NASA Science Mission Directorate
Oral History Project
Edited Oral History Transcript
John C. Mather
Interviewed by Rebecca Wright
Greenbelt, Maryland – 8 June 2017
Wright: Today
is June 8, 2017. This oral history session is being conducted with
Dr. John Mather for the NASA Headquarters Science Mission Directorate
Oral History Project at Goddard Space Flight Center in Greenbelt,
Maryland. Interviewer is Rebecca Wright, assisted by Sandra Johnson.
We thank you so much again.
Mather: Well,
thank you for asking.
Wright: We
appreciate your time. We know you are a very busy man. Especially
today, we would like for you to share information with us about your
role as the Senior Project Scientist with the James Webb Space Telescope.
It’s exciting times for all of us as we look forward to the
launch, but we would like you for you to go back a number of years
and tell us how you first became involved, and then what your duties
are, and how they evolved as they came through those years.
Mather: Starting
back more or less at the beginning, or before the beginning, in 1989
we launched the COBE [Cosmic Background Explorer] satellite, which
was my first project here at Goddard. It took us about four years
to finish off the operations of that mission, and some more years
to finish the analysis. That gets me up to mid-90s. What was I doing
then? I was writing a book about it called The Very First Light [The
True Inside Story of the Scientific Journey Back to the Dawn of the
Universe] with a coauthor John [I.] Boslough and thinking about, “Well,
what do I do next?”
My friends say, “We’re building the Spitzer Space Telescope”—but
it didn’t have that name yet—“and it’s not
going to be big enough. Can we maybe think about another way to get
a bigger telescope up there without spending too much money?”
So we started sketching telescopes that would unfold, but not be nearly
as hard as the Webb Telescope turned out to be. I presented this at
a seminar here, and people laughed at me and said, “We’ll
never do anything like that, that’s too hard.”
However, I think the day before Halloween of ’95, I got this
phone message from Ed [Edward J.] Weiler that said, “We are
doing a study. We want to get rolling on what’s the new telescope
going to be, so call up John [H.] Campbell. He knows what this is
about, and we need a proposal tomorrow. It only needs to be one page.”
It turned out Ed had just received the authorization to spend some
money on something, and this is what he wanted to do.
What I hadn’t appreciated at the time was that my friends were
involved in writing a report calling for this telescope to exist.
You have probably seen it. It’s the “HST & Beyond
[Exploration and the Search for Origins: A Vision for Ultraviolet-Optical-Infrared
Space Astronomy” (1996)] report from Alan Dressler and committee.
It was brilliant. It hadn’t happened quite yet then, but Alan
Dressler went down to meet [NASA Administrator] Dan [Daniel S.] Goldin.
They got acquainted, and they liked each other. Within a few months
of that, Dan Goldin went to the [American] Astronomical Society and
announced that what the committee asked for was much too small, and
we were going to build a bigger one. He got a standing ovation, which
is not exactly common for NASA Administrators. In fact, hardly any
of them ever go to the Astronomical Society, but he did. He liked
it. So that was our first peer review, as I like to say, a standing
ovation for the idea.
Dan knew something that we didn’t all know. He had worked in
TRW [Inc.], which had already been working on segmented telescopes.
He didn’t think they were impossible the way other people would
naturally assume, so “Of course we can build a bigger telescope.
How big do you want?” He announced this, and it was very clear
from that minute that the idea was a good one, and that NASA was going
to support it, so we almost didn’t have to recruit any interest.
People came to us and said, “Can I help do that?”
Pretty soon we recruited a study manager, Bernie [Bernard D.] Seery.
He is a really brilliant engineer and manager, and conducted the studies
necessary to figure out what are all the technologies you have to
have, and make sure that we got them all invented. He was our study
manager for a long, long time.
So what did we have to do? Well, we had to put out contracts to people
who promised that they could develop our new technologies. We had
a list of 10 new things we had to have, and we managed that from here.
We also started negotiating the international partnership, because
Headquarters said, “We need to make this international deal.
It’s good for us, it’s good for them. If we are going
to let other people use the telescope, they should help build it.
And, by the way, pay something for it.”
It turns out to have many other advantages too, but the difficulty
of arranging international partnerships is not so small. It took us
a long time and a lot of changes of the plan, but we did it. I guess
you know the upshot was the European Space Agency contributes the
launch vehicle, and one instrument, and a half of another instrument.
The Canadian Space Agency contributes the fine guidance sensor, which
makes sure we can lock onto a guide star and keep a sharp image, as
well as a research instrument.
All of them get to participate, and their scientists are guaranteed
a certain fraction of observing time. That’s what they get for
this effort. As well, of course, their local industry gets to have
work, so that’s another advantage that people have for supporting
the space business in each country. It keeps people busy, and helps
them be creative and invent things that they never had before, and
develop new products that they can sell.
Clearly, they have been important for our contractors. As you know,
we have got Northrop Grumman [Corporation] as our prime contractor.
They were chosen in 2002. They were still called TRW at the time.
It didn’t take them very long to say, “We are going to
build you a big model.” You have probably seen the pictures
of it, right?
Wright: Yes,
I have.
Mather: They
built that on their nickel, because that enables them to prove to
the entire planet that they can build a bigger telescope than you
can. Of course, they have other customers that would be interested
in telescopes.
Jumping back a little bit in time before that, of course Bernie Seery
had connections, and we made a connection to the DoD [Department of
Defense]. We did do jointly-funded technology development for the
mirrors with the DoD. At a certain point, we decided that we didn’t
have to cooperate anymore, because they needed different things than
what we needed, but it was a good help to get started.
That was the hardest one of all the technologies that we had. We had
to get I think 12 different contracts to different companies, and
they all said, “We can do that for you.” One of them was
correct. That’s why we do it our way with competitions and a
“prove it to me” kind of process, because you can never
actually take claims of future product on faith. And also why it’s
really good to have multiple choices, because if you really, really
have to have something by a certain date, you can’t just take
“yes” for an answer. You have to have two yes’s
at least. That’s a reminder. That’s how we got to the
Moon as well. James [E.] Webb, the man [former NASA Administrator],
made sure we had multiple ways to make sure we could get there.
Wright: A
sound business decision that has proven itself many times. Since we
are talking about the technology, when you mentioned the jointly developed
DoD technology, did you do that for efficiency and time, or for funding?
Mather: Some
of both. When two funding sources come together and say, “We
all need this,” it attracts a lot more attention from the people
who think they are going to develop a product they can sell. A lot
more people want to come forward and say, “We’ll do that
for you.” So that’s good to stimulate the creative force
of industry. It also makes some more money available, too. You can
get farther if it’s not your only nickel.
Wright: And
it develops that partnership or relationship. You mentioned there
were 10 new technologies. Were there some of those technologies that
were harder through the years, to get to the level that you wanted?
Mather: Yes, some of them turned out to be harder than we thought
they would be. We have a cryocooler requirement. You have one detector
system that has to run at 7 [degrees] Kelvin. We thought at the beginning,
we would just use a block of solid hydrogen, protected and well-insulated,
in a thing called a “cryostat” to keep our detectors cold.
That turned out to be difficult, expensive, and not having a long
lifetime. So, okay. The other alternative is make yourself a refrigerator.
We have got compressors, like you have in the kitchen, that compress
a gas that can go expand at the other end, and absorb heat and provide
cooling at the other end. The principle of that has been known for
a century. The practice of it to get it ready to go into space is
pretty hard. So that one was harder than we thought.
Something else that was harder than I thought it would be was the
big plastic sunshield. We’ve got a sunshield that’s as
big as a tennis court. You can’t buy a roll of plastic that
big, so how are you going to do that? The technology was developed
for us special. It involves ways to stick the small pieces of plastic
together to make the big piece. Also, what was harder than people
thought it would be was to get the right shape. We needed the edges
of the sunshield to be pretty much straight, as you see in the model
there [demonstrates]. Otherwise, there may be pathways where the cold
parts can see the warm parts directly, so we had to make sure that
couldn’t happen. That means the edges have to be straight. But
when you stretch out a stretchy thing, it’s pretty normal for
it to come out with a curved edge, so you have to work at that. A
lot harder than we thought.
