That’s an old trick for getting projects selected and landing contracts. In the formative stages, when the details are still fuzzy (note the 7, not 18 segment mirror in the image) people talk about how easy it will be using proven, heritage technology. Then, by the time they get to the preliminary or critical design review, it turns out that the heritage technology isn’t sufficient for the newer, more detailed and refined requirements. So they’ll have to do something new. In practice, either that original assumption about heritage was wildly optimistic, or they assumed there wouldn’t be any feature creep in phase A and B (also wildly optimistic given how many projects are managed.)
Yep. It’s used as a cover to explain how they can bid so low and pass the laugh text. The lowest credible bid wins as pointed out in the hearings. Once they win, the funds are locked up, their competitors are out of the way, and then they can milk it by cost overruns ng for a decade or so past the bid completion. This has been going on for a long time in the space industry. Now even the small business are doing it too if they have a key unique technology. Profit is computed on the total cost. There isn’t much incentive to reduce costs or get things right the first time if they can get away with dragging things on as long as they can get away with it.
One of the biggest costs due to these huge overruns is what’s known has an opportunity cost. The cost overruns tie up funds that could have produced more. The cost overruns are also a mechanism to starve out your competitors and grab more money. Look at the impact the cost overruns have had on many NASA projects.
I think the quote is, “If you’re not part of the solution, there’s good money to be made prolonging the problem.” But the right way to motivates people to do good work for a low and realistic cost is something I struggle with.
It would be incredibly unrealistic to expect a for-profit company to _not_ be motivated by making money. Whatever system or process we set up, they will find ways to maximize profits. And, as far as I am concerned, that’s just them doing their job.
Scientists are often willing to sacrifice money for opportunities to make discoveries (and the career prospects doing so provides.) But they also have to pay the bills and put their kids through college. And that combination sometimes results is minor or dubious results getting advertised as Earth-shaking.
I guess I think the current system prompted personal or institutional gain over what is best for the project or to achieve a goal. Changing that requires changing the system to reward and motivate people differently. I don’t know what that change would be.
There is no perfect system, but there are better systems. The Commercial Cargo (COTS) project had some definite advantages. For a fixed price of $500M NASA got 2 new LV and two new spacecraft. The potential reward for the companies involved was access to a broader market. For Orbital ATK it didn’t work out so well, though they still did OK out of CRS missions. But for SpaceX COTS proved to be a gold mine. They were able to parlay their COTS investment into a $12B launch manifest for F9. That’s the way govt contracting should work. Encourage the contractor to ‘get it done’ so they can get return on the investment from the commercial sector.
You know this reminds me a little of my own experience- perhaps peripheral, you decide.
I am a Landscape Architect- MLA from LSU. I’ve had many roles in nearly 40 years’ practice, many of which are unfamiliar to folks who don’t know about my profession- from site planning to planting design to representing developers trying to get permits from state, county, or federal agencies; street design, wildlife corridors. Long list.
These things all have one thing in common- rubbing shoulders with government employees. And if there’s one thing they all carp about, it’s this: “Come to us early! Let us be part of the process!”
And you know what? We do. We treat them with respect but in reality it more risible than anything. why? Because by and large these folks are under-qualified and armed only with a set of regulations. We play ball with them in order to serve clients, but truthfully they bring very little to the table. (Sometimes they put limits on outrageous clients, it must be said).
Is it universally true? Of course not. There have been many cases where these guys are helpful. But it is always in the realm of getting through the system, and nothing whatsoever to do with the creativity driving the work.
Bad attitude? Maybe. I’ve been a liberal my entire life, believing in the power of government to make real change. I still do. But often as not we are talking obstacles, not a hand up.
Now take this same folks and ask them to be design leaders and what do you think you will get? Obfuscation ,delay, constant meetings, poor decisions.
That’s what’s wrong with projects like Webb.
Do I have a solution? Sure. For acquisition, write reasonable performance specfications, and do it with the help of industry: “give us a bird that sees XX nanometers and lasts 10 years”. Naturally that’s too brief but the point is still spot on.
