The James Webb Telescope, the largest science project in US government history | Hacker News Reader

The James Webb Telescope, the largest science project in US government history

arstechnica.com

142 pointsTomte47 comments

This article also fails to mention that this beautiful, shiny telescope will assemble itself with atomic precision after having the snot shook out of it at several G's by the launch vehicle to get it there.

In exoplanet identification and characterization is where I bet the Webb will make it's place in history for the researchers. IR is a great frequency to show liquid / gaseous water in an atmosphere of an earth like planet, and if all goes well, this telescope will be able to detect that.

That is something a spy satellite doesn't have to do, have decent resolution for items next to suns many orders of magnitude larger and brigher than they are.

All of NASA will be on sedatives until this thing unfolds itself successfully. It's even more scary than the Curiosity sky crane landing system.

JWST seems like too much for NASA given the record. Galileo's high-gain antenna deployment failure, Hubble optics flaws, numerous near fatal problems with GP-B, MCO burning up... all of these and more speak to NASA's track record with complex space observatories.

At L2 JWST will be beyond the reach of astronauts at the time it's launched, so all this stuff; cryocoolers, folded sunshield, folded reflector, etc. have to `just work' with no opportunity for repairs. It just doesn't seem likely.

I desperately want it to work, but if it were up to me I'd trade it in for a large but mechanically simple optical interferometer, launched by spacex or whomever, bolted together at the IIS and operated from a serviceable orbit. That would be a lot closer to what one should expect of NASA.

I don't think it's possible to do the science they want JWST to do unless you get out to L2 or beyond. They are basically measuring heat, and Earth puts out a lot of that. And they want to be in an orbit that doesn't pass into shadow, for reasons of solar power. I understand, and somewhat share, your concerns, but JWST is designed the way it is because that's how it has to be done to accomplish its goals.

The cryocooler for the infrared imager has proven to be the toughest design issue with JWST, at least so far, and AFAIK it's still on the critical path for the mission. (http://spacenews.com/northrop-grumman-at-last-set-to-ship-tr...)

I was surprised the cryocooler was not mentioned in the article. The sunshade is there, but that's not nearly as big an engineering challenge.

Fair point.

But try to juggle context, tutorial scientific background, factual overview, news, human interest, and inspiration, and you'll be amazed how fast your allotted 3,000 words gets used up.

Sure, I get it. You covered a lot of ground.

"Putting an astronomical telescope in space was a natural solution, and it followed directly—and, according to some, very directly—from the spy satellite program."

Can anyone expand upon this? I know the NRO has recently donated some of its old gear to NASA, but is there much cross-pollination between the space-operating agencies?

Putting spy satellites in space presented many challenges and lessons that could be directly transferred... - Building, Launching and Orbiting very sophisticated sensitive sensors, lenses, elements and equipment into space. - Developing compression algorithms and comand and control protocols. - Calibrating readings against known ground-based objects

There were, of course, self-destruction and counter-intelligence measures that probably won't be needed...

According to many, hubble was a either a repurposed keyhole satellite, or was developed as a cover for a new generation of spy satellites. Rumour has it there are literally dozens of HST equivalents in orbit, but only one of them points away from earth.

Here's a source which makes this rather clearer - if this wasn't the case, why would the NRO happen to have two hubble-spec telescopes just sat around? http://www.planetary.org/blogs/guest-blogs/jason-davis/nasa-...

I also have trouble with this claim. Its been documented several times that problems NASA stuggles with have sometimes been solved by the DoD, but due to classification cannot be shared with NASA. No, this isn't some evil conspiracy, but NASA is open-ish and the contractors that do their work don't do so under classification, so anything given to NASA would quickly fall into the hands of America's enemies.

That said, I think this is true in a very general sense for more commodity space tech. Lets say, lens grinding on this level has been done for DoD projects, but because that's probably not classified (at least past a certain level), its easy for NASA's contractors to just re-use that lens grinding method than re-engineering a new one. Because the spy satellite programs are classified, we really can't know how much of it goes back into NASA programs, but I imagine something might.

Let's suppose you were an expert in optical design for space systems -- lenses, mirrors, coolers, detectors, electronics, thermal design -- and worked at NASA on a space telescope.

When that space telescope (say it was Hubble or Spitzer) was launched, you would be looking for another project. That's a small world. Maybe you would talk to a friend who works for intelligence agencies and find out what they are looking for.