The price of that sunshield is more than the price of a lot of satellites
we have put up, just because it’s all custom, it’s all
handwork, it’s all new ideas. We use a software called NASTRAN
[NASA Structure Analysis] for analyzing the mechanical properties
of solid things, but it’s not ready yet for something stretchy.
A big, stretchy piece of plastic that has the right shape after launch,
at another temperature, and made out of little pieces? One heck of
a hard problem. That turned out to be much more interesting than I
thought it would be.
Another one that was obvious from the very beginning was learning
how to focus the telescope after launch. I don’t remember if
we talked about that one before.
Wright: No,
we haven’t.
Mather: We
knew we had to do something different, because Hubble [Space Telescope]
was launched and it was out of focus. We did learn from experience
how to do that, again by running a competition by the way. We had
different teams around the country compete to say, “Well, we
know how to calculate this, and we’ll show you how to do it.”
Then we found out which one was the best, and we followed it up, and
it worked.
It turns out to be the essence of the idea we are using for the Webb
as well. It’s done by taking pictures of a star in focus and
out of focus by known amounts. Then you get all these funny-looking
patterns, but the computer knows what they mean. You can back up from
the fuzzy pictures to say, “Move all those motors that control
all those mirror segments just so, and it’ll be fine.”
That’s a huge accomplishment, and we had to demonstrate it,
because, as I like to say, “John Mather thinks it’ll work
is not the same as the hardware does work.” The hardware doesn’t
care what I think.
Anyway, we had many, many rehearsals. We made a small-scale model
test telescope to demonstrate that. But at the end we say, “Yes,
we are confident that will work.” We are about to demonstrate
it by test in the vacuum tank in Texas, at your place [NASA Johnson
Space Center, Houston].
The other technologies, the hard ones included improving the detectors.
We already had detectors that would work at the right wavelengths,
but they were not as large as we wanted, and they weren’t as
sensitive as we wanted, so all of those had to be improved. The way
I describe it, there are the manufacturers that have their secret
recipes with special mouse milk, and they know how to do it, but even
when they do know how to do it, it’s not 100 percent reliable.
You do it the same way the next time, and it doesn’t work that
time, and you say, “What did I do different?” And you
don’t know. So we had a lot of trouble with them. But they are
brilliant.
In the end, these detectors are so incredibly good that it’s
hard to express how good they are. Just to give you a little bit about
that, we measure how good a detector is by how many extra electrons
we get out of a pixel per hour. We get a few extra electrons per hour
per pixel, which means you can take time exposures for a really long
time and build up the light turning into a signal, and see something
incredibly faint. So, we are now able to say to the public, “You
know, if you were a bumblebee hovering at the distance of the Moon,
we’d be able to see you.” And you are nodding as though
you have heard this story.
Wright: I
was interested, yes, when I have seen that. It’s an interesting
statement. It’s very, very descriptive for people.
Mather: Yes.
One of the main reasons is the telescope is cold and big, but the
other one is the detectors are miraculously good. We didn’t
have them before. We had to pay good money for them. Then we had a
surprise with them too, because there was a period where we discovered
that we bought the good ones and we put them on the shelf, and they
went bad just sitting on the shelf. So, we had to figure that out.
And of course they did, they figured it out. There are places where
you just have to have time and money to go do it over.
Wright: I
think someone referred to it as it has a “long development history.”
That was a kind way of saying that things were put in place, and hopefully
exactly as you want it.
Mather: Yes.
But this of course is also the result of a lesson we learned from
Hubble. Hubble was really difficult as well, and it was scarier in
the sense that they counted on inventions to be completed on a schedule
that was not realistic. So we said, “No, we are not going to
do that. We are going to complete the inventions, and then we are
going to make the schedule.” It was still hard, but it wasn’t
so chaotic.
Wright: There’s
so much that you’ve put together, but you started out at some
point with a blank piece of paper. Then you pull in these lessons
from here, and this information from there. Can you share some of
the discussions that you had, of the ideas that maybe didn’t
make it, or that whole process?
Mather: We
can talk about the very, very beginning. The number one question is
where are you going to put the telescope? Because that controls everything
else. The first question was “Is there any possibility of building
one that you can keep near Earth, like Hubble is?” We worked
at it for a while, and the upshot was no, you can’t do it. You
would need a gigantic refrigerator that you can’t buy to keep
the whole telescope cold enough.
Where is the other place you can go? Well, this is called a Lagrange
point two [L2]. It’s a million miles away from Earth, and the
telescope will get cold all by itself if you can put up the umbrella.
So that was our plan, and it’s a different difficulty, but at
least it’s a possible mechanical engineering job to do. Put
it a million miles away and put up your big umbrella. That makes the
telescope cold. That implies a whole lot of other things. Number one,
you are not going to go there to fix it, at least not yet. It’ll
be possible in the future, but isn’t yet. That implies everything
about the reliability process.
A lot of technology that we already know has to be polished up a little
bit to work out there, but we tried to not invent anything that we
would have to invent. People said, “Well, can’t you use
optical laser communications?” And the answer was yes, you could,
but you don’t need to, so we won’t. No need to ask for
extra miracles. Nowadays, you could do that. It would be just as easy
now, but you don’t need it for this particular observatory.
Future ones yes, because they will produce 10, or 100, or 1,000 times
as many bits per day, and so they need different stuff.
We were on the story of how did we all get started. The first discussion
was where do you put it, and we answered that one right away. Then,
how are you going to keep it cold? What sort of umbrella do you need?
So we drew various shapes that you could have. Finally, I think it
was very early 1996, we had a meeting up at the Space Telescope Science
Institute [Johns Hopkins University, Baltimore, Maryland]. We had
a draftsman from Goddard, some engineers, some scientists, and the
people up there at the institute that had been thinking about this
for years. We all got together, and we started sketching. We sketched
something, and it’s turned out about like what we built. The
main features were obvious from the first day, practically.
Now I should back up and say what had been people thinking about before
even I got onto the project. There was a conference in 1988, even
before the Hubble was launched, and they had a big book that was published
about what do we want next, and how much would it cost, and what would
it look like, and what would it do? They were pretty close. At least
some of the articles in that book said, “We want a big infrared
telescope, and this is why, and this is how much it would cost.”
They were closer to it than our team was for a while.
The sort of short answer is, why do you want an infrared telescope?
Because it’s the next big opportunity. You can’t do it
from the ground. A telescope gets very bright at infrared wavelengths,
and the air is opaque at many wavelengths, so gosh, it gets pretty
hard to do astronomy that way. Obviously, we don’t know much
yet. That was even before the Spitzer Space Telescope had been launched,
so we hardly knew anything. But we knew it was going to be the next
opportunity.
Wright: Were
you able to apply a lot of lessons that you’ve learned from
Spitzer? You mentioned that you have used the lessons from Hubble.
Mather: Yes,
some. The detectors they developed for that, we sort of went farther
than they did, but similar ideas. Got personal experience in the end.
We recruited several scientists that worked directly on the Spitzer
to be on our team, so they had personal experience to make sure we
didn’t forget what they knew. Of course, we also recruited locally
here many people who worked on the Hubble Telescope. Their offices
are all up and down the hall here, of people who worked on the Hubble
before they worked on the Webb. That’s another way to not forget
what you know, is to get the same people.
So that was the very early ideas. We also started right away with
scientific teams discussing “What do you want to see?”
The book had already established the basic idea, so then we needed
to get a little more specific and say, “Well, if you want to
see the first galaxies the way the book says, what were they like?
