Since all the design and tool-up costs are sunk, why not build two or three? Even just as components.
Otherwise if we get to March 2020 and the launch fails, or the scope or shade doesn't deploy, that's $8 billion down the drain.
I'm using modest language ("very complex") - but the actual complexity is more like "uniquely mind-boggling". (I happen to know the MIRI cryocooler system engineer - just another Harvard PhD whose thesis required liquid helium.)
For more on the MIRI cryocooler, which is the critical path item in JWST, and has been for some time, see: https://en.wikipedia.org/wiki/MIRI_(Mid-Infrared_Instrument).
*
The cost of these flagship missions, and the all-or-nothing problem that you point out, has caused conflict in the Astrophysics community. It might be better to have a pipeline of lower-cost but more frequent missions.
A recent National Academy of Science report on exactly this question (https://www.nap.edu/catalog/24857/powering-science-nasas-lar...) concluded that the large missions did offer a unique benefit in pushing boundaries of what's possible. That report pinned the real problem on getting correct cost estimates, not the notion of large missions in itself.
Not really. The amount of scientific and technological developments that we get even from failed missions is substantial. When starting a new project, all the findings from previous projects are not ignored.
What you are proposing is not necessarily a bad idea but, if that was the case, I would expect some comment saying: why do they construct three equal telescopes when we could use all that money and man-power to build three different ones that would help us much more?
To extend it ab absurdium, it does remind me of the joke about the theoretical physicist going back to bed when he finds his house aflame, because he invested in a fire extinguisher so a solution exists.
If nothing else the intelligence agencies could probably come up with a use for it.
Even cubesats and GPS sats are special, you can't simply take a GPS sat and repurpose it for a TV station. The requirements can be vastly different (geostationary vs LEO, precise orientation vs simple attitude control).
The same goes for space telescope. You can't just take a ST sat body, ram in a LEO weather satellites electronics and expect that to stay in orbit or work reasonably.
a huge chunk of the cost of "building" a spacecraft is all of the testing, calibration, integration, validation, etc etc that goes on after the physical objects are assembled.
yeah, the machinists can crank out a second housing or structural element for nothing after they've made the first one. the electronics assembly techs make a handful of most every board, to start with.
but only one set of all of them is handed back to all the principal investigators to go through the months/years of prep work necessary to ensure the physical assembly that's been produced will function correctly, for years, on orbit.
Rarely do we as citizens ever hear of their planning (if it exists) for alternative outcomes.
To be clear, I'm exempting disaster preparedness and stuff like that.
But if we look at all sorts for things, from wars in Iraq/Afghanistan, to legislation like Dodd-Frank or the ACA, all we ever hear about is the best case scenario.
I think your comment is spot on. The problem is the government only has one mode of thinking, and that has to increase waste tremendously.
If anyone here can corroborate or contradict my theory, I love to hear either.
It's totally lost on Americans, but "Made in the USA" on the bombs other countries buy and then use against their adversaries is not lost on people who reap the destruction. Americans have forgotten how we ran out of shit to blow up during the Korean war, so we decided to blow up their dams in order to flood entire villages down stream. This is after the Geneva conventions were signed. The idea the United States is a peaceful country is very tenuous.
So I sent a note to my Congresscritter Cory Gardner, over a month ago about JWST and WFIRST with a really simple question, if he supported their continued funding to project completion. I included this link with it: https://www.forbes.com/sites/startswithabang/2018/02/20/canc...
The response I got was a POS canned non-response to the question. But at the same time he talked up how great Colorado aerospace is, pretty much making it sound like if the juice doesn't come to Colorado, he doesn't give crap about aerospace. But he doesn't have the integrity to actually commit one way or another to these two projects. It's disappointing.
But, I think your comment is getting at something very important, which is to have a pipeline of missions that run in some orderly way, so that you can build on prior knowledge and success.
Like the original Mars Pathfinder rover (1997) paving the way for Spirit/Opportunity (MER, 2004), and then Curiosity/MSL (2012) and next Mars 2020 (2020).
These have shared a lot of components, and some prototype systems that started out as "experiments" on MER made their way into "production" in MSL - such as autonomous data-taking (http://www.planetary.org/blogs/guest-blogs/2018/0313-automat...), and computer vision to detect obstacles during descent (https://www-robotics.jpl.nasa.gov/projects/projectImage.cfm?... -- this moldy old page is from MER, but the same idea was later used on MSL and slated for M2020).
If you can set up this kind of pipeline, it's very powerful.
Turning to page 160 of the (long!) Decadal Survey document (free!), we read:
"The committee carefully considered the alternative of several rover missions instead of sample return [for Mars]. It is the opinion of the committee that sample return would have significantly higher science return and a much higher science-to-dollar ratio. [...] Existing scientific knowledge of Mars makes it possible to select a site from which to collect an excellent suite of rock and soil samples to address the life and habitability questions..."
So, one way to reframe your objection is that we are in truth planning multi-mission coordinated observations, but in the context of sample-return. M2020 is going to select and store samples, and a subsequent mission is planned to return them. That's the kind of progression that seems to galvanize the science community.
Aside: Besides the above reports, which (I hope) provide reassurance that all options are being considered, there's also the context of the site-selection process for M2020 - see https://marsnext.jpl.nasa.gov/index.cfm. You might enjoy looking at the viewgraphs for the very last talk of the most recent (2017) workshop, which summarize some of the site considerations.
https://www.washingtonpost.com/national/health-science/nasa-...