The problem is, I’m not all that sure that the whole fixed-price contracting model would have worked for Webb. Where is the profitable larger picture for the contractor? Space Act Agreement contracts are actually spin-off programs. What does James Web Space Telescope tech spin-off into?
I can’t see the spin-off benefits, but that means you couldn’t use firm fixed price contracts for a telescope. In fact, that’s often how ground-based telescopes are built. The contractor estimates the cost, tacks on however much profit he wants and thinks he can get away with, and builds it. If his estimate was off in the wrong direction, the difference comes out of his profits. (Yes, I know, that isn’t exactly how it works, and construction contracts are infamous for cost overruns. But the legal framework is viable, and very different from a cost plus government contract.)
Earth-based telescopes are not one-of-a-kind. At least most of the costs aren’t buying one-of-a-kind things. I’m saying that there is no ongoing, repeat, commercial market for standalone, ginormous, Lagrange-based, infrared observatories. Most of the tooling, training, etc. will be used just this once. There is nothing for a business to grow on.
There are many aspects of a space telescope which have other applications. Solar arrays, attitude control systems, telecommunication, software to do really good, lossless data compression, etc. And it’s no secret that most of the mirrors in orbit point down not up. That’s why scientific spacecraft have to deal with ITAR and other export control laws. Knowing how to build a good space telescope gives you quite a bit of insight into building a good spy satellite.
But regardless of further applications, I don’t see why fixed cost contracts would not work for truely unique spacecraft. The contractor might tack on more profit, since spin offs wouldn’t be an extra bonus. Technically risk for doing something novel might also drive the price up (and make some people amazingly rich, if it really is as easy as it looked.)
For an initial estimate of $500 million, a sane contractor might put in a $2.5 billion bid. But we wouldn’t end up in the cost plus world where $500 million has turned into $8.8 billion and then hearing they really need another $800 million.
And importantly, if you could end the cost-creep (or cost-surge in the case of JWST), you could safely fund many more missions. That, at least to some degree, solves Bill’s “one-of-a-kind” objection.
And if you could stop the schedule overruns, you might have more researchers willing to let novel tech/instruments/science slip to the next mission, or the next. Increasing the rate any given type of mission (telescope/probe/lander/etc) reduces the sense that you’ll only get one mission in your working lifetime.
And you’d better believe that on a more fixed cost contract, the contractor is going to fight “requirements creep” a lot harder than they would on a cost plus contract. Requirements creep is a huge problem for any sort of contract whose goal is to build something.
For example, when you have a contract to build a house, the contractor isn’t going to let you swap out laminate flooring in the original contract for hardwood just because you as the customer didn’t think through all the downsides to laminate flooring before you signed on the dotted line.
Seriously, fixed price contracting is suited only for bidding on a statement of work that is based on knowing what a similar or nearly-similar article cost to build. JWST has many new technologies and some heritage hardware. But, in cost estimating for space programs, what really counts are (1) the scale of the end product compared to what has been built before; (2) the complexity off the systems integration and test; and (3) the skill of the engineering and technician workforce the contractors can bring to the job.
JWST is many steps up in the complexity scale compared to other space telescopes: it is not only big, it has to work nearly to perfection. Lots of starting unknowns. And, as can be read about over the past several years has very little tolerance for errors in design or manufacturing or testing. The systems engineering and integration challenges have been huge, largely due to the thermal control challenges.
The third means effectively how the profit/loss/reputational potential of the JWST development stacks up against the other development and production jobs N-G has in work. The programs are competing against each other for the best talent. Management’s decisions on where to place the best talent involve a lot of P/L/R considerations, and then added to those are the security credentials of the workforce. And, N-G has a lot of national security programs….
BTW, the US Navy had many overrun and delays when it used fixed price incentive contracting for new design ships. In FPI, every change order comes at a price in schedule and cost, AKA “get-well-opportunities.” And those change orders can stem from the mundane (departmental financial issues (e.g, continuing resolutions causing spoon-feeding contract funding), to the complex because the technology didn’t work as expected, to the inadequately specified in the SOW.