How faint are they, how far away they are, and what color are they?
What wavelengths are they bright?” And then you have to build
your telescope to see that, if that’s your objective. We pretty
quickly concluded that we could build a really big telescope—an
eight-meter was possible—and it would be enough to see what
we thought the first galaxies would be. So, by golly, that tells us
what to do. It’s really hard, but that tells us what to try.
Then, after that, we had a day where NASA and the Space Telescope
Science Institute staff invited industry people to come together.
We said, “This is what we are thinking of doing, and we want
you to understand it, because we are going to draw it and ask you
to build it.” Industry people told us, “Please go away
for a little while. We want you to listen to our idea. Give us an
hour, and we’ll come back to you with our idea.”
And their answer was, “No, don’t do that. Don’t
draw the telescope yourself and ask us to build it. Let us compete
with each other to show you our best ideas. We’ll draw the telescope,
and you’ll choose, but we’ll build. So let’s start
with a competition.” They called it a Cooperative Agreement
Notice, which is something NASA does to sort of share the effort with
our proposers.
So we said, “Oh, we’ll do that. We’ll take the entire
summer that year to do really quick, intense study, and we’ll
report out at the end of the summer on the various ideas.” We
had three teams, actually. Two contractor teams with their university
friends, and the government teamed with the Space Telescope Science
Institute team. We made these three gigantic [Microsoft] PowerPoint
presentations and documents, and we all presented them to each other
at the end of the summer. Of course, the companies didn’t present
to each other. They wanted their secrets. But they presented to NASA.
So now we had three teams all saying, “Yes, we could do what
you need,” and we could afford the budget. Now, we had been
commanded by Dan Goldin to please find a way to do this project for
half a billion dollars. I see you winking at that. We actually didn’t
know if it was possible, but we said we should try. We should look.
And we couldn’t do it in the end. Half a billion, even accounting
for inflation and the changes to the rules, was nowhere close to the
right answer. Eventually, as you know, it cost $8 billion to get to
launch, plus operations, and that’s just the U.S. piece. I don’t
know what the dollar equivalent of the Canadian and European parts
are, but overall the total mission is over $10 billion.
So Dan Goldin’s vision was a little too ambitious, to say the
least. But it wasn’t a command. It didn’t say, “Do
it.” It said, “Try to do it,” so we tried. We had
to go through a lot of ideas about how you might possibly do it that
way, but there are two major reasons why it can’t be done.
One is that we are not organized to do it in that fashion. To be organized
to do it in that fashion you would basically say, “Okay, bring
everybody together in one building, bring everything you need, and
go like crazy.” That’s the model that SpaceX [Space Exploration
Technologies Corporation] does, aluminum in one end and rockets out
the other.
There was no way we were going to do that. Our project is much too
big to do that, the expertise is too dispersed around the world, and
just not going to happen. Legally, it also doesn’t happen, because
we are run by procurement regulations, and you just can’t do
that. So there wasn’t a chance that it could ever have been
done that way.
Also, SpaceX has the advantage that they do it over and over, so they
learn each time to do it better next time, and we don’t get
to do it over and over. We get to do it once. It’s like having
to do it 10 times, but only fly it once. Since you don’t get
to fly it 10 times, you have to rehearse and rehearse, and practice
and practice, and design and fail and design and fail, and then fly.
That’s basically why it takes a lot more work than people like
to admit. So if we were going to do 10 of them, we could indeed do
them a lot cheaper than doing one, but we are not doing 10. By the
time we get around to wanting the next telescope, somebody will say,
“Well, that’s not where the next opportunity is. We want
something different.” We are not going to build another one
of these that I can imagine. So we’ll see.
Wright: What
did you learn that summer from the industry? You said you had those
three presentations.
Mather: We
learned that basically we all had very similar ideas about how to
do it. We realized that it was really hard, but we all had pretty
much agreement about the technologies we would have to finish off
also. The list of 10 things we had to invent was pretty similar in
1996 as it is later on, with the one exception that we had to shift
from a solid hydrogen tank to an active refrigerator.
So that’s how we got rolling. We also started having science
team meetings. Now, this is before there was an official science team,
so I called it the “volunteer science team.” People that
wanted to work on it, they came together, and we talked about what
would be the coolest science to do, and how to calculate everything
that was mentioned in the book. “How hard is this? What do you
want to do?”
Then we decided, “Well, it’s time to get official about
this.” So we said, “Okay, Headquarters will now solicit
nominations and proposals to be on our official science team.”
We called it the Ad Hoc Science Working Group. I chaired that group,
and our job was to say, “This is top priority for the project.
If you could build it, this is what we would use it for.”
This was important because we now have to decide exactly what are
those instruments going to be—how big, how good. And now we
are going to divvy up the project between the United States, Europe,
and Canada, to say, “Okay, who’s doing what part? The
instruments are something that different groups could produce, so
we’ll definitely have to decide how good they have to be, and
who is going to make them.”
We went through that for a long time. We voted on what are the top
priorities, and we wrote a book about what we thought was the best
and most important stuff, and we called it [Next-Generation Telescope]
Design Reference Mission [1998]. This is a sort of computer file saying,
“Observe this for so long, and with these colors and so forth,
and that’ll open up science for us.”
Wright: What
were you looking for when you were reviewing the nominations and proposals?
And how many people did you end up bringing in on?
Mather: I
don’t remember. I didn’t do the selecting, that was Headquarters
that did that. But we would certainly be looking for people who had
convincing expertise on the subject, that had been through the trials
and tribulations of doing other space projects, and that were respected
in the community and everybody knew that those people could do the
job.
We ended up choosing a fairly young crowd. They are not so young now,
but they were young then. We even had a little worry for a while that
these people are too young, but that didn’t last. We did have
external advice from more senior groups from time to time on various
different kinds of things.
Finally, we basically said, “These are the instruments we are
going to build,” and we negotiated our international partnership.
We got ready to choose, "“People have been working on this
a long time by now.” Because the members of that original group
might actually become competitors for the proposals that we are going
to solicit, we have to let them go and create a new group. So we created
a new thing called the Interim Science Working Group, and I chaired
that one as well. At this point, our job was more to advise the project
management and Headquarters about what do we think is going on and
what’s important? While that group was advising us, we were
at that point also preparing to choose the prime contractor, the big
contractor.
That ran along until we actually had the formal solicitation—and
Headquarters did this also—to choose the instrument providers.
The instrument providers included European, Canadian, and Americans.
The American one is the near-infrared camera [NIRCam], and that’s
Marcia [J.] Rieke at [University of] Arizona [Tucson].
The members of all of these teams became members of the new science
team, the current Science Working Group. They have been with us ever
since 2002, when they were chosen. At that time of course people didn’t
think it would take this long, so these people are no longer the youngsters
that would be considered too much of lightweights to have community
respect. They are the community-respected people, and they have all
stuck with us, which is a remarkable endurance feat considering how
hard the job’s turned out to be.
So that’s what I have been doing all this time with science
teams. I have been organizing meetings, conducting the meetings, making
people agree on things, conducting votes, keeping track of things.
Wright: How
often do you meet a year?
Mather: Usually
two or three times a year. And of course as the project grew, I no
longer do it by myself. We now have I think 12 other young scientists—slightly
younger scientists—here at Goddard, and they do almost all of
that. In particular, my Deputy [Senior Project Scientist], Jon [Jonathan
P.] Gardner, runs all these meetings now and he is really good at
that.
All of the other areas also have individually-assigned scientists
here at Goddard to make sure that some scientists are following the
engineering work in detail. We don’t do the engineering, but
we need to make sure that we understand it and do a cross-check. Because
once in a while, scientists think differently about things than engineers
do, and it’s a good thing. Now we could check each other.
The engineers say, quite rightly, “Well, if it’s not broke,
why, don’t fix it. Don’t improve it, it’s good enough
already.” And the scientist once in a while has to say, “Well,
it is not really good. We really should do better about that.”
