The Telescope of the 2030s | Hacker News Reader

The Telescope of the 2030s

nytimes.com

28 pointsdnetesn18 comments

Great article. I really enjoyed the part where he became a little more aware of the world in which he lives. We really do spend chump change on science.

"I used to think $10 billion was a lot of money before TARP, the Troubled Asset Relief Program, the $700 billion bailout that saved the banks in 2008 and apparently has brought happy days back to Wall Street. Compared with this, the science budget is chump change, "

TARP didn't just hand over $700 billion in cash. It was a financial instrument. According to Wikipedia, it actually made money -- $441 billion in revenue over $426 billion invested.

[0] https://en.wikipedia.org/wiki/Troubled_Asset_Relief_Program

Yes, it was the example used in the article. There are other ways to show how little money goes into science. Apple, for example, has more cash in the bank than the cost of the Apollo Space Program:

http://appleinsider.com/articles/15/01/28/by-the-numbers-app...

The U.S. is one rich country. It would be beneficial if we could funnel more money into science. It usually pays dividends.

Private funding pays dividends, public funding leaks money.

The problem is the lack of incentives for private funding.

public funding leaks money

Public funding's issue is that its value, while very real and substantial, is difficult to quantify because it's not in the business of making money. It doesn't deliver a product in the classical sense, it doesn't have a calculated ROI. Improving the lives of humans isn't something an accountant can plot on a spreadsheet.

Public funding is in the business of making new discoveries - many hold little immediate market value but a tonne of value when implemented by private industry years and even decades down the track. That's hard to put a solid number on.

Add to that, private industry leaks money as well. Any sufficiently large organization will require more resources to communicate a policy/directive/goal or pivot than a smaller one. Look at it like an organism - it eats more, wastes more but leaves big tracks in the sand and can attack the bigger problems that smaller organisms can't.

Wish the Irish bank bailout was going to make any sort of return in our investment.

Right now is more or less the worst possible time to plan out a new space telescope. There's one simple practical reason, the JWST hasn't been launched yet. Once it has then that'll end it's budget dominance and if it succeeds then it'll hopefully start to pump some fresh air back into the space science community, after having sucked most of it up for the past decade with its cost-overruns and schedule slips.

But far more important than that are the drastic changes coming to space launch. Reusability is on the horizon, and innovative launch concepts are being tested. Almost assuredly launch capabilities and launch prices 10 years from now will be very different. And that is the primary driver of cost and program complexity when it comes to orbital space telescopes. High launch costs means that spacecraft need to eek out the utmost performance from within tight payload margins, that raises cost. It means there's no room for contingencies like backup missions, which raises the stakes. And because the cost and stakes are so high that means the vehicle has to be over-engineered so that it has a very high probability of completing its primary mission, which further increases cost. Even though launch doesn't end up being the majority of mission cost, it still has an enormous impact on it.

In a decade or so it may be possible to launch payloads at somewhere between half to 1/10th of current costs. Which opens up a whole new vista of opportunities and changes the local cost/benefit maxima sharply. It will likely be possible to build, launch, and operate a Hubble-class telescope for less than a billion dollars, perhaps only a few hundred million dollars. That makes it more feasible to test innovative designs, it makes it easier to launch a fleet of similar or identical telescopes to increase science throughput, and it makes it feasible to simply replace failed telescopes. That may very well change the way we design and build spacecraft, relying more on extra mass instead of bleeding-edge over-engineering to provide longevity and reliability. But it'll also provide new insights into orbital telescope design, instrumentation, as well as abundant new science data, all of which will inform next generation telescope design, likely leading in different directions from the designs and budgets we would choose today.

So why can't the ISS be used to host the telescope, however large it may be? It seems to have all infrastructure in place already to be able to do something like that.. except maybe the starshade.

Probably something to do with ideal orbits for different purposes.

iirc, too low, and too much vibration.

It could resolve objects only 300 light-years in diameter — the nucleus of a small galaxy or a gas cloud on the way to collapsing into a star and planets, say — anywhere in the observable universe.

The article opens talking about exoplanets rather than very distant galaxies. What does that kind of resolving power - which is deeply impressive - mean for looking at stuff in our immediate neighborhood?

I wonder if basing such a large telescope, or group of telescopes, makes sense on the Moon for the ease of construction and maintenance. At either pole a large array of segmented reflectors could be constructed and added to.

I wonder if the thin atmosphere or moon-quakes make this non optimal.

Dust above the surface of the moon is a concern, but another point to consider is that it takes extra fuel to lower down into the gravity well of the moon. Fuel is mass and mass is money. It would be less expensive in terms of fuel to not go to the moon.

Well, I think that the moon wouldn't be too feasible since it does rotate. If you land telescopes on the moon, you wouldn't be able to observe the sky for a long period of time, like the hubble did with the deep field (50 days of time).

It also doesn't make sense to geosync the telescope behind the "dark side" of the earth for the same reason. Long exposures wouldn't be easy since the earth does move around the sun, so the sky moves with it. Eventually you will have to face the sun to get the same shot of the sky.

So, the only way would to be to have the orbit around the sun so the telescope can always face away from the sun. that way you can at least create the long exposures they need.

Is the location refered to, a million miles from Earth and orbiting the sun, a Lagrangian point?

I think so -- that's about how far we are from L1.

From the linked PDF report: "Nominal orbit: Halo orbit about the Sun-Earth L2 point."

The linked report also answers the "why there" question I immediately had.

> The HDST spacecraft and its operations will likely share many commonalities with the JWST architecture and operations. For HDST, as for JWST, an L2 halo orbit is preferable to a geosynchronous orbit (and most others, including low-earth orbit), because it is thermally stable and has lower fuel requirements for station-keeping

https://static1.squarespace.com/static/558adc44e4b002a448a04...

(if that link expires, it was from here: http://www.hdstvision.org/report )

If it's going to be human serviceable, it could also be assembled as well. No need to send it up in one piece. That opens up a lot more launch opportunities and makes everything much more flexible.

Reply on news.ycombinator.com