That’s the thing about Space Act Agreement contracts. The whole system is structured to work best for aggressively innovative companies with a larger goal beyond NASA…companies like SpaceX. It would be a ton of fun to watch Jeff Bezos playing that game some day 🙂
Or they know that it will require something new that is coming available inside their timeline…but that they’ll need to develop and refine to their needs. And they know it could cause schedule creep, so “shhhhh”.
I’m pretty sure Keith is down on (in this case) NG who seem to have profited from the continual budget increases, and who continually promise “we’ve got it right this time, trust us; another billion will see us through, trust us.”
Keith, I think, is consistently down on NASA’s current way of doing business. JWST’s problems are current and high-profile, and really too hard to pass up as an example.
That’s probably the biggest impact- the huge opportunity cost. It also gives N-G a competitive edge to suck up the funds to starve out the competition. N-G is laughing all the way to the bank on the backs Of the taxpayers.
As a civil servant, I’m retired now, the difference in the commercial world business practice and the aerospace industry is staggering. For example, a commercial contractor was putting in a clean room floor in one of my labs. Their subcontractor used the wrong materials and didn’t follow procedures. The contractor did even hit us up for a re do- he apologized and immediately offered to re do it. In the aerospace business practice, with the tax payers footing the bill, we would have had to keep paying the contractor to re do it with profit over and over until it’s done right. It’s actually very profitable to screw up.
JWST is very exciting and we hope finally it does really launch and work. However, there were many other exciting stuff impacted by the cost overruns. In fact N-G is stifling these other program and benefitting by eliminating competition too. That’s why the aeropace industry milks the big flagship efforts for all they can get. The JWST hearings expose just the tip of the he iceberg of whe aeropace incompetence, complacency, and greed. Most of All I fault NASA and congress. They let these companies do this. We all deserve better than this with our hard earned tax payers money.
“The … telescope can point to any target over more than half the sky.”
Not any more. Webb can only point to a circular swath of sky perpendicular to the sun. Like an Earth-based telescope that can only look at the sky just above the horizon; you have to wait for the sky to rotate into that band. Except that only occurs twice per year. How great for repeat and stacked observations when the telescope’s major working life is limited to 4yrs. All for a bargain price of $9 billion… and rising.
Or targets of opportunity. When something unexpected happens, like the Shoemaker-Levy 9 impacts on Jupiter or a nearby supernova, many observatories delay planned observations to look at it. JWST can’t do that, unless the even happened to be in the right part of the sky.
Well, this is basically true of every space telescope built in decades: Spitzer, WISE, etc. It’s not just the the power requirement, it’s the need to stay behind the sunshield. Typically, that permitted zone is several tens of degrees wide. You’re also forgetting that HST has similar limitations, and that being in LEO half the sky is unavailable to it at any given time. That’s why half the observing time is thrown away. The real horror is the slew time. Of course, HST hides this during the earth eclipse – the typical slew rates on these telescopes are about the same as the minute hand on a clock.
HST isn’t nearly that bad. Well, the Earth is in the way and the slew times are long. But it can point as close as 45 degrees of the Sun. They generally don’t, but as close as 90 deg. isn’t uncommon. And there is a “Continuous Viewing Zone”, roughly perpendicular to the plane of HST’s orbit, were you don’t have the Earth in the way for half an orbit. You’re thinking of infrared telescopes which are really pushing low temperatures.
The problem is that, even if one constraint like this isn’t too restrictive, it can add up. Once you start adding more constraints (solar arrays to Sun, unobstructed field of view for star trackers, etc.) you can paint yourself into a corner. Also, there are fewer clever tricks the observers can come up with. (The discovery of some of Pluto’s moons involved rotating HST to put the diffraction spikes in different places; JWST can’t do that due to the thermal constraints.)
That’s an old trick for getting projects selected and landing contracts. In the formative stages, when the details are still fuzzy (note the 7, not 18 segment mirror in the image) people talk about how easy it will be using proven, heritage technology. Then, by the time they get to the preliminary or critical design review, it turns out that the heritage technology isn’t sufficient for the newer, more detailed and refined requirements. So they’ll have to do something new. In practice, either that original assumption about heritage was wildly optimistic, or they assumed there wouldn’t be any feature creep in phase A and B (also wildly optimistic given how many projects are managed.)
