How might NASA change under Trump? Here’s what is being discussed

"Elon get those rocket ships going because we want to reach Mars before the end of my term."

See full article...

 

[dzid]

Why do people always assume the worst about Trump? Why isn’t anyone writing about the countless blunders Biden has made? This bias is getting old.

Trolling skills need work...

 

[monogoto]

I agree that Mars is a better target than the moon. But not for manned missions. Send in the robots.

Surely at this point it has to be autonomous Optimus robots. Even if they're in secret demo mode being controlled by employees.

 

[rockscirick]

Supercomputing clusters are all about their interconnects. The machines at datacenters aren't optimized in this aspect. NASA doesn't actually do this integration. They contract with IBM or nVidia to build the clusters. NASA's involvement is on the software side of practically using the horsepower (and on making the architecture decisions for the next round of machines). That's where the value in Ames' location lies. They're in the middle of the highest concentration of software expertise on the planet.

I would agree, the machines at most data centers are not optimized around interconnects. However, AI/ML (artificial intelligence/machine learning) and complex physics simulation are both consumers of very large matrices, which benefit from superior interconnects. At some point, if you are serious about AI/ML, you have pursue this. At least, I expect Google to do this; they seem to be architecturally more aggressive than Amazon, and certainly more aggressive in AI/ML algorithms. (... which makes me wonder, what does Meta/Facebook run its AI/ML workloads on?)

Alas, I think Ames has been a late comer to clusters of GPUs. Unlike Dept of Energy labs, Ames seems to have been highly focused on massive clusters of CPUs until recently. Perhaps the software they have is not yet well-suited to GPUs. I can't imagine Ames computing being consolidated to Marshall, but could it be consolidated to DoE labs? (I'm not advocating this, but if I was forced to think out of the box, this would be hard to ignore.)

 

[dzid]

5. Increased Federal funding and commitment to robotic and Human outpost on TRAPPIST-1 system by July, 2038.

Trappist-1? 40 light-years is pretty ambitious, to say the least. I'm certainly interested in more detailed observations of the system in the shorter term, though.

 

[LibraryCommoner]

Cheap, Fast, Good. They can only pick two, but how much do you want to bet the incoming administration will want all three (and punish others for not achieving it).

you assume that "Good" in the Trump administration could be recognized as "good" to anyone outside his cadre. A central theme of the incoming admin is enriching Trump's oligarch pals, everything else is just gravy.

 

[TVPaulD]

Why do people always assume the worst about Trump? Why isn’t anyone writing about the countless blunders Biden has made? This bias is getting old.

What you call “assum[ing] the worst about Trump” is actually ‘observing his track record and taking seriously all the awful things he has literally said he is going to do’ and the notion that no one is “writing about the countless blunders Biden has made” is completely counterfactual. Trump is an incompetent fascist with a failed insurrection attempting to overturn an election he lost fair & square under his belt and he got more favourable treatment in the media than Biden did throughout the year.

 

[Helmore]

*snip

Biden shouldn't have withdrawn from Afghanistan. People never forgave him for that. His approval rating dropped like a rock and never recovered.

The lesson is pretty simple: Never be the President who ends a war, no matter how much people say they want you to.

snip

The irony here is that it was Trump who made the decision to move out of Afghanistan, Biden even delayed the withdrawal by a year or so. Now, Biden could have probably overturned that decision, but I guess he valued being seen as up holding past agreements more than not following through. It might even be that he agreed with the decision. But it wasn’t his signature on that particular decision.

 

[Wickwick]

Why do people always assume the worst about Trump? Why isn’t anyone writing about the countless blunders Biden has made? This bias is getting old.

Perhaps because the article was a forward-looking one at what's going to happen to NASA under Trump.

 

[Wickwick]

You forgot that the early SX engineers worked 15 hours or more a day. SX has shifts covering all 24 hours. Artemis (Orion/SLS) program Contractors doesn't work like that. If they did, Orion/SLS would be on their 6th crewed trip to lunar orbit. There would be 24 international crews with lunar deep space mission experience and Lunar Gateway orbital assembly would be completed. My opinion.

What the hell am I paying for if they're not working 24/7? I'm paying (my share of) > $2B per rocket and the contractors can't be bothered to spin up a third shift?

 

[Wickwick]

Sorry but I am not seeing exactly any results out of these research centers. Still using the same fuels that were used during Apollo and Shuttle. Same materials used for building the rockets.

It was a private company that developed methane fueled rockets and developed the landing and re-use of the 1st stage boosters in a way where it wasn't crazy expensive, like the Shuttles, were to re-fly.