We have had pretty darn good luck about that. We wrote specifications
that we have not had to relax in most areas.
The engineers said, “We understand why that’s important.
We’ll do that.” And they worked really, really hard to
make it happen. There are only a few things where we had to say, “Well,
that’s too hard. We are going to back off.” The places
we did back off were places that didn’t actually matter that
much. It means that we have an engineering marvel to match the incredible
scientific demand. And you could say, “Well, that’s also
why it’s expensive,” but to tell the truth, there is no
gradual thing in this stuff. If you said, “Well, why don’t
we just leave off all the instruments?” We’ll get no science
and it would still cost almost as much. Which is shocking, because
there are so many large pieces of a big project like this, five or
six different pieces. You have to have all of them, and they are all
expensive. So they say, “Well, why don’t we cut way back
on that?” Well, it’ll still have most of the rest.
Wright: Did
you find that you were making sacrifices?
Mather: We
did have one really important change to make at a certain time. Shortly
after we chose our prime contractor they said, “Okay, the budget
has to go up a lot.” And it became immediately obvious that
we couldn’t afford everything we had been asking for. Even before
we selected them, we said, “Well, we can’t afford the
eight-meter design. We are going to make a 6.5-meter telescope, please.”
We got far enough with the technology development for the mirrors
to say, “We could never afford the time and the money to build
all the big mirror that you said. And besides, it won’t fit
anyway. The rocket’s only so big, and it’s a really snug
fit.” Maybe we could have made it fit, but it sure is snug.
Another thing about the mirror technology—it turns out to be
heavier than we thought. If the mirrors are heavier, you can’t
have as many.
Anyway, that’s all the process we had to go through. The science
team was deeply involved in all of these discussions about “What’s
the real requirement?” But ever since 2002, where we said what
the real requirement was and who the players are, we basically haven’t
had to change anything.
Wright: That’s
an achievement in itself.
Mather: It’s
a major achievement. It means that what we said we would do, we could
do. But the flipside of it is, it was a whole lot more expensive than
we expected.
Wright: Anywhere
along that path that you thought you might lose the project because
of the expense or the budget?
Mather: Yes,
it was obvious that you could lose it, but I thought, “I am
not going to think about that part. I am just going to do my best.”
That’s what we all did. You just can’t focus on the bad
part. You have to focus on the opportunity.
Wright: You
seem to have had the support, though, from the NASA Administrator.
Mather: Yes,
we have had. It fluctuated. Some people were more interested than
others. By about 2011, we knew we were in big trouble. We had been
asking each year for more money, and the curious process in the federal
government is you don’t actually ask Congress directly. There
are so many people between the project manager and the Congress that
it’s like a game of telephone. Each level above you says, “Well,
I don’t think I can ask for that much. I am going to ask for
less.” By the time it gets around to Congress, people are basically
asserting that we can do fine with a small budget, and that’s
not true.
By 2011, it was embarrassing to everyone. NASA was embarrassed, [Maryland]
Senator [Barbara A.] Mikulski—who was a really strong backer
of science—was embarrassed. She wrote us a letter and said,
“Please come back with a plan we can believe. And please get
an external committee to review it, too.” So we did that [James
Webb Space Telescope (JWST) Independent Comprehensive Review Panel
(ICRP)].
That was chaired by John [R.] Casani at [NASA] JPL [Jet Propulsion
Laboratory, Pasadena, California], a very famous project manager of
great honor in his own world. We felt the committee was kind of hostile,
but in the end they wrote a [2010] report that basically said this:
“Did great work, didn’t ask for enough money.” That’s
about the best words you could hear. That’s my version of what
they said. They had a lot of detailed recommendations, but I thought
you couldn’t ask for better.
Congress actually heard these words. I don’t know the process
at Congress—I didn’t have anything to do with that—but
somebody said the right things and Congress said, “Okay, we’ll
send the money.” And they kept that plan. NASA put that into
the budget every year, and Congress said yes every year since then.
So for six years now, we have had a do-plan which we have stuck to.
The Congress sent the money, and we do the plan. Knock on wood, because
it isn’t a guarantee of future performance, but we have done
awfully well. I do not know exactly how we accomplished that. I know
Bill [William R.] Ochs and the project managers made that happen.
For every technical problem that comes up, you have to find a work-around
that you can afford, and they did. That’s another miracle.
Wright: While
we’re talking about management and connections, you are the
project scientist, and you just talked about the project manager.
Can you talk about your responsibilities and how they fit in with
the other parts of the management, and how you have ended up communicating
the necessary things to each other to have it work?
Mather: At
the beginning we said, “Well, these are the general scientific
objectives, and this is what we want to look at.” Then we calculated
some more and we said, “Now, this is how big the telescope has
to be, and how long you have to look, and how good the detectors have
to be. And if you can do that, that would be great.”
We wrote that all down in a book called the Science Requirements Document,
and that got translated into detailed engineering requirements by
engineers. We worked together to make sure that the translation was
correct. And then after that, it’s pretty much been engineering.
Engineers understood perfectly—well, we met and once in a while
they said, “That’s too hard. Can you back off?”
And we said, “Yes, maybe we could.” Mostly, they understood
no, you can’t back that off, because we know why that was important
to us.
So they struggled and struggled, but they solved the tough problems.
And that’s actually good. It’s astonishing how much effort
you can waste saying, “Why don’t we cut back on something?
Are you sure you need that? Maybe 1.5 sandwiches is enough for lunch
instead of two?” You could spend a year figuring that out. In
that time, you are wasting your time. It’s much better to say,
“This is the requirement. Let’s go fix it, let’s
go make that happen.” So, people did.
In the early days that was not quite the case. In the early days we
had people arguing, “Well, I would like to redesign your observatory,
and it’ll be so much cheaper my way.” That was kind of
confusing. People knew it would be expensive, and they were afraid,
for a good reason. What if you go do all this work and people say
no? But I guess my observation is stick to your plan, and be inspired
and inspirational about it, and people will eventually see that. So
they did.
Wright: The
engineering task—you are the project scientist, is there a project
engineer specifically?
Mather: Yes.
We have many project engineers, but my nearest counterpart is the
systems engineer. His office is right next door, Mike [Michael P.]
Menzel, and he is brilliant. He is just a wonderful guy to work with.
He understands what you say and why you mean it, and he knows how
to accomplish what you want. And he knows how to exert himself in
the team to make good things happen.
Of course there is not just one systems engineer, there are many.
Each subsystem has its own systems engineer and managers. A typical
engineering team has a manager and a systems engineer at the top,
then they make sure all of the people that report to them have done
the right technical work, and have gotten it done on budget and schedule.
That’s what that team does.
Our job as scientists is to hand our science job on to their manager
and the systems engineer, and say, “Please build this, please.”
When they fully understand the job, that’s what they do. We
like to participate in their discussions to make sure that they do
understand, but they understand. It’s very rare for a scientist
to come up and say, “Well, I have a better idea about how to
do your engineering job.” The engineers are really good at what
they do, and very rare for a scientist to have a better idea. Once
in a while, we have something to contribute, which is something nobody
has ever done before, and I’m happy when we can do that. But
mostly, it’s scientists cheering on the engineers and checking
that their work is what we meant.
Wright: And
then you have a project manager that oversees it all?
Mather: Yes.
Wright: And
relays to Center management, and then on to Headquarters as well?
Mather: Yes.
Bill Ochs, his office is down the hall. Have you interviewed him?
Wright: No,
I have not.
Mather: Then
you will want to do that sometime. Right now, he is really busy. In
fact, I think that’s true until we’re done, because he
is the focal point for everything about success. So I am cautious
about asking for his time. I don’t want to take his time. I
don’t need my hand held, I just want to know that he has got
what he needs.