Yep. It’s used as a cover to explain how they can bid so low and pass the laugh text. The lowest credible bid wins as pointed out in the hearings. Once they win, the funds are locked up, their competitors are out of the way, and then they can milk it by cost overruns ng for a decade or so past the bid completion. This has been going on for a long time in the space industry. Now even the small business are doing it too if they have a key unique technology. Profit is computed on the total cost. There isn’t much incentive to reduce costs or get things right the first time if they can get away with dragging things on as long as they can get away with it.
One of the biggest costs due to these huge overruns is what’s known has an opportunity cost. The cost overruns tie up funds that could have produced more. The cost overruns are also a mechanism to starve out your competitors and grab more money. Look at the impact the cost overruns have had on many NASA projects.
I think the quote is, “If you’re not part of the solution, there’s good money to be made prolonging the problem.” But the right way to motivates people to do good work for a low and realistic cost is something I struggle with.
It would be incredibly unrealistic to expect a for-profit company to _not_ be motivated by making money. Whatever system or process we set up, they will find ways to maximize profits. And, as far as I am concerned, that’s just them doing their job.
Scientists are often willing to sacrifice money for opportunities to make discoveries (and the career prospects doing so provides.) But they also have to pay the bills and put their kids through college. And that combination sometimes results is minor or dubious results getting advertised as Earth-shaking.
I guess I think the current system prompted personal or institutional gain over what is best for the project or to achieve a goal. Changing that requires changing the system to reward and motivate people differently. I don’t know what that change would be.
There is no perfect system, but there are better systems. The Commercial Cargo (COTS) project had some definite advantages. For a fixed price of $500M NASA got 2 new LV and two new spacecraft. The potential reward for the companies involved was access to a broader market. For Orbital ATK it didn’t work out so well, though they still did OK out of CRS missions. But for SpaceX COTS proved to be a gold mine. They were able to parlay their COTS investment into a $12B launch manifest for F9. That’s the way govt contracting should work. Encourage the contractor to ‘get it done’ so they can get return on the investment from the commercial sector.
You know this reminds me a little of my own experience- perhaps peripheral, you decide.
I am a Landscape Architect- MLA from LSU. I’ve had many roles in nearly 40 years’ practice, many of which are unfamiliar to folks who don’t know about my profession- from site planning to planting design to representing developers trying to get permits from state, county, or federal agencies; street design, wildlife corridors. Long list.
These things all have one thing in common- rubbing shoulders with government employees. And if there’s one thing they all carp about, it’s this: “Come to us early! Let us be part of the process!”
And you know what? We do. We treat them with respect but in reality it more risible than anything. why? Because by and large these folks are under-qualified and armed only with a set of regulations. We play ball with them in order to serve clients, but truthfully they bring very little to the table. (Sometimes they put limits on outrageous clients, it must be said).
Is it universally true? Of course not. There have been many cases where these guys are helpful. But it is always in the realm of getting through the system, and nothing whatsoever to do with the creativity driving the work.
Bad attitude? Maybe. I’ve been a liberal my entire life, believing in the power of government to make real change. I still do. But often as not we are talking obstacles, not a hand up.
Now take this same folks and ask them to be design leaders and what do you think you will get? Obfuscation ,delay, constant meetings, poor decisions.
That’s what’s wrong with projects like Webb.
Do I have a solution? Sure. For acquisition, write reasonable performance specfications, and do it with the help of industry: “give us a bird that sees XX nanometers and lasts 10 years”. Naturally that’s too brief but the point is still spot on.
The problem is, I’m not all that sure that the whole fixed-price contracting model would have worked for Webb. Where is the profitable larger picture for the contractor? Space Act Agreement contracts are actually spin-off programs. What does James Web Space Telescope tech spin-off into?