NASA can't even build a mobile launch tower now for anything less that a few billion dollars? Not exactly sure, why a mobile launch tower is even necessary? Sure seems like a giant waste of money. Do you really need to be a rocket scientist to figure that out.

I think you're unaware of the activities of NASA in the first "A" of the name or perhaps of the entire science directorate.

I would agree that NASA is not leading the charge on advanced propulsion technology development (that would be the Space Force and formerly at the Air Force). But they still accomplish a lot in areas related to global monitoring and aircraft flight.

 

[Wickwick]

I would agree, the machines at most data centers are not optimized around interconnects. However, AI/ML (artificial intelligence/machine learning) and complex physics simulation are both consumers of very large matrices, which benefit from superior interconnects. At some point, if you are serious about AI/ML, you have pursue this. At least, I expect Google to do this; they seem to be architecturally more aggressive than Amazon, and certainly more aggressive in AI/ML algorithms. (... which makes me wonder, what does Meta/Facebook run its AI/ML workloads on?)

Alas, I think Ames has been a late comer to clusters of GPUs. Unlike Dept of Energy labs, Ames seems to have been highly focused on massive clusters of CPUs until recently. Perhaps the software they have is not yet well-suited to GPUs. I can't imagine Ames computing being consolidated to Marshall, but could it be consolidated to DoE labs? (I'm not advocating this, but if I was forced to think out of the box, this would be hard to ignore.)

This is not my area of expertise, but my understanding of how the large matrices for ML are manipulated is fundamentally different for the gradient descent work than for a full-blown CFD analysis. For ML there is a large matrix, but it can be copied many times and operated on separately to determine the partial derivatives in as many dimensions as desired then reassembled into a new matrix once the next step direction and magnitude has been determined. For CFD you're working to invert the entire matrix all at once. Yes, both require a fair amount of sharing of data, but the CFD needs it even more.

 

[EllPeaTea]

The irony here is that it was Trump who made the decision to move out of Afghanistan, Biden even delayed the withdrawal by a year or so. Now, Biden could have probably overturned that decision, but I guess he valued being seen as up holding past agreements more than not following through. It might even be that he agreed with the decision. But it wasn’t his signature on that particular decision.

I don’t think that were or are very many Americans opposed to the concept of a US withdrawal from Afghanistan. The problem was the way the withdrawal was executed. US soldiers being killed by suicide bombers, and pictures of people crowded into planes made it look like it was worse than the US evacuation of the embassy in Saigon in the 70s. And to make it worse, I don’t think any of the military top brass suffered any consequences.

 

[director_mr]

What is exciting about this is there is a clear goal, a willingness to not just accept the status quo with these statements and it just changes the expectations around what NASA should be doing when you say something like this. Whether people react against or for these proposals, I think the fact they are being proposed is going to be a positive thing for NASA. Need to go to Mars in 4 years? We'll need funding for that and dedication to a goal. We now have something to measure what is being done against.

The SLS system has been a terrible disaster that (almost) everyone here at arstechnica has railed against because it had almost no chance of success. While I don't think we can land people on Mars in 4 years, I think we can land something using the human landing systems they intend to eventually use on Mars. And people can claim their win, because people can claim we just needed to land there in 4 years and we did it. And that will still be an amazing accomplishment.

I'm hopeful that there will be positive steps in space exploration from this. I don't know how an honest person can feel like Elon Musk is not pushing space technology and science and exploration in a positive direction. Trump leaning on him to help guide Space initiatives seems like a common sense move if you are only going to look at who has done space launch things successfully lately. The commoditization of space launching, increasing space communication capabilities, and lowering the costs of launch to the degree he already has is an incredible positive.

 

[Wickwick]

Anything NASA not related to defense and country safety is a waste of tax money.

Given that NASA is a non-military agency, that's going to be difficult to square. Arguably, the thing most responsible for country safety is their earth monitoring programs. Water and food security is a requirement for a safe country.

Have no fear the Air and Space Forces are looking to develop space-based technologies for military purposes. Those avenues are being pursued - just not by NASA. There's rarely a need to launch an interplanetary probe with 6 minutes warning.

 

[Cheshire Cat]

Water and food security is a requirement for a safe country.

It sure is. Also monitoring of critical resources and climate change.

 

[fl4Ksh]

There is no way that we’re going to Mars in 4 years. We’re barely going to make it (back) to the Moon in that timeframe.