And the reverse happens too. If you've explored the design space for an intelligence agency, you would be well-positioned to re-use that knowledge for a space telescope. You mention the classification barrier, but the basic physics drivers will be more apparent the second time around.

Case in point: Larry James (http://www.jpl.nasa.gov/about/bio_james.php), who is the Deputy Director for JPL, but who used to be Air Force Deputy Chief of Staff for Intelligence, Surveillance and Reconnaissance at the Pentagon, and who before that was a Shuttle payload specialist. He's a technical executive, but there are pure technical examples as well.

OK, and a second example, of a segmented mirror demonstrator: http://www.nps.edu/About/News/NPS-New-Home-for-Giant-Segment... ("NASA’s James Webb Space Telescope, named for the former NASA administrator and scheduled for launch in 2014 [erp!], will deploy the next generation of SMT technology on a mission to image light from the earliest galaxies in the universe.")

No, there is not much cross-pollination between science agencies and the military agencies. When I was in astronomy one of the professors in my department was on the committee to figure out what to do with the old telescopes that the NRO donated to NASA. From what he said, no one in the astronomy community had any idea that these telescopes existed and they learned about them from the New York Times, just like everybody else.

It has turned out to have put the astronomy community in a bit of an awkward position. Although the telescopes are free, they still need to be outfitted with appropriate instruments, and, of course, launched into space. Each of those projects costs roughly as much as the telescope itself, so the donation really only covers ~1/3 of the total cost of putting a working telescope in space. Unfortunately, the community just doesn't have the budget to do that. Furthermore, the telescopes aren't really ideal for astronomy since they were designed to look down rather than up. (That is, they are designed to focus on relatively nearby objects rather than on objects "at infinity.") But politically we can't turn the telescopes down, since that projects a bad image to Congress. Beggars can't be choosers, after all, so we have to look grateful while we try to scrape together the funds to get something useful out of them.

Nice article, but the diagram halfway down that shows the EM spectrum.... I'm scratching my head over why "Audio" is apparently now electromagnetism.

I was thinking exactly the same thing. Did my professors skip over something in the electromagic classes?

Bet you a dollar that you've got an unmodulated electromagnetic audio signal not two feet from where you're sitting (if not stuck directly into your ears).

More importantly: I think it's fair to assume that someone reading an EM spectrum has the knowledge to figure out what they mean.

I'm a physicist and I have absolutely no idea what you mean by an "electromagnetic audio signal". I think what you mean is an EM wave which happens to carry an amplitude-modulated audio signal , which is not an audio wave.

Audio waves are compressional waves formed in a medium, and require a medium for transmission through, such as air, or rock, or water, or even interstellar gas.

Electromagnetic waves require no medium, and are brought about by energy transference between electric and magnetic fields in free space.

> I have absolutely no idea what you mean by an "electromagnetic audio signal". I think what you mean is an EM wave which happens to carry an amplitude-modulated audio signal

...So you did know what they mean.

> I'm a physicist

Useless nitpicking gives us a bad name. Please stop.

Audio signals may - and usually are - represented by EM waves. Do not confuse "audio signals" with "acoustic signals". The diagram is not only easy to understand, but also technically correct.

The same image appears on https://simple.wikipedia.org/wiki/Electromagnetic_spectrum with the text:

"NOTE: The AUDIO entry in this graphic is there for comparison only. Sound waves and light are two different things entirely."

Imo still somewhat confusing. I've never heard anyone refer to something on the EM spectrum as audio before.

You went to the simple image wikipedia just to find a citation that agrees with you, while the english Wikipedia has a whole artcicle describing it in quite an opposite way: https://en.wikipedia.org/wiki/Audio_signal

If it is confusing, is because you do not understand what a "signal" means. A signal is not the phenomenon itself, it is a representation of it.

An audio signal is a signal that represents sound, but it is not sound itself. If it were, you would not need to call it a signal, it would be simply "sound". In a slightly different use of adjective, an acoustic signal is a signal (that may or not be audio, eg, ultrassound) that is carried by sound waves.

Actually, it only gets confusing if you try to be pedantic, as GP did. Everyone knows what an audio signal in a EM chart.

> The diagram is not only easy to understand, but also technically correct.

It's not easy to understand and I still don't understand it. What is audio doing on this diagram?

Audio is a name for signals that represent sound waves humans can hear. As we are only sensitive to sound in some frequencies, there is a range of frequencies in which audio signals may exist.

This diagram shows it, probably as a more mundane reference for scale.