Wright: And
you are going to be pretty busy. Are you shifting gears now as you
are getting closer to launch?
Mather: Yes.
As time has passed, I have accumulated so many young people to work
with me that I don’t do technical work much. Most of my work
now is communications, and making sure our general public is aware
of what we are doing and why we are doing it, and looking at big-picture
items.
But in general, what our scientists are doing now is getting ready
for flight. We have got the Space Telescope Science Institute in Baltimore,
and they are going to do the scientific operations there. The control
center is there too, with NASA participation. So we say, “Okay,
these are the commands we are going to send every day, and this is
okay or it’s not okay.” I think we have an intricate,
interlocking process between the NASA engineering team and the people
we hire up there. Their job at the institute is to make sure we know
where to point the telescope and why. So they are going to complete
a process of soliciting proposals, reviewing them, putting the selected
ones in order, and making the observations according to a schedule.
That’s a hugely difficult challenge. It’s a big team up
there that is required.
After we take the data, then it all has to be processed through the
computer and made available to the users and say, “Your star
is so-and-so bright at each wavelength,” or “Here is a
picture of the thing you wanted us to take the picture of for you.”
And it’s all been processed. Everything that we know of that
represents the instrument has been removed, so now you can say, “This
is in standard units, and you can trust what we are giving you.”
So we do that part.
Those scientists that are chosen, by the way, are all around the world—American
ones, some European ones, some Canadian ones, and who knows where
else. That’s being figured out now. Me personally, I don’t
have any particular observing plan at the moment. I will have to write
a proposal too, if I want to observe something. I have some ideas,
but they are not particular radical or interesting, so I don’t
know what I am going to do at that point personally. I could say,
“I really want to do something else, because this is a big notch
on my accomplishment list, but I want to do something else.”
Or I could say, “Well, now I have a really great idea of something
I want to observe.” So I don’t know what I am going to
do.
Wright: If
you could tell me some more about the science team that you have been
working with. As you mentioned, they have been on since the early
2000s, and hung in there. I guess that’s a compliment to their
institutions, that they have allowed them to stay on this project.
Each one has their own separate duties, but yet you will all work
together to talk about things. Kind of give us an idea of what happens
with that.
Mather: Right.
We have got several categories of these people on the science team.
The first category are people who are in charge of an instrument,
who make sure that it is built and provided to NASA. Those are called
“principal investigators” or “team leaders.”
In the U.S. we pay them, because we are going to send them whatever
it takes to buy that hardware. And of course their institution is
happy for that.
In Europe, they have different arrangements. They have to pay their
own scientists. In Canada, ditto, they have their own fund sources.
But their fund sources have been happy with them. I think everybody
recognizes this particular project is really the big thing that we
would be proud to work on, so everybody has managed to come through
with their support.
That’s the first category, and those people all have guarantees
of some observing time. They have already been required to submit
their list of targets, so they get dibs on things that they say, because
they are guaranteed-time observers. And they get a total of about
4,000 hours altogether for all these teams.
There is a second category who are called “interdisciplinary
scientists.” Those are people who observe but do not build.
They also get a share of that several-thousand hours, so they also
have submitted their list of proposed targets to look at. There is
a third category of people that are more like me, who are in the job
because of a position in our agency. So I am working for the government,
for the taxpayer, to do the right thing. A lot of us are on the committee
for that reason, representing our countries or our agencies. So it
is the mix of these three categories. None of those ex-officio people
have any guaranteed time, so we all get to propose if we want to use
the telescope.
Wright: Do
you find that the people from each of those categories have different
inputs or feedback depending on what your conversations are about?
Mather: Yes.
Yes, we have very different perspectives, and useful.
Wright: Can
you give me an example of maybe a topic or two or something that you
all discussed that it was good that you have that type of a mix of
people? That it seemed like it advanced the discussion, maybe kept
from a failure? Just an example that you take different type of scientists
and how it works well. Because sometimes we think that we should all
have people that think like us to get anything advanced.
Mather: Okay.
So diversity of opinion, if not of appearance.
Wright: There
we go.
Mather: Clearly,
the instrument builders know the requirements of their hardware totally,
and they are responsible for delivering stuff. So if we want to know
“Is this going to work?” we ask those people. This is
hugely important. If you ask me “Is it going to work?”
I don’t know, because I don’t know the details.
On the other hand, I have the job of trying to make sure the entire
planet knows how good this is, and why they should be building it
and supporting it. We also have a little bit of difference of priority,
too. I work for the public, so my job is to make sure we get the best
science for the entire observatory, and the individuals that have
guaranteed time, they do not. They have a job to make the best of
what they have been given, but that’s not the same as the general
public. This is not exactly the same interest. It’s a little
bit different job.
It gets to be interesting when you say, “Well, what are we going
to look at first?” Is it the people who have guaranteed time?
The guaranteed-time observers get to keep their information private
for a year, but that’s not necessarily the best for the world,
particularly if there is some chance that the observatory doesn’t
last forever. Similarly, if you find something really exciting, we
all want to follow it up, so please don’t keep that secret forever.
Those are competing interests, so that’s a thing we currently
think about, is how do we resolve that? We have a process, and basically
it is a combination of ask politely if you will please let the data
out sooner, and also provide some motivations.
We also developed, based on input from advisory committee, something
we call the “early release science.” This is different
from what the guaranteed time investigators have. We said, “Okay,
we are going to ask our entire world community for ideas about what
we should look at first. This particular small amount of time—a
few hundred hours of observing time—is going to be whatever
you all say, and it will all become public right away. So it’s
great that you are going to volunteer to help us do this, but you
don’t get to keep it to yourself.”
That’s an answer to the question of “What if you see something
really cool but you keep it to yourself?” That’s part
of what we do. Again, our project is so large that I don’t do
any of this personally, though we have a process. So that’s
how it goes.
Wright: Interesting
statement that you made, that the answers or the discoveries belong
to the whole planet. Which is different from when SpaceX might launch.
They are doing it for their company, and for their customer, but science
belongs to those who have an interest in it.
Mather: Yes,
yes. There are different kinds of reasons for people to do stuff,
but a great nation does this. We give stuff away. We get credit for
it, but we give it away, and that demonstrates who we are. So we encourage
people to do that a lot, as much as we can.
But the people who don’t want to give it away right away have
their own reasons, and fair enough. Particularly, university professors
have graduate students, and the graduate students need a shot at getting
their degrees from analyzing this information, so they don’t
really want to be competing with the people who could write a paper
tomorrow about the same picture. And there are people who can do that.
When we published the COBE maps a long time ago, the next day there
were published papers about them. People had been waiting for that
to come out. They said, “We already know what we are going to
say if it’s this way, and we already know what it’s going
to say if it’s that way, so the whole paper is written. We just
have to plug in some numbers.” I know some people are like that.
No graduate student can compete with that, so the university people
have multiple reasons for wanting to do it their way. We’ll
see, it’ll be fun.
Wright: You
mentioned earlier about some of the lessons that you learned that
you applied from the Hubble, and of course one was the fact of trying
to get everything exact as possible, because you won’t be able
to repair anytime soon. How much of an impact was that? I read something
about how everything was measured twice, and I am not sure if that
was a statement that you made?
Mather: I
would have said something like that. The carpenters say, “Measure
twice, cut once,” and they are right. Here, our version of it
is if you really, really require something to work, you better measure
it twice independently. The error that was made on the Hubble mirror
was about a ruler. In the end, it was a ruler. It’s a reminder
that if your two tests use the same ruler, they could both be wrong
because the ruler is wrong. Do not use the same equipment to test
the thing as you use to build it with. So we made sure. The Webb Telescope
mirrors are measured differently when we accept them than they were
when they were built, and we have multiple ways to do that. That matters
a lot to us. Everything that matters, you’ve got to be sure.