SpaceX has to perfect refilling Starship main propellant tanks in LEO before any Mars mission will be possible with Starship. That's scheduled to happen in late 2025 barring any lengthy stand down due to a major failure in one of the Starship IFT flights in that year.

The launch next Earth-to-Mars launch window occurs in Nov-Dec 2026.

An uncrewed Block 2 Starship flight to Mars would require two or three uncrewed Block 2 Starship tanker launches to LEO for the refilling process depending on how efficient that refilling process turns out to be.

The Mars Starship would then have enough methalox propellant for the trans Mars injection (TMI) burn to put it on a 270-day transfer to Mars. That trajectory is the lowest energy Hohmann trajectory and results in an arrival speed at Mars of ~5.5 km/sec.

The Mars entry, descent and landing (EDL) would be a modified version of the EDL that the Ship (the second stage of Starship) is doing now in the IFT flights. But the trajectory would be significantly different because the low density of the Martian atmosphere causes that Starship to spend a long time at ~10km altitude where atmospheric braking is sufficiently effective in reducing the speed.

Elon plans to send humans to Mars in the next window (Dec 2028-Jan2029) if the uncrewed Starship landing attempts at Mars in 2026 are successful.

That crewed Mars Starship will need ~50 kWe of solar electric power. The main tanks will be empty, but the header tanks will be filled with the methalox propellant for the landing burn on Mars. The standard header tanks are not insulated very well and would not be adequate for a 270-day trip to Mars. Consequently, the header tanks on the Mars Starship will need to be double-wall super-insulated zero boiloff tanks (ZBOTs) that probably will be located in the payload bay because of their large size.

 

[Nooge]

A lot of that infrastructure is extremely old and in desperate need of replacement. Maintenance costs are already unaffordably high, and continue to spiral out of control.

This has been reported on for years (including by Ars and even nasa's OIG), but here's a recent article on it:

https://www.thespacereview.com/article/4885/1
If there's an existing need to rebuild a bunch of stuff, that makes right now the cheapest time to consolidate.

Besides, I expect the rise of hardware-rich development, and much cheaper launch, will remove the usefulness of some of these unique testing facilities. We're probably already close to the point where it's cheaper to just test stuff by launching it.

Once we get down to single digit $100's per kilogram (which could happen in only a year or two), the extremely expensive testing facilities aren't going to be worth the cost anyway.

(Not saying we won't have testing facilities at all, just that ground testing facilities that cost billions of dollars and can't be easily rebuilt don't make sense when you can launch a whole satellite for a tiny fraction of that cost)

That’s probably true for a minor of test stands. But there’s several reasons we need terrestrial test stands. For example, even more important than testing is modeling long before you make the hardware (even SpaceX does copious modeling). Your models are mostly going to be garbage for anything cutting edge until you can get some test data to calibrate the model. There’s tons of component testing that we don’t see that’s critical for that.

Another reason is you can launch name times successfully and still find new issues that only occurred due to some rare combination of factors that’s easy enough to test for but may not show up until the 1000th try. If you’re building a mega-constellation, that can be a very expensive lesson learned. In fact, you might not even be able to figure out the exact root cause because you have no hardware returned to examine.

Finally, there’s some payloads you just can’t iteratively test, like JWST or really any expensive or long lead time payload. Launch cost is only one small factor and is nowadays usually not the biggest part of the lifecycle cost of a satellite (except cubesats).

 

[fl4Ksh]

Given that NASA is a non-military agency, that's going to be difficult to square. Arguably, the thing most responsible for country safety is their earth monitoring programs. Water and food security is a requirement for a safe country.

Have no fear the Air and Space Forces are looking to develop space-based technologies for military purposes. Those avenues are being pursued - just not by NASA. There's rarely a need to launch an interplanetary probe with 6 minutes warning.

IIRC, President Kennedy gave NASA "Brickbat" priority for its Apollo program. That put NASA first in line for materials, manufacturing assets, etc. needed for that Moon program ahead of the military with some caveats (e.g. ICBMs, nuclear warheads).

I doubt if Trump will need to go that far for Starship missions since Elon currently is building Starships on a mass production line at Starfactory in Boca Chica.

 

[Redglyph]

Good idea, put Trump and Musk in a rocket to Mars. "Go ahead, we're right behind you!"

 

[Chopfuna]

Yup, I totally believe that.

I highly doubt most of the other NASA centers are significantly different, though. NASA as a whole has been bogged down by incredibly inefficient pork for decades, and has a lot of trouble hiring good engineers at all. It's gotten to the point where even related orgs like JPL aren't executing well anymore.