Author here. "Audio" is there for reference. It never occurred to be that this would cause confusion, but you're not the only one to object to it.

"Audio sideband" is a common term for AM or FM sidebands from dc to a few tens of kilohertz or a few hundred kilohertz. Not exactly the same as calling a 10-km wavelength EM wave an audio wave, though.

I work at NASA Goddard where JWST is being assembled and by coincidence I went over to look at it today and take a picture. Looking at it from the observation area, I am once again reminded how enormous this thing is! Getting a look at something like this can help restore one's perspective, when one is having a bad day, and be reminded of the amazing things that are done here. I am not attached to JWST but the group I'm currently in did some work on the detectors years ago. NASA has a lot riding on this instrument, to be sure. I look forward to the amazing things it will show us when it's finally on station doing science.

Humans for scale:

https://upload.wikimedia.org/wikipedia/commons/a/a3/JWST_peo...

Can you share that photo?

And also who would I ask about getting a tour? ;)

I don't have an account on any photo sharing service. I'll have to set one up.

Goddard has a visitor center, but I don't think there are public tours of the center itself. We do have open house events from time to time. The best way it to ask someone who works here to show you around. Are you in the DC area?

I'm not sure if this works, but let's try:

https://www.facebook.com/photo.php?fbid=10207140470964842&se...

Worked for me! Thanks!

Are the people working on this stressed out? I couldn't imagine working for years/decades on a super complex billion dollar device where you have only one try to get it right.

So there will be one pointed out at the stars, but I wonder how many NRO will point back at us?

Isn't the largest science project in US the Superconducting Super Collider? Yes, it was cancelled, but JWST was almost cancelled as well.

I'm not sure what you are trying to reason. X was cancelled but Y was almost cancelled. That makes no sense.

According to Wikipedia, $2.2 billion was spent on the SSC before it was cancelled in 1993.

https://en.m.wikipedia.org/wiki/Superconducting_Super_Collid...

It's unfortunate that it was never completed, it was going to be bigger than the LHC. Everything we are learning now could have been learned a generation ago.

Claiming we'd have learned everything the LHC taught us a generation ago also doesn't make much sense, considering the vast improvements in sensor technology and data handling that happened between the last generation and this. It's just as likely we'd have seen little or nothing because our "eyes" weren't sensitive enough or we couldn't find the needle in the data haystack.

Perhaps, but in high energy physics, size matters. Or at least size matters more than most other dimensions.

I'd suggest computing matters more.

The SSCO would have been crippled by the sensor and computing technology available at the time. It would have been able to make very small things go "boink" a lot, but it would have had problems capturing a fat enough stream of data to record the results.

LHC wins because signals are statistical, and the more data you have the more reliably you can say "Uh huh - that's a real signal and not just one of those random things that happens sometimes."

See also the current debate about whether or not something real is going "boink" at 750GeV.

Only with respect to generating events, not with respect to detecting and analyzing them.

Adjusted for inflation, the Manhattan project cost nearly US$ 2 billion in 1940s money, or about $26 billion in 2016 dollars says https://en.wikipedia.org/wiki/Manhattan_Project .

I'd wager on Apollo. $25.4 billion in 1973, adjusted to $109 billion in 2010 dollars.

https://en.wikipedia.org/wiki/Apollo_program#Costs

I'd argue Manhattan Project and Apollo are tied. They both could be measured in % of GDP, at .4% each.

I think it might have been better if the "largest" claim had not wound up part of the title, because it's led to a great deal of distracted quibbling. NASA claims that the Webb is its largest-ever science project, and others have judged it to be the largest in US history. If it's important, we can note that the Manhattan project was military, not science, although of course plenty of science was done there. Apollo might not be counted as a single project, or perhaps put in a different category (exploration rather than science?).

Manhattan wasnt one project any more than Apollo was and most of the money went to Uranium enrichment.

http://blog.nuclearsecrecy.com/2013/05/17/the-price-of-the-m...

The page you linked to says:

> Another way to look at this is to say that we usually talk about the atomic bomb as project focused on scientific research. But one could arguably say that it was more a project of industrial production instead.

I emphasized the "a project of industrial production" to show that your own reference says that it is "one project", and it refers to the entire Manhattan project as a single project, and not multiple projects as you seem to argue.

i tried to edit my comment to point that out, i ran out of time.

i was somewhat agreeing with you except classifying it as industrial building construction instead of defense.

Reply on news.ycombinator.com