I can’t see the spin-off benefits, but that means you couldn’t use firm fixed price contracts for a telescope. In fact, that’s often how ground-based telescopes are built. The contractor estimates the cost, tacks on however much profit he wants and thinks he can get away with, and builds it. If his estimate was off in the wrong direction, the difference comes out of his profits. (Yes, I know, that isn’t exactly how it works, and construction contracts are infamous for cost overruns. But the legal framework is viable, and very different from a cost plus government contract.)
Earth-based telescopes are not one-of-a-kind. At least most of the costs aren’t buying one-of-a-kind things.
I’m saying that there is no ongoing, repeat, commercial market for standalone, ginormous, Lagrange-based, infrared observatories. Most of the tooling, training, etc. will be used just this once. There is nothing for a business to grow on.
There are many aspects of a space telescope which have other applications. Solar arrays, attitude control systems, telecommunication, software to do really good, lossless data compression, etc. And it’s no secret that most of the mirrors in orbit point down not up. That’s why scientific spacecraft have to deal with ITAR and other export control laws. Knowing how to build a good space telescope gives you quite a bit of insight into building a good spy satellite.
But regardless of further applications, I don’t see why fixed cost contracts would not work for truely unique spacecraft. The contractor might tack on more profit, since spin offs wouldn’t be an extra bonus. Technically risk for doing something novel might also drive the price up (and make some people amazingly rich, if it really is as easy as it looked.)
For an initial estimate of $500 million, a sane contractor might put in a $2.5 billion bid. But we wouldn’t end up in the cost plus world where $500 million has turned into $8.8 billion and then hearing they really need another $800 million.
And importantly, if you could end the cost-creep (or cost-surge in the case of JWST), you could safely fund many more missions. That, at least to some degree, solves Bill’s “one-of-a-kind” objection.
And if you could stop the schedule overruns, you might have more researchers willing to let novel tech/instruments/science slip to the next mission, or the next. Increasing the rate any given type of mission (telescope/probe/lander/etc) reduces the sense that you’ll only get one mission in your working lifetime.
And you’d better believe that on a more fixed cost contract, the contractor is going to fight “requirements creep” a lot harder than they would on a cost plus contract. Requirements creep is a huge problem for any sort of contract whose goal is to build something.
For example, when you have a contract to build a house, the contractor isn’t going to let you swap out laminate flooring in the original contract for hardwood just because you as the customer didn’t think through all the downsides to laminate flooring before you signed on the dotted line.
Seriously, fixed price contracting is suited only for bidding on a statement of work that is based on knowing what a similar or nearly-similar article cost to build. JWST has many new technologies and some heritage hardware. But, in cost estimating for space programs, what really counts are (1) the scale of the end product compared to what has been built before; (2) the complexity off the systems integration and test; and (3) the skill of the engineering and technician workforce the contractors can bring to the job.
JWST is many steps up in the complexity scale compared to other space telescopes: it is not only big, it has to work nearly to perfection. Lots of starting unknowns. And, as can be read about over the past several years has very little tolerance for errors in design or manufacturing or testing. The systems engineering and integration challenges have been huge, largely due to the thermal control challenges.
The third means effectively how the profit/loss/reputational potential of the JWST development stacks up against the other development and production jobs N-G has in work. The programs are competing against each other for the best talent. Management’s decisions on where to place the best talent involve a lot of P/L/R considerations, and then added to those are the security credentials of the workforce. And, N-G has a lot of national security programs….
BTW, the US Navy had many overrun and delays when it used fixed price incentive contracting for new design ships. In FPI, every change order comes at a price in schedule and cost, AKA “get-well-opportunities.” And those change orders can stem from the mundane (departmental financial issues (e.g, continuing resolutions causing spoon-feeding contract funding), to the complex because the technology didn’t work as expected, to the inadequately specified in the SOW.
This model is used by millions of companies every day in America, including mine.
That’s the thing about Space Act Agreement contracts. The whole system is structured to work best for aggressively innovative companies with a larger goal beyond NASA…companies like SpaceX.
It would be a ton of fun to watch Jeff Bezos playing that game some day 🙂
Or they know that it will require something new that is coming available inside their timeline…but that they’ll need to develop and refine to their needs.