(APL seems to have avoided this fate, but they're like 95% defense R&D anyway, civilian space is a tiny part of what they do)

Meanwhile, SpaceX seems perfectly fine with relocating all their testing facilities to an area of Texas that's far more rural than Huntsville. They haven't seemed to have any problem getting stuff done.

Maybe if you actually do exciting stuff instead of endless fruitless paperwork, you can convince engineers to work for you? I think that's a far bigger deal than where facilities are physically located.

Another reason for relocating to Texas is SpaceX needed a place to launch to the east that they could control. There is a risk that old space can slow them down in Florida. It makes sense to have manufacturing and testing close by.

 

[Chopfuna]

There are several options, both current and in development, that would be cheaper, faster AND better than SLS/Artemis as it is currently designed. Also, the Falcon 9 is proof positive that "cheaper, faster, better" is completely possible, though difficult.

Cheaper, faster, better can only happen if you make a revolutionary step like moving the meat industry’s disassembly line into the auto industry’s assembly line or you find a way to land your rockets and reuse them. I can’t think of any other examples.

 

[KyleK29]

Is that a real quote? Because I can totally see Trump trying to put his stamp on history by using Elon to get the US to be the first to Mars. Safety be damned.

And I don't think SpaceX would have a shot to do it in his term, it's just not feasible. There is so many engineering hurdles that they would have to overcome, the only conceivable way is if they started doing manned flights using just simulated models for a lot of those hurdles and made them one-way trips. Mars is not the moon. Mars is a planet that can be very hostile.

 

[Chopfuna]

And yet, tens of thousands of engineers currently work in Huntsville.They have the highest per-capita number of engineers of any metro area in the US (roughly 60 per 1000 employees, though I didn't dig very deep on it).

Regardless of your opinion, tons of engineers ARE willing to work there, and in other conservative states like Texas and Florida.

If you work in aerospace your company needs to be as close to the equator as possible and have an ability to launch to the east Texas and Florida are the only choices.

 

[flutefreak7]

Huntsville has tons of aerospace engineering talent. It's one of the main hubs for aerospace in the nation. Easily a quarter of the aerospace grads in my lab back in grad school got jobs in Huntsville after graduation, and my school wasn't even in the same state.

Plus, I thought Ars was generally in favor of spending government money to stimulate economically disadvantaged areas? Why are you arguing to spend it in some of the wealthiest areas of the country instead?

Not sure what y'all's problem with Huntsville is, other than general dislike of seeing your political opponents have nice things happen to them. It makes far more sense to keep a center in Huntsville than someplace like Goddard, for example.

I agree regarding Huntsville. Also worth noting that the DoD is like 30X the size of NASA and the size of a region's DoD footprint is a much better indicator of a region's concentration of talent than the unique facilities at a given NASA center. I also 100% agree with those saying that the unique facilities at any given center would take years to move, which is true for basically every center. The push for commcercialization of specialized work would be more in favor of shutting down and outsourcing unique NASA capabilities than moving them, which is a sad prospect, but would mean that NASA would be more fluid/mobile and less tied to specific locations where they've invested decades in test facilities and labs. NASA is already that way with Glenn running Plumb Brook Station, Marshall running Michoud, and White Sands being run by Johnson for example. Following this existing operating model of field centers controlling the smaller field labs, it makes sense that consolidating field centers would mean the unique facilities at existing field centers would just get downgraded to being field labs which would be run by a field center elsewhere while the thousands of desk jobs would consolidate elsewhere and/or become telework/remote jobs.

 

[jaabdad]

Cheap, Fast, Good. They can only pick two, but how much do you want to bet the incoming administration will want all three (and punish others for not achieving it).

SpaceX has already done cheaper, faster, and gooder by going back to first principles, accepting risk, and owning production. Now that SpaceX has done it, others will follow. Dilbert worked for Old Space.

 

[blackhawk887]

Finally, there’s some payloads you just can’t iteratively test, like JWST or really any expensive or long lead time payload. Launch cost is only one small factor and is nowadays usually not the biggest part of the lifecycle cost of a satellite (except cubesats).

If it were true that launch cost is only a small factor, then JWST could have simply added 1,000 kg of mass and made the sunshield much cheaper and more robust. Launch is cheap, right? What's another 1,000 kg? $30 million?

The marginal cost of launch past the limits of LV performance (and flagships are usually at the limits of LV performance) is enormous. On the order of $1B/kg enormous.

