasting.

Okay that should be enough to put a faraday cage up and the gear the network down.

Reply on news.ycombinator.com