And they know it could cause schedule creep, so “shhhhh”.
Maybe he meant the sunshield itself was new technology, not the method to deploy it.
Keith, do you not like the JWST? You’re always down on it.
I’m pretty sure Keith is down on (in this case) NG who seem to have profited from the continual budget increases, and who continually promise “we’ve got it right this time, trust us; another billion will see us through, trust us.”
Keith, I think, is consistently down on NASA’s current way of doing business. JWST’s problems are current and high-profile, and really too hard to pass up as an example.
Working at NASA I can attest to how much of a budget suck it has been.
That’s probably the biggest impact- the huge opportunity cost. It also gives N-G a competitive edge to suck up the funds to starve out the competition. N-G is laughing all the way to the bank on the backs Of the taxpayers.
As a civil servant, I’m retired now, the difference in the commercial world business practice and the aerospace industry is staggering. For example, a commercial contractor was putting in a clean room floor in one of my labs. Their subcontractor used the wrong materials and didn’t follow procedures. The contractor did even hit us up for a re do- he apologized and immediately offered to re do it. In the aerospace business practice, with the tax payers footing the bill, we would have had to keep paying the contractor to re do it with profit over and over until it’s done right. It’s actually very profitable to screw up.
JWST is very exciting and we hope finally it does really launch and work. However, there were many other exciting stuff impacted by the cost overruns. In fact N-G is stifling these other program and benefitting by eliminating competition too. That’s why the aeropace industry milks the big flagship efforts for all they can get. The JWST hearings expose just the tip of the he iceberg of whe aeropace incompetence, complacency, and greed. Most of All I fault NASA and congress. They let these companies do this. We all deserve better than this with our hard earned tax payers money.
JWST was a disaster in 2011. It is an unmitigated disaster now. Senator Mikulski made a deal with the SLS mafia and this slid under the radar.
The senators and congressmen who pushed this crap through should resign – democrats AND republicans.
Oh, please. This instrument might be important to some, including me, but the dollar amount is pocket change to those critters.
From that early design:
“The … telescope can point to any target over more than half the sky.”
Not any more. Webb can only point to a circular swath of sky perpendicular to the sun. Like an Earth-based telescope that can only look at the sky just above the horizon; you have to wait for the sky to rotate into that band. Except that only occurs twice per year. How great for repeat and stacked observations when the telescope’s major working life is limited to 4yrs. All for a bargain price of $9 billion… and rising.
A fact often forgotten.
Or targets of opportunity. When something unexpected happens, like the Shoemaker-Levy 9 impacts on Jupiter or a nearby supernova, many observatories delay planned observations to look at it. JWST can’t do that, unless the even happened to be in the right part of the sky.
Well, this is basically true of every space telescope built in decades: Spitzer, WISE, etc. It’s not just the the power requirement, it’s the need to stay behind the sunshield. Typically, that permitted zone is several tens of degrees wide. You’re also forgetting that HST has similar limitations, and that being in LEO half the sky is unavailable to it at any given time. That’s why half the observing time is thrown away. The real horror is the slew time. Of course, HST hides this during the earth eclipse – the typical slew rates on these telescopes are about the same as the minute hand on a clock.
HST isn’t nearly that bad. Well, the Earth is in the way and the slew times are long. But it can point as close as 45 degrees of the Sun. They generally don’t, but as close as 90 deg. isn’t uncommon. And there is a “Continuous Viewing Zone”, roughly perpendicular to the plane of HST’s orbit, were you don’t have the Earth in the way for half an orbit. You’re thinking of infrared telescopes which are really pushing low temperatures.
The problem is that, even if one constraint like this isn’t too restrictive, it can add up. Once you start adding more constraints (solar arrays to Sun, unobstructed field of view for star trackers, etc.) you can paint yourself into a corner. Also, there are fewer clever tricks the observers can come up with. (The discovery of some of Pluto’s moons involved rotating HST to put the diffraction spikes in different places; JWST can’t do that due to the thermal constraints.)