Also, when a test flight is $200M minimum (using Ariane 5), there aren't many tests that you can justify, even if flight testing would be helpful.

So while it's true that launch usually ends up being a small fraction of the mission cost, it's not at all true that launch cost plays a small factor. Launch costs (and launch performance, which is just another way to describe cost) drive almost every decision related to a mission.

 

[fl4Ksh]

A lot of that infrastructure is extremely old and in desperate need of replacement. Maintenance costs are already unaffordably high, and continue to spiral out of control.

This has been reported on for years (including by Ars and even nasa's OIG), but here's a recent article on it:

https://www.thespacereview.com/article/4885/1
If there's an existing need to rebuild a bunch of stuff, that makes right now the cheapest time to consolidate.

Besides, I expect the rise of hardware-rich development, and much cheaper launch, will remove the usefulness of some of these unique testing facilities. We're probably already close to the point where it's cheaper to just test stuff by launching it.

Once we get down to single digit $100's per kilogram (which could happen in only a year or two), the extremely expensive testing facilities aren't going to be worth the cost anyway.

(Not saying we won't have testing facilities at all, just that ground testing facilities that cost billions of dollars and can't be easily rebuilt don't make sense when you can launch a whole satellite for a tiny fraction of that cost)

You're right about ground test facilities. NASA largest facilities are wind tunnels, large vacuum chambers, gigantic stands for testing complete rocket stages, and huge movable launch towers, all dating back to the days of Old Space in the 1960s.

SpaceX has made those types of ground test facilities obsolete in its Starship development process. Now, the Starship stages (Booster and Ship) are tested on relatively small setups at the Massey's Site. The Orbital Launch OLM A is all SpaceX requires to ground test the Booster's 33 Raptor 2 engines at full thrust for ~10 seconds.

Those engines are tested individually at the McGregor engine test site at full thrust/full duration on relatively small test stands during development, qualification and acceptance testing.

SpaceX has done away with the gigantic Vehicle Assembly Building (VAB) and the custom-built huge crawler transporters like the ones at KSC in Florida. Instead, commercial self-propelled modular transporters (SPMTs) are used to move the largest rocket stages ever built. And the mechazilla arms on the Orbital Launch Integration Tower are used to stack the two Starship stages.

 

[NetMage]

If you agree that nasa has too many centers (and they absolutely do), you're gonna have to move something. Every center is going to have some argument for why they deserve to exist.

But you could try doing it gradually e.g. all new probes are controlled from Huntsville and as old probes shut down, Goddard shrinks.

 

[jimlux]

To address your edits. First, Huntsville has a population of 230k people. Your idea of "lots" of biomedical is not going to match that of Boston or even South Florida. I won't dispute that there is biomedical research in that area. But it's not one of the top 5 areas in the country for that sort of work. Alabama doesn't even show up on the most recent census map for bioengineers and biomedical engineers - even down to the map by area.

Secondly, I don't doubt that you can recruit people to move to Alabama. But the people taking that offer are self-selected to be willing to work in Alabama. If one is to move an entire NASA Center there, the question is, "what fraction of employees from Ames would be willing to move to Huntsville?"

One might look at Disney’s abortive move of Imagineering to Florida from California. Significant pushback from workforce.

 

[miker289]

The team that came up with these recommendations must be smoking something funny! Goddard is a high functioning center with good proven space exploration professionals. They managed the Delta ELV programs for years then did really good mission management for the New Horizons (Pluto flyby) mission and the Webb Space telescope to mention but a few of their successes. Ames is a hybrid center which works on in space and aerospace. They have a good staff of California education aerospace professional.
Marshall is the center whose employees brought the Space Shuttle and some post Apollo missions into existence. Since about 2000, they have overseen Shuttle replacements vehicles including the Artemis launch vehicle. The restrictions Congress placed on these efforts has led to the disastrous state of affairs today. If a center deserves punishment, it should be MSFC in my mind.
I read comments about Voyager and other missions that were run primarily by JPL and no other center. JPL is the crown jewel of space exploration and deserves better treatment because they general succeed with a brilliant staff!
The team looking to improve NASA might be advised to close down portions of Marshall. Assuming Artemis is cancelled, there is nothing for them to do. It would be crazy expensive to transfer employees to a new center and expect them to perform as if nothing has happened. The disruptions would catastrophic for the employees and their families.
So much for the new White House leadership saving the country money. I see little value in anything they have proposed. There are things that Congress can do to fix NASA's problems, but doing the stuff the advisory team has proposed won't help. (Writer is a former NASA aerospace professional with about 40 years of experience)

"JPL is the crown jewel of space exploration and deserves better treatment because they general succeed with a brilliant staff!" Yeah, this. Those people in Pasadena do some pretty amazing things, for sure

 

[MostlyAnger]

To an extent I agree. However, CA has at least a dozen Republican House members. Trump's not going to get much of anything done if he pisses off that contingent.