It’s a general lesson for life. If you really care about something,
better make sure more than one way. You can’t just ask John
Mather if it’s okay. If you need an opinion, you have to at
least ask two people. If you need a measurement, you better have two
independent measurements. If it’s a theory or a calculation,
you better have two different people do it, with two different tools.
We found, for instance, that the tools were incorrect in places. You
buy a computer code that’s supposed to calculate how much stray
light is going to get in because starlight is bouncing off the mirror
in a certain way. We had several different computer codes we were
using, and they were all incorrect in some different ways, so we were
able to find it. When it matters, you have got to really push. And
who’d have known that one? You would never suspect that the
computer code is incorrect, now would you?
Wright: No,
no.
Mather: But
it can be, it can be. And of course, the reason that we would find
it and other people didn’t is we are applying it in extreme
circumstances. Our telescope is cold, it’s dark. Everything
is different about our computer simulations than it is about anybody
else’s. They are enormous, too. Our computer simulation of that
hardware has got millions of nodes in it. You know about nodes and
simulations?
Wright: Some,
but not as much as you.
Mather: Basically,
it says, “We are going to divide our physical object into little
bits.” The computer will know how they all connect to each other.
The more little bits you do, the better it could be, but then also
the harder it is to do it. So in our case, the hard part was things
like the joints between carbon fiber structures.
Carbon fiber is inhomogeneous material. It’s carbon cloth made
with glue. You come to a joint and you say, “Well, neither of
those is the mix that I’ve gotten in the joint. It’s a
different kind of glue.” So you have to make a computer model
of the joint of two pieces of carbon fiber stuffed together that goes
down to tiny, tiny pieces to account for the fact that it’s
a variation of properties from place to place.
That’s one heck of a hard thing to do, but we had to master
that one, because that big structure is enormously complicated, and
we didn’t want it to twist, bend, warp, and other things as
it cooled down. And even if it does, after it’s cold you don’t
want it to be changing. It has to just sit there. So a huge investment
of that was required by our engineering teams. A hundred years from
now you would say, “Well, why was that so hard?” But today,
it’s hard.
Wright: Yes,
it definitely is that. October 2018 is the current launch date. You
will be busy until then, I am sure.
Mather: Yes,
for sure.
Wright: We
talked about the user community and getting those things ready. I
think another term that you have used—whereas Curiosity [Mars
rover] might have had the “seven minutes of terror,” you
said you were going to have six months of terror. Talk about that.
The launch is just the beginning of the wait.
Mather: Yes.
It’s actually not that bad, but the launch is the beginning
of a checkout period that lasts six months.
But ours is different from the Mars lander. The Mars lander had to
do it all by itself. There was no possibility of us noticing a problem
and fixing it on the way down, because the landing, seven minutes
long—it takes longer than that just to get the information back.
That little guy had to do it all by himself, all alone, so everything
that could have gone wrong they had to fix it in advance. That was
hard. It took imagination, it took skill, and it took a lot of computer
code checking.
We have a different plan. We do it step-by-step, under individual
commands, and at every step you say, “Did I get the right answer?”
And if I didn’t, this is what we know we have to do about that
problem. Our system is redundant. Every motor has two sets of electrical
winding on it, and two sets of controllers, so if it doesn’t
work we can switch to the other. And so we will be doing that.
Ideally, we don’t have to do any of that switching to the backup,
but there is always a backup. And we have time. We have got quite
a long time allowed for focusing the telescope. We have about 10 steps
that it takes, each of which has many sub-steps. When you first unfold
the telescope, it is nowhere near correct. The mirrors, none of them
are in place. We know about where they are supposed to be, but they
will have just survived a launch—a heck of a lot of vibration—so
nothing will be exactly right. A lot of time allocated to figure that
out, yes.
Wright: And
you get to enjoy the moment.
Mather: Yes.
But as long as it’s not dead, we have just routine engineering.
That’s the plan. That is something we are ready to do. We already
have rehearsed all this stuff. A year and a half before launch, we
know how to focus that telescope. We have got simulations, we have
got the scripts written.
A vast amount of stuff has already been done so that you would get
pretty close to the right answer if you did what we have ready today.
We will have plenty of opportunity to check. The tests that we are
about to do at Johnson, many of those scripts will be tested. We’ll
focus the mirror in the big vacuum tank there to make sure that works.
Wright: From
what I understand, it’s going to be behind closed doors for
about 100 days. Is that correct?
Mather: Yes.
It will be in the vacuum [JSC Chamber A], so really closed door, yes.
Wright: It’s
amazing that it’s going in there. Talk some about why it went
to Johnson, and the work that Johnson had to do to prepare for it.
Mather: Okay.
Well, I only know the top-level version of it. A long time ago, when
we were just beginning this project, we thought, “Oh, they have
a really gigantic vacuum tank at [NASA] Glenn [Research Center], out
in [Cleveland] Ohio.” So people imagined that was the place
to go. Then when we got down to it, it turned out no, that’s
not a good place after all. Without going into details, it would have
been really hard and expensive to do that. So what are we going to
do? Well, we had better look around. Then we found Johnson is the
place to go. They have a big tank, it’s leftover from Apollo.
It’s a good tank. It hasn’t been much used, but it’s
available.
But then, what we had to do was pretty big. We had to really scrub
it and clean it, because we are going to put something in there that
cannot be contaminated. And we had to cool it down. This is something
they didn’t need for Apollo. We have to make sure it is going
to be so cold that it’ll be simulating the temperature the telescope
will have out there. We had to buy a gigantic refrigerator. It’s
huge. I have seen the warm part. It’s a gigantic red thing,
about 30 or 40 feet long, and uses a megawatt of power to cool the
insides. There is also just the precooling, and we take truckloads
of liquid nitrogen to cool the outer shells.
So this is a huge engineering project just to get it cold. We spent
quite a long time getting it ready, cleaning it up, making sure there
are no sandwich bags at the bottom, all that stuff that could have
been in there. It’s hard, but they did it, and so it’s
ready.
Then, after we got the chamber pretty much ready, we have to get the
equipment ready. You can’t say just, “Well, I’ll
push the telescope in and make it go.” You have to test the
test equipment. So build all the test equipment, build a simulator
for the telescope so you can test the test equipment on a simulated
telescope. We did all that. It takes a long time to do this, but you
can’t avoid it, you have to do it.
Wright: And
they modified the building itself to build an actual cleanroom to
protect it, correct?
Mather: Right.
Since the telescope has to be clean, you open it up in a cleanroom,
which we didn’t have. So around the front door of the tank,
we put a room with clean air, and the process for doing that.
Wright: Have
you been down there? I know that you are going in a couple of weeks,
but have you been down there already to see it so far?
Mather: Just
briefly, I was there last week. We had a visitor, [Congressmen] Brian
Babin and Randy [Randall K.] Weber, whose districts are right there
[in southeast Texas]. So yes, I went for that, and I have a phone
message about “Brian Babin just came in this afternoon,”
so he is interested. He is clearly very interested in the Webb Telescope,
and in NASA. Glad to see that.
Wright: Yes.
It’s always good to have a champion somewhere on the [Capitol]
Hill to keep it going. When you were developing those new technologies,
we always use the expression that they are “state of the art.”
As you have been working on these, do you feel like the technologies
that you developed and designed so many years ago are still the ones
that you need to go where you are going?
Mather: Yes.
Wright: Or
you always have that fear that something better is going to come along?
Mather: That
happens in other parts of the world, where there’s other customers—if
you want a better computer processor, somebody else always wants one
too. So if you wait, one is going to happen. In our world, there is
nobody else that wants what we want, so there is no other, better
one anywhere.
There is one small area where better ones—or bigger ones anyway—are
coming along, and that’s the infrared detectors. There is a
new generation currently in progress that’s got a larger format,
more megapixels. I don’t think it’s more sensitive, so
we really didn’t have to have it. We are getting just about
as good a result as we would have gotten if we could have had the
new generation. That’s the only one that I am aware of where
progress has continued after we stopped developing.