In fact it has nine. But anyway Ames is in a "blue" district* so I'm not so sure it would very much bother Republican representatives or their districts' voters.

* interestingly, its Democrat majority in November was the smallest among the congressional districts bordering San Francisco Bay. But not really small. Like ~ 58% D

 

[Wickwick]

You're right about ground test facilities. NASA largest facilities are wind tunnels, large vacuum chambers, gigantic stands for testing complete rocket stages, and huge movable launch towers, all dating back to the days of Old Space in the 1960s.

SpaceX has made those types of ground test facilities obsolete in its Starship development process. Now, the Starship stages (Booster and Ship) are tested on relatively small setups at the Massey's Site. The Orbital Launch OLM A is all SpaceX requires to ground test the Booster's 33 Raptor 2 engines at full thrust for ~10 seconds.

Those engines are tested individually at the McGregor engine test site at full thrust/full duration on relatively small test stands during development, qualification and acceptance testing.

SpaceX has done away with the gigantic Vehicle Assembly Building (VAB) and the custom-built huge crawler transporters like the ones at KSC in Florida. Instead, commercial self-propelled modular transporters (SPMTs) are used to move the largest rocket stages ever built. And the mechazilla arms on the Orbital Launch Integration Tower are used to stack the two Starship stages.

Please explain how SpaceX's testing has done away with the need for a wind tunnel that can test full-sized helicopters, e.g. And no, one can't do the testing with computers or else NASA wouldn't bother with the testing.

Or a large-cross-section icing tunnel for aircraft icing studies. One obviously cannot simulate physics if that aspect of physics hasn't been explored before.

Many of the large, unique facilities are for the first "A" in NASA - aerospace. SpaceX has done nothing to change the needs or the value of these facilities.

 

[JoHBE]

Good idea, put Trump and Musk in a rocket to Mars. "Go ahead, we're right behind you!"

And no need for aerobraking or parachutes or skycranes .A nice artificial crater will do.

 

[Statistical]

3. We have no incites into the true cost of Falcon 9 because SpaceX is privately held. For all we know Musk is launching Falcon 9's at a loss merely to capture the whole market and deter competition.

As a SpaceX shareholder all I can say is lol.

 

[Tod R. Lauer]

Yes. Let's send more robots with drills that can't penetrate the ground.

The reality is that while dollar-for-dollar one might get more science done with robots, robotic exploration would never receive the amount of funding that a manned mission will. Therefore, manned missions will always accomplish more because they're given more resources.

"Always accomplish more?" Robotic missions have been to every planet in the solar system, orbiting and even landing on a bunch of them, not to mention gobs of asteroids and comets. One's flying within 3.5M miles of the Sun just now. Then we have robotic space observatories mapping the Universe in every form of radiation from gamma rays to microwaves. Pictures from the Hubble and Webb are iconic.

Humans got to the Moon in 1968, but have gone past low Earth orbit since 1972. Space exploration has been a robotic game for 5 decades.

 

[MeValeMadre]

NASA, as it functions today, is dying. The SLS/Orion/Gateway projects are consuming all the available funding and bleeding all the other science and exploration projects to death. NASA used to be about pushing exploration with the side affect of bringing jobs to various contractors. Now the jobs aspect has metastasized and is killing the host. Something has to change. Jobs will be lost, centers closed, it is inevitable. Best to do that with some measure of control, and pick out some centers, and some projects to save. THEN, get out of the jobs business, and back in the exploring, risk taking business where NASA belongs. If Isaacman / Trump can even just get NASA and Congress back into the business of taking acceptable risks, that'd be a worthwhile accomplishment.

 

[OrvGull]

I don’t think that were or are very many Americans opposed to the concept of a US withdrawal from Afghanistan. The problem was the way the withdrawal was executed. US soldiers being killed by suicide bombers, and pictures of people crowded into planes made it look like it was worse than the US evacuation of the embassy in Saigon in the 70s. And to make it worse, I don’t think any of the military top brass suffered any consequences.

Fewer soldiers died in the withdrawal than had died in the previous year of occupation. Is this just "you touched it last" logic? Going in to begin with was a bad idea and it was always going to end messily.