Wright: Is
the project living up to the expectations you wanted it to when you
started 20 years ago?
Mather: Yes,
it really is. I had no idea how hard this was going to be, nobody
really did. But it was really clear to me that this was the next great
opportunity, so it didn’t take me five seconds to say yes, that’s
what I wanted to work on when I was given the shot. So yes, it’s
what I want. And I am sure we’re going to discover something
amazing. I just don’t know what it is yet.
Wright: Well,
that was going to be my next question. A lot of people have lots of
speculation of what you feel like it’s going to be able to do,
but I was curious if there is something you really want it to do?
Mather: Yes.
I’ll just give a few guesses about things that might turn up
that we can be sure about. I think we are going to get a big surprise
about the first galaxies. They are unknown to us, so we don’t
know how the black holes were formed that are in the middles of big
galaxies. We don’t even know how the little ones are formed
that are making gravitational wave bursts, now that we have seen three
of those.
Black holes are still a very big mystery. I have a speculation that
there is some form of small early galaxy that came and went, that
was formed and then disappeared. That would be an explanation for
why don’t we have any of them locally. So I am guessing there
is something like that out there.
We have something we didn’t expect in the beginning, the ability
to study planets around other stars, exoplanets. I think we are going
to get a surprise there, because everything we know has been a surprise.
The laws of physics are not changing, but what nature does with them
is always a surprise. I have a friend who works on the exoplanet subject,
and he says we are still waiting for our first successful prediction
of anything about the exoplanets. Everything has been a surprise.
That’s cool. So whatever we find, it’ll be a surprise
as well.
We could get a surprise about something ordinary—quote, “ordinary”—like
how do stars form? They do it in places that are invisible, mostly.
You have seen our beautiful pictures like the Eagle Nebula [Messier
16/NGC 6611], and they are beautiful to look at because you can’t
see inside. So gosh, what is going on inside? Well, I’d like
to know. We’ll make some progress with that.
Here is a weird example. What if there is a little class of sub-stars
that are just zooming around and filling the local space, and the
nearest-by luminous object is nothing like that we ever saw? What
if little Jupiter-sized objects are zooming around in-between the
stars? Well, I think it’s pretty likely. We are beginning to
think so, have found a few. There is even a star out there that I
call a “chilly star.” It’s called a brown dwarf
star. It’s a real star, but it’s cooler than the Earth.
Wright: What
kind of impact do you want the Webb Telescope to have across the planet?
What are you hoping it’ll do?
Mather: Well,
I would like us to be able to put newer, better pictures in every
astronomy book, and new stories about how things actually happened.
Our general idea is we wanted to look at everything from the Big Bang
until now, so the first things we hope to see are the first stars
and galaxies, and black holes. We want to see the galaxies grow.
We have a story, we’ve got lots of computer movies we didn’t
used to have, and they might be right, but they might not. We would
like to see the stars and planets being formed, and we are just getting
a hint about how to do that. And we’d like to see everything
about planets and how they form, and might even turn into life-supporting
planets. We don’t think we are going to see life on another
planet with Webb, but we are certainly going to look and see what
we can see. It is thought maybe we could see an Earth-like planet
having enough water to have an ocean. On the other hand, we have no
idea whether to expect that or not. We know of a few really nearby
planets. The nearest star has a planet, Proxima Centauri. You saw
that, right? There is no particular reason to think it’s like
home.
I think the most likely thing about planets around weird stars like
that one is that they are round rocks, that they are just lifeless,
bare things. Some of them will have atmosphere, and some of them will
be roasting hot. I don’t think very many of them will be home.
But we have got to go look.
The idea that you could find the TRAPPIST-1 system of planets starting
with a telescope on the ground? That telescope on the ground is about
this big [demonstrates], and they were able to detect three of the
planets with that. So then they said, “Well NASA, can we follow
that up with your telescopes in space?” And yes, now we have
got seven [planets]. So we have got surprises coming to us from a
lot of directions.
Did you see a week or two ago, they announced they had found a disappearing
star?
Wright: Yes.
Mather: You
saw that? I think I believe it, and what can happen to a star that
disappears? Well, it could turn into a black hole. That’s the
first explanation, and a popular explanation. Now, what else? I don’t
know. Maybe it surrounded itself by dust, but I think they have already
argued against that.
Wright: You’ll
have the next 20 years to enjoy all of the discoveries at the lab.
So what are you planning to do after this? Not that you’ll have
a lot of free time.
Mather: Well,
a lot of possibilities. I could go retire and go to the beach, but
I don’t think I will. I could write a book, I have some ideas.
I could observe with the Webb, I could actually help contribute to
the new observatories we are working on. NASA is currently studying
four really wonderful observatories to be considered in our next decadal
survey. Have you seen that stuff? Paul [L.] Hertz, our Director of
Astrophysics at Headquarters, has decided to study them all so we
can have a serious discussion in 2020.
They have also selected a handful—maybe six or eight—probe-class
missions for further study. These are smaller, and you could do more
of them. All those are interesting, too. I got to contribute to at
least two of them to help them win, so I was really pleased with that.
Those are smaller, quicker, could be opening up new territories.
One that I am fond of is an X-ray telescope that would look for X-ray
flashes. If you could see the X-ray flash associated with some event
like a gravitational wave burst, that would be exciting. Then we would
be able to say, “Okay, point the Webb Telescope”—or
any other one—“over there today, see what that was.”
There is so much happening. It’s all I can do just to keep up
with the popular stories about it, much less contribute to details.
I could imagine going back to school. I wish I was a graduate student
again. The kids coming out of school now, they know so much.
Wright: When
we came in, you had a group of them. You were commenting about projects,
that these are never done alone.
Mather: Yes.
Wright: Talk
about what it takes to put one of these projects together, the dedicated
people, and all these new—as you mentioned—young people
that are excited about learning more, and giving their time for discovery.
Mather: Yes.
I think everything we know is community knowledge, in a way. I came
across a really lovely article that pointed this out. People are concerned
these days that kids don’t know how to understand evidence.
And nobody really understands I think in a way because we are all
embedded in a context. I know to trust science because I have grown
up with science.
Other people know to trust their neighbors. You know, where do I get
the best tamales? Well, ask your friends. They say, “Oh, go
over there. That’s where they are.” We all have our local
sources of knowledge, and that’s a universal phenomenon of people.
Our science community is a real community, but it’s spread out,
it’s dispersed. We have got scientists in every country. If
you want to know about your neighborhood, then your neighbors are
your community.
That’s sort of an aside from this, but it’s a reminder
that we can only do what we do because of the communities, and our
communities in technical and scientific things are dispersed. If I
need a widget that does something, I may have to go to Kansas or California,
or Azerbaijan, or wherever they are made to find them. That is changing
quickly, since everything can be posted on the internet in a flash,
for good or bad. The community is also spread out in a way that it
never was, so when you say, “Well, I think I am going to invent
something,” the first thing you do is Google it, see if it already
has been invented. Then you say, “Oh, oh no! Somebody already
invented it. I’ll just buy it. So now what? What am I going
to invent next? What else do we need that we don’t have? Oh!
I’ll Google that.” Or, “Nobody has invented that.
Who could help me?”
This has changed so dramatically, and our technology has changed so
dramatically, too. When I came here to Goddard in 1976, people were
designing satellites with pieces of paper and sharp pencils. And they
were good at it, but it was really hard work, because it’s really
hard to visualize something that you’ve drawn on a piece of
paper. If you have gotten a computer design now, say, you can rotate
it in front of your eyes, you can put the other parts together with
it, you can see when that will all fit together. In the old days with
pieces of paper, you had to build it and see if it would fit, because
you couldn’t visualize well enough.
So things have changed a lot. We can 3D-print parts you could never
possibly construct before. So whatever it is you want, we’ll
be able to do it. I heard the futurist of Microsoft [Corporation]
talk once and he said, “If you can imagine it, you can build
it.” It’s almost true. If you want a thing that’s
made out of an atom here, and another atom over there, and another
atom over there, if you really want it, we can make it.
I used to think, well, travel to the stars is obviously impossible,
but one of the members of the COBE team was behind this idea of the
Starshot. Have you read about the Starshot?
Wright: No.
Mather: It’s
now being supported by private funding, as well as some by NASA. It’s
called the Breakthrough Starshot. The idea is you get one heck of
a powerful laser, and one heck of a tiny projectile that’s large
and very lightweight. It reflects the light back, and you can push
on the thing with the laser beam, and can accelerate it to a quarter
of the speed of light. It’ll get to Alpha Centauri in 20 years.
It is impossible today, but it’s not impossible under the laws
of physics. This is my sort of motto for things. If it’s not
impossible according to the laws of nature, try it and see how far
you can get. They are trying it, they are already working on it. And
you can draw what you need. You can’t build it yet, but you
can draw what you need. I like to think about things that are really
hard and really impossibly difficult, then see if we can do them.
Wright: When
you were talking about the design of the technology, and then you
were also talking about how the engineering and mechanical and the
scientific all came together—when someone looks at the model
of the JWST, it’s very simple, but yet it’s very complex.
It also, when it’s standing there with its glorious mirrors,
reminds you so much of artwork. I’ve actually thought of it
being more of a “space-art-craft” instead just a spacecraft.
Mather: Yes,
yes.
Wright: Then
you shared with us too, when we were walking in about how Goddard
had an art day. I’m looking at an art piece on your desk, but
explain how this other piece of artwork came out to typify what’s
going on with the telescope. It’s very unique.
Mather: We
had an art day that was dreamed up by one of our public outreach specialists,
Maggie [Margaret E.] Masetti here. She organized a day for artists
to come and look at the telescope through the cleanroom window. They
looked, and they thought, and they imagined, and they drew. They wrote
poetry, they painted pictures, they sketched, composed music—all
kinds of things happened.
And one artist had the idea to make this thing that I’m showing
you [demonstrates]. This is the small version. She has got a Webb
Telescope mirror made out of 18 hexagons, as the real one is, but
she’s got arms coming out of it. These are arms. One of them,
this is me. This is Amber, I guess, I think that’s Amber [N.]
Straughn [Deputy Project Scientist]. I think this is the artist herself
[Ashley Zelinskie].
She does a 3D scan with a portable scanner of each person’s
arm. I just held my arm here like this, and she walked around my arm.
So there it is. It’s made small, and coated with real gold.
Made with a 3D printer. All of this is something you could only have
dreamed of a long time ago, and it’s done by high-tech [technology]
stuff now. If you said, “Michelangelo [Renaissance artist],
can you make me one of those?” he could have, but it would have
been a very different process.
Wright: I
have my ideas of what the artist is saying, but what do you think
she is saying with that? Or what did she tell you she was saying?
Mather: She
didn’t say what the story was that it was telling, really. I
have a piece of paper that summarizes how it was made, but it doesn’t
tell the meaning of it. So the meaning is whatever you give it, I
think.
Wright: I
think so, too. I think that’s how art works, right?
Mather: Yes.
Wright: It’s
an amazing spacecraft, because that’s what it is. When you take
all the different types of arts and sciences and put them together
to create, and of course to bring back what it’s going to bring
back.
Mather: Well,
I like to say that science is about imagination. A scientist has to
imagine something that you cannot see—and not only imagine it,
but imagine it so well that he or she can say, “And this is
how it works.” And if you are really good at it, then you can
say, “And here are the equations that describe it.” Or
maybe you can’t do that. Maybe you just have a story, but you
say, “This is how chromosomes work.” Well, I’ve
never seen a chromosome, but I know how they work, sort of. I have
seen movies about how the chromosomes are ripped apart by little chromosome
readers in the cell, and then put back together after they’ve
been copied. We have got a movie of the Xerox machine that runs inside
the cell to duplicate a chromosome. Isn’t that astonishing?
It was all done by imagination.
Similarly, our engineers have to say, “Well, we are going to
build you something that was never built before. We are going to imagine
it, and then we are going to build it.” They have to imagine
also all the ways that it might go wrong, and make sure that doesn’t
happen.
People might think differently about us. I think science has been
so badly taught in so many places that people don’t know what
we do. Some people think, “Well, scientists are just those people
who know so much.” My picture of it is we are the people who
know the least. We are always thinking about the stuff we don’t
know. That’s kind of the opposite. We are up against the crossword
puzzle of the universe. How does it all fit together, and can you
tell a story about how it really works? If you understand that, can
you build something you want that’s based on that?
That’s what we have been up to while the public thinks we are
trying to make them learn things they didn’t want to learn in
physics class. We are imagining the world as you cannot see it, and
trying to make it go the way we want it. And that’s really different.
Wright: It
is, it is that. Is there anything else you have you thought about
you’d like to add about your project?
Mather: I
always think it’s important to mention that individual human
beings made this project possible. Our first team leader was Bernie
Seery, and I guess you know his name right, already?
Wright: Yes,
yes.
Mather: Then
we had Phil [Phillip A.] Sabelhaus, and he died a few months ago.
He had retired from NASA. He was having a good time, he was going
to go off on travel to go to the beach and see all kinds of things,
and he died unexpectedly soon. Our current Project Manager is Bill
Ochs, who you know, right?
Wright: I
know the name, yes.
Mather: Yes.
And he was just down the hall, you’ll want to talk with him.
I think, personally, that the art of project management is much more
difficult than the art of science. You can get a building full of
managers, but only a few of them would be able to pull off a project
like this one. We have one, so I am totally, eternally appreciative
of what they do. I don’t think I could possibly do it myself.
I could study and learn, but I’d never be able to do what they
do. That’s how I feel. It’s like when you watch Roger
Federer [tennis player] play, you say, “Well, I’m glad
he can do that, but I can’t do that.”
I am so appreciative of the individual human beings, and also of the
system that we have built up here at NASA. We struggled at the beginning,
we couldn’t beat the Russians at first, and now what we can
do is astonishing. I am just amazed at what we can do. The people
who built the system that we are in, we don’t even know who
they were. But they built the system.
There was a system that enabled NASA to recruit me. There was a system
that enabled NASA to recruit all these people that enabled us to build
the hardware, and the buildings, and the tanks. And Congress, who
has decided they like us and they will send us money. And there was
the Soviet Union that forced us to do it. Yes. Who’d have thought
that was an important part of our system? But it was. They scared
the bejesus out of us here in this country. I don’t know where
you were in 1957, but people were scared.
Even before that, they were teaching little kids to hide under the
desk in case of nuclear attack. Even when you are seven years old,
you think that’s stupid. But it just showed us we were scared.
And then they said, “Oh, that must be the fault of the educational
system that we didn’t do the right thing over there, so we are
going to make sure the kids have science.” So I got to benefit
from that. The country spent a fortune, but look what we got.
Wright: And
are still getting. Or you wouldn’t be having those students
in your office ready to work hour after hour, and write word after
word.
Mather: Yes.
We are discovering stuff, and this is an extraordinary part of history.
We are right at the beginning of a new era, and I don’t know
what it’s going to be. It’s going to be different. But
it’s very exciting.
Wright: It
is exciting, and I wish you the best of luck in the next 18 months,
and then after that.
Mather: Well,
thank you. We are working hard to earn our luck. We do the test, we
argue and think. And that’s what it takes to earn our luck.
Wright: Well,
we look forward to exciting times and reading more things.
Mather: Okay,
Rebecca. Thank you for coming to ask.
Wright: Oh,
thank you so much for your time. We appreciate it.
[End
of interview]
Return to JSC Oral History Website